Hello, and welcome to today's Journal Club Spotlight on Pharmacy webcast, which is supported by the Society of Critical Care and Medicine's CPP section. My name is Michael Haas. I'm the Surgical ICU Clinical Pharmacy Specialist at UMass Memorial Medical Center University campus in Worcester, Massachusetts. I will be moderating today's webcast. A recording of this webcast will be available to registered attendees. Log into mysccm.org and navigate to the My Learning tab to access the recording. A few housekeeping items before we get started. There will be a Q&A session after each of today's speakers. To submit questions throughout the presentation, click into the question box located on your control panel. You will also have the opportunity to participate in several interactive polls. When you see a poll, simply click the bubble next to your choice. SCCM provides the following disclaimer. This presentation is for educational purposes only. The material presented is intended to represent an approach, view, statement, or opinion of the presenter, which may be helpful to others. The views and opinions expressed herein are those of the presenters and do not necessarily reflect the opinions or views of SCCM. SCCM does not recommend or endorse any specific test, position, product, procedure, opinion, or other information that may be mentioned. And now I'd like to introduce your speakers for today. Each will give a 15-minute presentation followed by a Q&A. Our first presenter today is Miranda Wong, PGY2 critical care resident at Stanford Health Care in Stanford, California. Our second presenter is Andrew Webb, PGY2 critical care resident at Oregon Health and Science University in Portland, Oregon. And our third presenter is Emily Harmon, PGY2 critical care resident at Memorial Hermann Texas Medical Center in Houston, Texas. And now I'll turn things over to our first presenter. Hi, thank you, everyone. My name is Miranda. I'm a PGY2 critical care resident at Stanford. And today, I am going to be presenting on the effects of dexamethasone on days alive and ventilator-free in patients with moderate or severe ARDS and COVID-19. This is also known as the CODEX trial. By the end of this presentation, I want you all to be able to recall the rationale and previous evidence regarding the use of corticosteroids in patients with coronavirus disease 2019 or COVID-19. Second, analyze the outcomes of the CODEX trial. And finally, evaluate the place and therapy of dexamethasone for patients with moderate to severe ARDS and COVID-19. Starting off with some background, as you all may know, the first case of COVID-19 was discovered in December 2019 and has quickly become a global pandemic. The COVID-19 disease is caused by the SARS-CoV-2 virus, which is shown in the center. This is a spiked envelope positive sense RNA virus. Infection with the SARS-CoV-2 virus can lead to severe complications, including hypoxemic respiratory failure, ARDS, cytokine storm, disseminated intravascular coagulation, and many more. As of earlier this week, the World Health Organization reported that there have been greater than 213,000 deaths in the U.S. alone. And it's estimated that there's about a 26% mortality rate amongst hospitalized patients. However, this may be higher in our ICU and critically ill patients. Starting off with the disease course of 2019, or sorry, just starting off with the disease course of COVID-19, it's split into three phases. First, the early infection phase, two, the pulmonary phase, and three, hyperinflammation. This graph has time course on the X-axis and severity on the Y. Depending on the stage of infection, patients may display different clinical signs and symptoms. For example, in the early infection phase, patients may display nonspecific signs such as fever, dry cough, and diarrhea. However, as patients progress to the more serious stages, such as stage three hyperinflammation, they might display ARDS, go into shock, or cardiac failure. Additionally, the degree of inflammation increases dramatically as patients progress through these stages. So there are elevated inflammatory markers, including LDL and interleukin, as patients progress to stage three. The staging of COVID-19 also has implications of possible therapies and the timing that these therapies might be helpful. For example, in the early infection stages, antivirals might be more useful. However, it's hypothesized that in the later stages, corticosteroids, interleukin inhibitors, and JAK inhibitors might be more beneficial. Currently, there are many investigational therapies being studied. However, there's insufficient evidence for guidelines to provide a strong recommendation on any novel agents. Previously, the ACT-2 trial showed that REM deaths severe short in time to recovery from 15 to 10 days. However, this did not show to have a reduction in mortality. Dexamethasone is a long-acting corticosteroid with potent anti-inflammatory effects. And prior to the recovery trial, there were really limited quality evidence showing its use in COVID-19 patients. As you can see in the diagram on the bottom, corticosteroids decrease inflammation, specifically mediators such as TNF-alpha, interleukin-1, and interleukin-6, and decrease capillary membrane permeability. In terms of current guideline recommendations, the Surviving Sepsis Campaign on COVID-19 Management, which was published in March of this year, suggested low-dose corticosteroids, 200 milligrams of hydrocortisone a day. However, this was a weak recommendation. The World Health Organization recommends 6 milligrams of dexamethasone either orally or intravenously daily or hydrocortisone for 7 to 10 days. And this was a strong recommendation published after the recovery trial. And finally, the NIH recommends the use of dexamethasone, 6 milligrams daily, for up to 10 days. And they stratify these patients and the recommendation based on the level of mechanical support they are requiring. Going into some background on the use of dexamethasone for ARDS and COVID-19, first in February 2020 in the Lancet, dexards was a multi-center randomized control study of over 250 patients with moderate to severe ARDS. They showed dexamethasone with a dosing strategy of 20 milligrams daily for 5 days, followed by 10 milligrams daily on days 6 to 10, decreased the number of days that the patients were on ventilators. This trial had limitations, including that it was stopped early due to safety. Next, the GlucoCovid published in June of 2020 was a multi-center, partially randomized open-label study of only 85 patients with confirmed COVID-19. They found that methylpregnancy alone decreased the composite of death admission to ICU and requirement for non-invasive ventilation compared to placebo. This study was limited and also that it had a limited end was inadequately powered and patients who received steroids also got other investigational treatments. And finally, the Recovery trial published in July 2020 in New England Journal of Medicine was a multi-center, randomized controlled open-label study of over 11,000 patients looking at dexamethasone 6 milligrams daily for up to 10 days. And what they found was that 28-day mortality was much lower in the dexamethasone group compared to standard of care. Moreover, a subgroup analysis showed that there was a mortality benefit specifically in the patients requiring invasive mechanical ventilation and oxygen. This study also had limitations in that 15% of patients were excluded per the investigator and adverse events were not noted. So as you can see previously, there are many different dosing strategies for steroids and different dosing as well as number of days, both for our ARDS patients and our COVID-19 patients. Now I'm going into the CODEX trial. This was published September of this year in JAMA. It was a multi-center, randomized controlled open-label study. The purpose was to determine whether intravenous dexamethasone increased the number of ventilator-free days amongst patients specifically with COVID-19 associated ARDS. They looked at 299 patients and they were randomized in a one-to-one fashion to intravenous dexamethasone, 20 milligrams daily for five days, followed by 10 milligrams daily or until ICU discharge plus standard of care versus standard of care alone. The primary endpoint was days alive and free from mechanical ventilation during the first 28 days. Secondary endpoints included 28-day all-cause mortality, clinical status measured by the six-point ordinal scale, mechanical ventilation duration, and SOFA score. In terms of statistics, initially a planned sample size of 350 patients was calculated to adequately power the study. All patients were included in the primary analysis and secondary outcomes were adjusted for age and PaO2 to FiO2 ratio. Of note, no adjustments were made for multiplicity, so secondary outcomes were considered exploratory. In terms of inclusion, patients were included if they were from one of 41 participating centers in Brazil between April to June of this year. Patients were 18 years of older, confirmed or suspected COVID-19 by PCR or serology testing, and receiving mechanical ventilation within 48 hours of meeting criteria for moderate to severe ARDS according to the Berlin criteria. Patients were excluded if they were pregnant, actively lactating, had a history of dexamethasone allergies, were using steroids in the past 15 days for non-hospitalized patients, or using steroids within the prior day prior to enrollment, using steroids for other indications such as septic shock, immunocompromised, expected to die within 24 hours, or they refused to participate in the study or there was lack of consent. Here is a table of the baseline characteristics. As you can see, the average age of these patients was approximately 60 years old. Moreover, the time since the onset was about nine days in the dexamethasone group versus 10 days in control. 95% of these patients were confirmed COVID positive. And as you can see, most of these patients were enrolled within 24 hours of qualifying for moderate to severe ARDS with a mean P to F ratio of about 130. All baseline characteristics were generally well matched. For their primary outcome of ventilator-free days during the first 28 days, you can see that it was higher in the dexamethasone group versus standard of care, so 6.6 days versus four days. This diagram on the left, you can see for the green bars, there are more ventilator-free days as the number of days progress. And the image on the right just further highlights the absolute difference of 2.26 days for patients who were treated with dexamethasone plus standard of care versus standard of care alone. And again, statistically significant with a P value of 0.04. This also looked at a subgroup analysis for predictors. And there was no statistically significant interactions for the primary endpoint of ventilator-free days during the first 28 days. So they looked at age, P to F ratio, SAPS-3, time since the onset, position at randomization and vasopressors, and there was no statistically significant interactions. In terms of secondary outcomes, I did want to point out a statistically significant difference between groups. As you can see, the SOFA score at day seven was lower in the dexamethasone group versus standard of care, 6.1 versus 7.5 with a P value of 0.001. And the SOFA score estimates ICU mortality. So as you can see, a difference of 6.1 to 7.5 correlates with a 5 to 10% difference in expected mortality. The SOFA score at 48 hours and 72 hours were not statistically significant between groups. And then in terms of other secondary outcomes, there were no statistically significant differences. However, you can see that there was a trend towards decreased mortality at 28 days in the dexamethasone group versus standard of care. So an absolute difference of 5.2% lower in the dexamethasone group. And in terms of adverse events, there were no statistically significant adverse events. For serious adverse events, it was actually lower in the dexamethasone group, 5 versus standard of care, which had nine serious adverse events. There was a trend towards a higher incidence of hyperglycemia in the dexamethasone group at 31% versus 28.4. However, this was not statistically significant. So the authors' conclusions is that among patients with COVID-19 and moderate or severe ARDS, the use of intravenous dexamethasone plus standard of care compared to standard of care alone resulted in a statistically significant increased number of ventilator-free days over the first 28 days. I did want to point out some strengths and limitations. Some strengths is that it was multi-centered across 41 different ICUs, and groups were well-balanced at baseline. Also, relative to the recovery trial, adverse event data was recorded. Some limitations are that other clinical interventions were left to the discretion of the ICU team, leading to some possible heterogeneity. Also, it was open-labeled, so some investigator bias might have occurred. The trial stopped early, so notably it was underpowered, and 35% of patients in the standard of care group actually received at least one dose of corticosteroids. So just doing a side-by-side comparison with the recovery, I just wanted to highlight the differences. So in terms of patient population, the recovery trial had more patients over 11,000, and they looked at all patients who were suspected of COVID-19 confirmed. However, the CODEX trial specifically looked at patients with ARDS, so, in fact, a little bit sicker. However, their N was much smaller at 299. Also, time from symptom onset, the CODEX patients, on average, presented one day later. However, you can see the inter-quartile range overlaps. In terms of intervention, you can see the dosing is much higher in the CODEX study in that they used dexamethasone 20 milligrams for five days, followed by 10 milligrams daily for five days, whereas the recovery trial looked at six milligrams daily for 10 days. And then lastly, you can see that the recovery trial overall for 28-day mortality was much lower at 22.9% versus the CODEX study, the mortality was much higher at 56%. This might be because the patients were much sicker, and also because these patients didn't have any other investigational therapies that were utilized at the time. So some takeaways and unanswered questions that in patients with moderate to severe ARDS, the use of dexamethasone plus standard of care increases the number of ventilator-free days compared to standard of care alone. Some notable trends that were not statistically significant but notable is that there was lower all-cause mortality in the dexamethasone group and greater incidence of hyperglycemia. Of note, this trial was stopped early and underpowered, but had the trial gone on, I wonder if these would have been statistically significant. Moreover, I think an unanswered question is the most optimal dose and duration of corticosteroids for COVID-19, specifically in patients with moderate to severe ARDS. So some future directions are to compare adverse events reported in the CODEX trial to that reported in the recovery once the trial is completed, as the one that was published in July was just a preliminary report, and then possibly identify treatments with actual mortality benefit in patients with less severe disease so that they do not progress to ARDS. Now going on to my polling questions, the first question is, what corticosteroid dosing strategy does your institution use for COVID-19 positive patients with moderate to severe ARDS? A, dexamethasone 6 milligrams daily for 10 days, dexamethasone 20 milligrams daily for five days, followed by 10 milligrams daily for five days, hydrocortisone 60 milligrams every six hours, or steroids are not regularly used for these patients. It would be interesting to see if institutions have readily adapted what was presented in the CODEX trial versus recovery. Here at Stanford HealthCare, we still use 6 milligrams daily for 10 days for all patients who are requiring oxygen support or mechanical ventilation. We have not yet adapted what's in the CODEX trial. And then second polling question, let me see if I can advance this. Should guidelines be updated to reflect the higher corticosteroid dosing presented in the CODEX trial, the 10 to 20 milligrams a day versus the 6 milligrams a day for up to 10 days for COVID-19 positive patients with moderate to severe ARDS? A, yes, B, no, or C, more research is needed to compare the dosing strategies head to head. Okay, so it looks like maybe I'm unsure if guidelines should be updated at this time. It's interesting, it will be interesting to see if guidelines do become updated, especially those providing sepsis guidelines, which was published prior to both the recovery and the CODEX trial. Here are my references, and at this time, I would love to take any questions that you may all have. Thank you, Miranda. Excellent presentation. We do have one question currently, so a question from the audience is, what was standard of care in this study? So, standard of care was not with steroids. In fact, no patients actually received remdesivir because it was unavailable in Brazil at this time. Some patients did receive azithromycin, and about two-thirds received hydroxychloroquine. However, other than that, it was really supportive. Okay, we have one more question. Given the substantially higher rate of mortality as seen in the CODEX trial, does the generalizability apply to countries outside of Brazil? Yeah, that's a great question. The trial did have a much higher incidence of mortality, and they noted that because maybe it is a second or third world country, and their predicted mortality at the time was much higher. I think that we would probably see similar trends, and that we would see decreased mortality. However, I don't know if it'd be as substantially different between groups. I think what we might see more closely reflect the recovery trial, and that our mortality seems much lower so far in our COVID patients. There was also a meta-analysis published in JAMA earlier of this week, and they showed an overall mortality rate of about 35%, which also pretty closely mirrors the recovery trial. I think we have time for one more question. The next question is, what was the most common corticosteroid regimen that the standard of care group received to the family? The most common... Sorry, I was a little bit staticky. The most common corticosteroid dosing strategy in this control group or standard of care group? Correct. Okay. I'm actually not positive what the most common steroid dosing strategy was, since some of these patients actually only received one dose. It seems like many of these patients did receive corticosteroids in the standard of care group for septic shock. I would think it might be 50 milligrams Q6 hours, but that's something I would probably have to double check for you. All right. That concludes our Q&A session. Thank you, Miranda. Now I'd like to introduce our second presenter, Andrew Webb. Hi, everyone. My name is Andy Webb. I'm the current PGY-2 critical care resident here at OHSU. Today, I'm going to be presenting on the MIDUS trial, which was a prospective trial looking at mitadrine to help wean off vasopressors in ICU patients with persistent hypotension. I'm sure we've all had that patient in the ICU who's been on the unit for five, six days, doing well, but otherwise on that really low dose of norepinephrine, just keeping them on the unit. All of us are kind of scratching our heads, trying to get them off that vasopressor, trying to find reversible causes of hypotension. And one of the things that's kind of become more popular over the past few years has been putting them on mitadrine to try to get them off that agent. So to start my presentation, I have no financial disclosures, although I'm certainly open to them. And I'll start with some polling questions just to kind of get a feel for the audience's practice before we head into the study itself. All right. So we'll start before the publication of this trial. If you had these four patients on your unit, which one would you be most likely to initiate mitadrine in? So an 82-year-old woman with sepsis of a urinary source on day six of norepinephrine at a relatively low rate, maybe a 65-year-old man post-CABG, slowly weaning off epi, a patient who's just post-op day one from a cervical decompression on a relatively low dose of norepinephrine, or your 42-year-old COVID positive patient on ICU day 17 on and off norepinephrine. Yeah, so I'd probably agree with the audience here that the two patients I'd be most likely to initiate or at least recommend mitadrine on would be those patients who have just kind of been on either on and off pressers or the one who's just taken a real long time to get off. So I would certainly agree with that. I'd just like you, everyone, to keep that in mind as we go through the rest of the study. Now, my next polling question is, I want to get a feel what everyone's threshold for their maximum dose of mitadrine is. The 10q6, 20q8, 20q6, or maybe up to 40q6. I understand not every regimen is going to be listed on this slide, and some institutions only do q8, but if you were to choose one of the above, which would you choose? All right, yeah, so 20q8 is what most people have selected here, and I think you'd likely agree with what the study did, but some institutions do kind of push that higher, including some that may all go all the way up to 40q6. So just, I would like everybody to keep that in mind as we kind of go through the rest of the article. So we'll jump into it. So mitadrine is an oral medication that's actually a prodrug for its metabolite desglymitadrine, which mitadrine undergoes enzymatic hydrolysis within the circulation, and desglymitadrine is an alpha-1 receptor agonist, as many people will call it, kind of oral phenylephrine. Peak levels of the active metabolite are reached about one hour, and it'll increase blood pressure by both venous and arterial vasoconstriction. And the interest in mitadrine has really come into practice within the last 10-ish years, starting particularly with a study by Levine and colleagues, which found that oral mitadrine allowed surgical ICU patients to wean off their phenylephrine infusions at a faster rate than prior to starting mitadrine. This led to Poveromo and colleagues to publish a study showing that mitadrine may actually accelerate time to ICU discharge after pressure discontinuation, followed by a study by Whitson and colleagues that showed that mitadrine appeared to decrease IV vasopressor duration overall. Rizzi and colleagues showing that cumulative vasopressor doses decrease after initiation, but with all of these, they were all retrospective, were potentially confounded by indication, and had various endpoints that they reported. So a meta-analysis first electronically published last year, but in press this year, kind of cast a doubt on some of these findings, and really called for a prospective evaluation of mitadrine's role in getting patients off vasopressors. So the MIDUS trial is our first randomized placebo-controlled trial evaluating whether or not mitadrine has a place in our ICU. So to kind of jump into things, MIDUS's inclusion and exclusion criteria was kind of ideally what you would expect in terms of what patients would be eligible. So adult hypotensive ICU patients on a single-agent vasopressor who were just having difficulty meaning. Now they defined flat doses as their inclusion criteria, defining less than 100 micrograms per minute of phenylephrine, less than 8 mics per minute of norepinephrine, or less than 60 mics per minute of an Australian pressor by the name of metaraminol, which I'll get into a little bit later on in the presentation, as the minimum inclusion criteria to get in. Patients had to be unable to wean after 24 hours of meeting their blood pressure goal, and they were excluded if they had hypovolemic shock or adrenal insufficiency, as these are reversible causes of hypotension, or if they had liver failure, chronic renal failure, heart failure, and bradycardia. And some of those I'm going to get back to later on in the presentation, but I'd just like you to kind of reflect on which patients you may be most likely to recommend mitadrinone, and whether or not this patient population is reflective of that. In terms of the intervention, in patients who were ultimately deemed to be eligible for the intervention, they were randomized to either mitadrine at 20 milligrams every eight hours, or a matching capsule placebo. And the intervention was continued on either until ICU discharge, or the patients had worsening hypotension, had worsening hypotension, developed an epinephrine requirement, given the fact that mitadrine is a pure afterload increaser and has no effect on inotropy, has signs of hypoperfusion, which they define mostly as an increasing lactate, or if their blood pressure was maintained for 24 hours on mitadrine, at which case they were weaned with a standard weaning protocol. The primary endpoint of this study was time to IV presser discontinuation in hours, and they defined that kind of discontinuation as requiring a 24-hour vasopressor-free period to account for patients who were kind of coming on and off in that kind of few-hour period after discontinuation. They powered their study to estimate a six-hour reduction in vasopressor time, estimating 50 patients would be needed in each arm to reach that. Some of the secondary endpoints they were interested in was time to ICU discharge, ICU and hospital length of stay, as well as the rate of ICU readmission. And there are several post hoc analyses that they did as well, which I'll mention later on in the presentation. So I'll just briefly go over the population here. I think some kind of key things to focus on, if you just look at the top of that table, so there's about 50-50 in terms of the gender, but patients range from about 60 to 70 years on average and weighed about 80 kilograms. What I'd really like to focus on is kind of the characteristics of the critically ill population. In terms of the day one SOFA score, they were about four to five, so not exactly the most kind of organ dysfunction in many of these patients. I didn't list it here on this table, but the mean APOCHE2 score at baseline was around 14 to 15. And in terms of the ICU indication, about 60 to 70% of the patients were in the ICU for a surgical indication, most commonly post-operatively, with only about 20% having septic shock. Additionally, one point that I do want to highlight is the duration of pressors at the initiation of mitadrine was only 35 hours in both arms. And the equivalent dose of norepinephrine in both arms was about 0.06 micrograms per kilogram per minute. So in these patients who were randomized to either mitadrine or placebo, in terms of their primary endpoint for the time to discontinuation of vasopressors, they did not detect a meaningful difference, with 23.5 hours being the time to discontinuation for mitadrine versus 22.5 hours in placebo. That one hour was not deemed statistically significant. In a post hoc time to event analysis, you can see the time to event of discontinuation was well matched between both arms. In terms of secondary endpoints, similarly, there was really no difference in endpoints. So there was no difference in time to ICU discharge readiness, no difference in ICU length of stay, no difference in hospital length of stay, although there was a numerically smaller difference in the mitadrine arm, and there was also a numerically smaller rate of ICU readmission in the mitadrine arm. Now, I apologize. I think my slide may have blocked one of the boxes, but one of the post hoc analyses was the proportion of patients who had epidural analgesia, and they actually found, did find a subgroup of patients who had epidural analgesia that actually was able to wean off vasopressors at a faster rate at about 14 versus around 30 hours. However, it was a very small subgroup of this study, so it's unclear whether that is just a type one error and is really just exploratory in nature, but is worth noting that they did find a difference in that one subgroup. In terms of safety, this is one thing I do want to highlight. In terms of both arms, the overall rate of adverse drug reactions was not different, at least not statistically significantly different, with about 30% experiencing any adverse drug reaction in the mitadrine arm versus 25% in the placebo arm. One thing that's not particularly surprising, however, is the rate of bradycardia was significantly higher in the mitadrine arm. Considering mitadrine is primarily that afterload increaser with no effect on cardiac contractility, you would expect that autonomic reflex to potentially cause bradycardia, as you may frequently see with ephenolephrine infusion. So about 7.6% of patients who received mitadrine experienced bradycardia, which was largely defined as a 20% decrease in heart rate goal set by the team or a heart rate less than 40 beats per minute versus 0% in placebo. Rates of hypertension, which was defined as a greater than 20% increase over the systolic blood pressure goal, were numerically higher but did not reach statistical significance and neither did rates of atrial fibrillation. So the author's conclusion was that the use of mitadrine did not affect time to vasopressor discontinuation, nor did it hasten time to ICU discharge, but it did lead to more bradycardia in patients who are randomized to mitadrine. But there are several limitations that I do want to highlight with this study. One is that it's a largely surgical population, and while this certainly isn't necessarily a limitation, as surgical ICU patients do require vasopressors a large portion of the time, it would be difficult to extrapolate these findings to a largely medical population. One additional limitation is that a significant portion of the patients were randomized in the Australian hospital arm, and about 45% of the overall population received metaraminol, which certainly is a vasopressor that I doubt many of you are familiar with. But to give you a kind of brief overview, metaraminol is somewhat like an IV infusion of ephedrine, which if you work in a surgical ICU, you may be familiar with your anesthesiologist quite liking it. And the reason why this is significant is because metaraminol, while it is an alpha-1 agonist, also works by releasing norepinephrine from synaptic vesicles. And because of that and prolonged utilization, you may see tachyphylaxis with increasing doses required to achieve the same pharmacodynamic goal. And that's important because patients who had worsening hypotension on the same dose of metaraminol may simply be experiencing tachyphylaxis rather than some underlying pathophysiology that may confound the effect of midadrenaline in that patient population. Other limitations include the lack of a mention of vasopressin or other vasopressor doses, as it has been found that using vasopressin as your final pressor before discontinuation can actually lower the rate of hypotension compared to using norepinephrine as your last vasopressor. Additionally, at least in my experience, the patients that I ultimately recommend midadrenaline on are generally on vasopressors for much longer than 35 hours. So the fact that the bulk of these patients were only on vasopressors for about a day and a half makes me wonder whether they would have discontinued vasopressors not long after the initiation of midadrenaline anyway, and that midadrenaline really didn't do anything to kind of help it along that path. So it would be good to only include patients who were on vasopressors for a longer duration than that 35 hours. The other thing that's curious is that this study had been going on for several years prior to its publication, and they excluded a large number of patients for things like high-dose vasopressor requirements at enrollment or at evaluation. So it would have been good if they had explained potentially whether they continue to evaluate patients after that initial evaluation, as patients who may have been initially evaluated their first day in the ICU may have been great candidates for the study three, four, or five days into their ICU stay, and we're really missing a bulk of that population. And lastly, several of the exclusion criteria are of patients of key interest, particularly those with liver failure. Now, in the protocol that was published in 2017, they explain that they exclude patients with liver failure because midadrenaline is metabolized in the liver. However, we know that midadrenaline is safe in our liver failure patients, as it's part of kind of a standard regimen for hepatorenal syndrome. Additionally, there are pharmacokinetic studies which demonstrate that midadrenaline reaches safe but higher concentrations in the blood. So these patients may actually be particularly, see a particular benefit to midadrenaline. As in my clinical experience, oftentimes it's our patients in liver failure who are on low-dose vasopressors for the longest duration of time. Now, to kind of sum this up in terms of some of the outcomes that were reported, as I had mentioned previously, there was a meta-analysis that was published last year prior to the publication of MIDUS. So I included some of the endpoints that were reported in the MIDUS trial with the summarized findings from that meta-analysis. So there's only two endpoints that matched with what they reported in the meta-analysis, the first being the ICU length of stay. As you can see, there is no difference in MIDUS. And so this study really did not add anything to our understanding of midadrenaline's effect on ICU length of stay. Additionally, for total vasopressor discontinuation, midadrenaline also did not really move the needle as to what was previously understood from the meta-analysis published last year. I will say that evaluating the effect of midadrenaline on total vasopressor duration retrospectively is not really fair, as when you look retrospectively, the patients who are given midadrenaline are going to be the ones who are on vasopressors for a longer duration of time. So with this study, I don't really believe that it added a whole lot to our current understanding. However, there are several future directions that I'd like to see this research go. One is I'd love to see a larger medical population. Our patients in septic shock are often the ones that we have the most difficulty weaning off vasopressors. An unfortunate minority of patients in this study had septic shock. Additionally, it'd be interesting to see higher doses. This study maxed out at 20 milligrams Q8, but there are several reports of centers using up to 40 milligrams Q6, and there's a linear relationship between plasma levels and systolic blood pressure, as was reported in the early 90s when this was being studied for orthostatic hypotension. I'd also love to see patients be included after longer periods on IV vasopressors. As I've mentioned previously, the patients who are on pressors for four, five, six days, as you also said in that initial polling question, were the ones that we really want to see whether midadrine has an effect on. And lastly, it would be interesting to explore alternative oral antihypertensives. Now, in a perfect world, I'd love to see droxydopa evaluated in this indication. That being said, droxydopa is quite pricey, so I think you might have a hard time convincing a department to study that. Initially, pseudoephedrine has been reported in case series, but has never been formally evaluated and may have a potential role as the additional activity of being able to release catecholamines from synaptic vesicles as well. So to kind of sum up, that final question is whether midadrine does have a place on your patient's MAR. And I think it's worth considering that patient with a prolonged MICU course with difficulty weaning pressors, but of course having that continuous evaluation, where if it's not effective or not tolerated, it certainly should be discontinued. And with the body of evidence that we have from our retrospective studies, as well as from MIDUS, it's likely ineffective in our SICU patients or in patients who are kind of still in that early course of vasopressors, as they may kind of wean off on their own after a day or two. And of course, considering midadrine can worsen bradycardia, it's important to likely avoid this in patients who are either baseline bradycardic or at risk for bradyarrhythmia. So with that, thank you for listening, and I'm excited to hear what questions you guys have. Thank you, Andrew, for that excellent presentation. A question has come through. So the first question is, would you use this in SCI with elevated map goals, so maps of greater than 80 to 85? Yeah, that's a great question. I think that that is probably the one place where this may have a role, as phenylephrine is often the agent of choice in those kind of spinal cord surgeries, where you're trying to get the map up to, you know, 80, 90. And if you have patients who are having difficulty weaning off phenylephrine, I think midadrine makes sense. And they didn't necessarily break out what the types of surgeries were in MIDUS, so it's hard to say what subgroups may have seen or not seen benefit. But in that subgroup that they kind of go into the discussion about whether epidural analgesia and the potential benefit of midadrine in that subgroup might be related to the etiology of hypotension from spinal manipulation. So I think it'd be worth considering in somebody who's having difficulty to wean off phenylephrine. I certainly wouldn't discount it entirely, but I'd like to see more kind of prospective research to definitively say that that has benefit in that population. Okay, the next question is, does the presence of mixed etiology of shock, such as cardiogenic shock, influence your recommendation to initiate midadrine to wean off vasopressors? Yeah, I think it certainly would influence my decision, especially in the mixed state of having cardiogenic shock. Considering midadrine has no effect on cardiac contractility, I would certainly concern that trying to wean someone off vasopressors who has a cardiac abnormality would ultimately worsen the effect of midadrine on cardiac contractility. Would ultimately worsen the underlying etiology of that shock. And the worst thing that I would want is to make a recommendation to try to get someone off their pressor, start that medication, and the starting of that medication worsens the underlying disease. So considering this causes bradycardia will only worsen the cardiac function. I would avoid this in patients with mixed shock, especially if there's a cardiogenic component. Okay, I think we have time for one more question. The last question would be, was there any reporting from the authors of how many patients were continued on midadrine therapy upon hospital discharge? That is a great question. So part of the hospital or the study protocol was to discontinue midadrine at ICU discharge. I am familiar with the research that's likely being referenced here as there is a significant portion of patients who when started on midadrine in the ICU may transfer to the floor on it and maybe potentially transfer or discharge home on it. So I think it's very important as ICU pharmacists to be cognizant of that transfer medication reconciliation as there are medications like antipsychotics, midadrine, or other things that are really just supportive care in the ICU that likely should not be continued on the floor or continued home. As if patients are still requiring midadrine to support their blood pressures in the ambulatory setting, that should trigger additional evaluations as to the underlying cause of the hypotension, especially as midadrine does have a black box warning for supine hypertension that may be dangerous if patients were to continue a high dose midadrine at home. So to answer the question more directly, they did not report the proportion of patients who may have been discharged home on midadrine, but it was in their protocol to discontinue upon ICU discharge. Okay. I think that concludes our Q&A session. Thank you, Andrew. Now I'd like to introduce our third presenter, Emily. Hello, everyone. My name is Emily Harmon, and I am a current PGY-2 in critical care at Memorial Harmon Texas Medical Center, located in Houston, Texas. And today I'll be speaking on the effect of aspirin on deaths associated with sepsis in healthy older adults, or the study called antisepsis. Now before we begin, we see in the title that this trial looks at healthy older adults. So I'd like to begin with a polling question looking at the age you typically take care of in your unit. So our first polling question will look at the age of your unit going from less than 20 years for A, B, 20 to 39 years, C, 40 to 64 years, and D, for greater than 65 years. And I know we see a multitude of patients, but just choose the one that best represents your unit. And I see most of us take care of 40 to 64-year-olds and then greater than 65, which is wonderful. Maybe this trial will be a benefit for you. As we begin, I'd like to go into a little bit of the background of sepsis. Sepsis is a major cause of death and disability and something we encounter often as healthcare professionals in the critical care field. If there's a preventative treatment available, the incidence of mortality associated with sepsis may be reduced. This trial investigated the hypothesis that once daily low-dose aspirin would reduce death or hospital admissions associated with sepsis in elderly patients. Now, what inspired this trial? In 2017, a propensity-matched analysis of individual patient data from published retrospective observational studies was published. This analysis evaluated outcomes related to sepsis when patients were taking low-dose aspirin and included over 680,000 patients, the majority of which came from one large insurance database study. This study found an average 6% reduction in mortality when aspirin was used prior to sepsis onset. Some hypotheses of how aspirin is beneficial in sepsis include its effect on pro-inflammatory cytokines like tumor necrosis factor and interleukin-6. In sepsis, immune cell receptors recognize diverse PAMPs mediating intercellular signaling events that result in NF-kappa-B activation and transcription of TNF. NF-kappa-B activation is inhibited by aspirin, and while we don't know the concentration of NSAIDs required for inhibition, they have been measured to be lower than that of daily low-dose aspirin. Another proposed mechanism is through lipid mediators. In sepsis, they reduce established inflammation by mechanisms like the restoration of polymorphonuclear apoptosis, which limits continued production of pro-inflammatory cytokines in the tissues. Low-dose aspirin is shown to increase lipotoxins and nitric oxide, which in animal models was associated with lower mortality. And finally, we look at how aspirin affects platelets in sepsis. Platelets become activated due to interaction with invading bacteria, either by binding of bacteria to plasma proteins, that are ligands, for platelet receptors, through direct bacterial binding to platelet receptors and secretion of aspirin-binding bacterial product-like toxins. Aspirin reduces the activation of platelets through the inhibition of COX-1, as we know, therefore reducing activated platelet contributions to our microvascular occlusion and organ failure. Before we look into the antisepsis trial, we first need to know more about ASPRi, of which antisepsis is a sub-study. ASPRi, or the effect of aspirin on all-cause mortality in the healthy elderly, was a primary prevention study looking to prolong the healthy lifespan of older adults. This trial had a primary endpoint of disability-free survival, and primary composite endpoints of event of death, dementia, and persistent physical disability. ASPRi included about 19,000 patients, and found 1,052 deaths occurring in the 4.7-year follow-up period. They found risk of death from any cause was 12.7 events per 1,000 patient years in the aspirin group versus 11.1 events per 1,000 patient years in the control group, and found an overall higher cause of mortality in the aspirin group, which they then attributed primarily to cancer-related deaths. Keeping these trial methods and results in mind, antisepsis was conducted. This was a randomized, placebo-controlled, double-blind, community-based clinical trial. The primary endpoint was death associated with sepsis, and secondary endpoints included admission to the hospital for non-fatal sepsis, or admission to the ICU for fatal or non-fatal sepsis. An endpoint adjudication committee of five physicians established whether a potential septic event met endpoint criteria. Two committees would review the de-identified cases and determine if sepsis contributed to death. If they did not agree, the case was then brought to a third committee member. Antisepsis inclusion criteria included absence of life-limiting chronic illness, free from diagnosed cardiovascular disease, dementia, or disability, and no major risk of bleeding or aspirin hypersensitivity. Patients were excluded if they had any of these previous conditions or disabilities. Patients were randomized one-to-one to aspirin 100 milligrams daily, first placebo, remotely through a web portal according to a computer-generated randomization schedule. Randomization was stratified by general practice and age, grouping them 70 to 79 years, and then 80 years and above. All participants, investigators, and staff were blinded to the treatment. Endpoint data was found from aspirin patient logs, in-person visits, and telephone calls. Sepsis was defined as infection plus at least two of the four SERS criteria measured during a 24-hour period. This study design was completed before the Sepsis-3 definition was released. Case summaries related to endpoint events were prepared for adjudication from hospital records in which sepsis was described. They also looked at death certificate documents that indicated sepsis as a cause or contributor to death. Hospital records, including notes, vitals, discharge summaries, and pathology reports were evaluated, as well as searched following any participant's self-report of hospital admission or mention of sepsis infection or other similar diagnoses in primary care clinic notes. This trial used intention to treat univariate Cox proportional hazard models, as well as an adjusted analysis with variables selected a priori known to influence mortality. These variables included age, diabetes, current alcohol use, a history of cancer, and current smoking status. Since a participant may have had multiple hospital admissions associated with sepsis, the time to the first secondary endpoint was used as event time. Baseline characteristics were similar between the groups with notable low rates of diabetes, and with about 70% of both groups having a history of hypertension and dyslipidemia. Another notable characteristic is the very low percentage of non-white participants limiting generalizability. This trial looked at about 20,000 patients that were assessed for eligibility, and participants were enrolled from March 2010 to December 2014. About 16,000 patients were randomized, with about 8,000 assigned to aspirin and 8,000 assigned to placebo. Now, aspirin was stopped prematurely in June 2017 when an interim analysis showed no benefit in the intervention for primary outcome. 880 potential events were analyzed by investigators, finding 203 events meeting the a priori definition of death associated with sepsis, and 413 events to be classified as hospital admission associated with sepsis. When looking at the sepsis diagnoses themselves, it was predominantly due to pneumonia, a urinary tract infection, or bacteremia, and they found E. coli was the most commonly pathogenized. And they found E. coli was the most common pathogen identified. Of note, most pneumonias did not have a pathogen identified and were therefore confirmed based on clinical parameters and diagnostic imaging. A total of 912 deaths in this study were found, which like mentioned previously, 203 were attributed to sepsis, and this equaled about 1.2% of participants in both groups. Adjusting for the variable set a priori that we spoke about earlier had no effect on the hazard ratio comparing mortality associated with sepsis. The adjusted analysis was similar to the unadjusted and there were no missing covariate data. And you can see from the chart exactly what was just spoken about. 105 events were classified as repeated hospital admissions associated with sepsis. And as you can see in this graph, no difference was shown in the outcome. There was also no difference found in any of the other secondary outcomes mentioned previously. And while 61 participants were admitted to the ICU for sepsis, no difference was shown between these groups either. Adverse effects were reported in aspirin and therefore not reported in antisepsis. However, aspirin showed higher rates of clinically significant bleeding, including hemorrhagic stroke, with 8.6 events occurring per 1,000 patient years in the aspirin group compared to 6.2 events in the placebo group. Primary prevention strategy using daily lotus aspirin did not improve endpoints of death or hospital admission associated with sepsis. Now the authors had a couple of proposals for why this may be, one of which was due to mild cases of sepsis. As there were a few patients that were admitted to the ICU for septic events. Now this is plausible, however, they noted this may also be due to admission policies, as many older adults with advanced malignancy were often not admitted to the ICU. The findings of this trial are different from the previous trials mentioned earlier. And when comparing antisepsis to those trials, we see the opposite effect occurring when using aspirin as primary prevention. And this is suggested perhaps due to the different patient populations evaluated. As patients were younger in the others trials compared to antisepsis, where they had mean age ranges of 58 to 73, where antisepsis had to be 70 or above. However, no study has looked at age prospectively, and so we cannot attribute the difference to only that variable. It is notable that many deaths associated with sepsis occurred in patients with advanced malignancy, and 76 deaths were deemed to be associated with sepsis and had cancer as a proximal cause in the ASPR trial. The authors noted this, and after a post hoc analysis found that there was no effect of aspirin when cancer deaths were considered alone, and therefore there does not seem to be a misattribution of deaths associated with sepsis or associated with cancer. Compliance is another notable finding in this trial. Overall, 74% of patients were compliant with the assigned therapy. However, 4% of patients were also taking open-label aspirin at some point during the four-year follow-up period, and with our intention to treat analysis, this should not affect the study outcome found by the authors. And finally, this trial used SERS criteria to define sepsis and their outcomes, and this was appropriate as sepsis three definition had not yet been published when the protocol was completed. However, SERS criteria have poor specificity, and the authors noted that there were deaths associated with sepsis as in patients with bacteremia that recalled cases of suspected sepsis and not included in the primary endpoint since the SERS criteria were not fulfilled. This is especially important to note as patients over 70 years may not meet SERS criteria even when an infection is present. This may also have prevented identifying septic events in patients who were not admitted to the hospital. As patients going to a clinic may not have had blood work done, limiting SERS criteria, as well as may not have had cultures drawn, preventing these patients from being identified with a true infection when using the definition of infection given by authors in this trial. This trial design was limited by their ability to follow up with patients and patients being identified afterwards through record searches. As study personnel had to dig through the records to determine cause of death, this may have led to inaccurate inclusion or exclusion of patients. Sepsis is also a very multifactorial disease state, and there is no mention of how long a patient needs to be present with sepsis before passing for it to be associated with death. There is also no evaluation if patients had received adequate treatment at any point after the diagnosis of sepsis, and if they had been adequately treated but still passed during that same hospitalization course, if they would have met criteria for sepsis-associated death. While the authors did their best to define terms like sepsis and infections, the nature of the disease state lends itself to ambiguity when deciphering if there is a true association or if there's another explanation. So in conclusion, this trial did not find that aspirin used as primary prevention in patients over 70 years old reduced the rate of death or hospital admissions associated with sepsis. And I think it's an interesting hypothesis that we consider using aspirin for primary prevention, as we do use it in other disease states. And thinking through the mechanism of action of the medication, as well as the inflammatory state of sepsis, there may have been benefit. Unfortunately, this study did not support this hypothesis and no benefit was seen. And this brings me to my last polling question. Knowing the current literature available, should additional studies evaluate aspirin for prevention of death or hospitalization associated with sepsis? Please choose A for yes or B for no. Wonderful. I agree. I do not think we need to continue to evaluate this. I think it was a good hypothesis, and with the previous literature available, something that should have been investigated, but I don't believe this would be beneficial at this time. Thank you so much for your attention, and I will now open the floor for questions. Thank you, Emily, for that great presentation. Let's see if we have any questions. I guess I can start off. So did the authors report any difference in the rates of adverse events between the two groups? So in antisepsis, they did not report adverse events. They only reported aspirin adverse events, which did find higher rates of statistically significant bleeding in the aspirin group compared to the placebo group. All right. The next question is, did the authors address the standard of care for these patients, whether it be antimicrobial prescribing or other adjunct therapies? That's a great question, and you're talking about their treatment while they were inpatient? So speaking to that, the authors did not mention what was happening inpatient, as this was looking at prophylactic medication. And so we don't have data about what they may have received or appropriate prescribing of antibiotics or other sepsis treatments, which I agree would be beneficial to know. All right. I think that concludes our Q&A session. Thank you, Emily. Thank you. Thank you to our presenters today and the audience for attending. Please join us on the third Friday of every month from 2 to 3 p.m. Eastern Standard Time for the next Drone Club Spotlight on Pharmacy. And that concludes our presentation for today.

Journal Club Spotlight on Pharmacy

dexamethasone

ARDS

COVID-19

ventilator-free days

mitadrine

vasopressors

ICU patients

aspirin

sepsis