Hello, and welcome to today's Journal Club Spotlight on Pharmacy webcast, which is supported by the Society of Critical Care Medicine's CPP section. My name is Craig Whitman, Medical Respiratory ICU Critical Care Pharmacy Specialist at Temple University Hospital in Philadelphia, Pennsylvania. 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 after each of today's speakers. To submit questions throughout the presentation, type 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. You may also follow and participate in live discussion on Twitter following hashtag SCCMCPPJC and hashtag PharmICU. 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, physician, 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 Griffin Colmy, PGY-2 Critical Care Resident and St. Joseph Mercy Oakland in Pontiac, Michigan. Our second presenter is Sterling Torian, PGY-2 Critical Care Resident at University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences in Aurora, Colorado. And our third presenter is Kristen Brenninger, PGY-2 Critical Care Resident at Oregon Health and Science University Hospitals and Clinics in Portland, Oregon. And now I'll turn things over to our first presenter. All right, thank you for the great introduction. I guess Sterling and I are tag-teaming the balanced versus normal saline debate today. So thank you for joining me. I'm Griffin Colmy. I'm a PGY-2 Critical Care Pharmacy Resident from St. Joseph Mercy Oakland in Pontiac, Michigan. And my study I'm sharing with you is the results of the BASICS trial, specifically the manuscript looking at balanced solutions versus normal saline. So to set the stage a bit, I'll refresh everyone on the contents of fluids discussed in this presentation. So plasmalite we know most closely matches human plasma compared to the others. It's isoosmolar and uses acetate and gluconate as buffers, and it does not contain calcium. Meanwhile, lactated ringers does offer lactate as a buffer. It contains calcium, but overall is hyponatremic and slightly hypotonic. And then lastly, normal saline is both hyperchloremic, hypernatremic, and is exacidic. So in the past decade, I would say there's really three major trials in this space. That's the SPLIT, SMART, and SALT-ED trials, which all had enrollments of over 1,000 patients. There was also a Cochrane meta-analysis in 2019 that combined the results of all previous smaller trials with the SMART trial. So going over each, SPLIT was a small but high-quality RCT that did not find any advantages of plasmalite over normal saline in the ICU for either renal outcomes or in-hospital mortality. And then later on, SMART and SALT-ED came around, and they were a bit more pragmatic. There were sister studies that were unblinded and randomized in a cluster fashion based on the calendar month. So SMART was the ICU study, and then SALT was conducted in ED patients who were either admitted to a non-ICU service. And then in these studies, after admission, providers could switch fluids to their preference. So these two studies both found reductions in major 30-day kidney events defined as a composite of death, new renal replacement therapy, or persistent renal dysfunction. However, this was just a secondary endpoint in the SALT trial. And then lastly, in the 2019 Cochrane meta-analysis, the authors found that there was at least high certainty evidence that there was no effect of balanced solutions on mortality was what they stated, and there was no effect that they could find on AKI outcomes, albeit with a lower certainty of evidence. So I wanted to start with a question here. Does your institution have any policies or guidelines regarding criteria for use of one IV fluid over another? I'll just give everybody a couple seconds to answer yes, no, or you have no idea. You haven't looked into it yet. All right. So a few hospitals do have policies, but most generally don't. That's similar to my hospital. We don't currently have any policy on this. All right. And then so for the next audience question, at your institution, which IV solution is most commonly used for resuscitation of critically ill patients? Normal saline, lactated ringers, plasmolyte, AKA normal cell, or something else, maybe colloid. All right. So a lot of people are using lactated ringers, next sodium chloride, and then lastly plasmolyte. I would say my hospital, surprisingly, is probably one of the normal saline hospitals still. Alright, so that is getting us to the study today, next in chronologic order. This is titled the effect of IV fluid treatment with balanced versus saline solution on mortality and critically ill patients. So the central question of the study was, you know, for critically ill patients who require IV fluids, does the use of balanced solutions over normal saline improve survival, prevent kidney injury, or improve SOFA scores? So this study was a large multi-center double-blind randomized controlled trial across 75 different Brazilian ICUs. It actually had a 2x2 factorial design with a sister paper that was published examining the rate of the bolus infusion. So then this manuscript specifically, you know, included patients admitted to the ICU who were, you know, required fluid expansion and had at least one risk factor for AKI, which I have listed on the side. They excluded patients who were already expected to require renal replacement therapy, anybody who had a severe sodium or potassium derangement at baseline, or anybody that was palliative care only, expected to die within 20 hours or already pronounced brain dead. So then, you know, of the plasmolyte versus normal saline part of the factorial design, this study randomized 10,520 patients in a one-to-one ratio to plasmolyte 148, also known as plasmolyte A, versus 0.9% sodium chloride, also known as normal saline. And their primary endpoint was 90 days survival, while their secondary endpoints included new renal replacement therapy, AKI, and SOFA scores, which were assessed separately on days 1 through 3 and day 7. And then they also had some tertiary outcomes, which included ICU mortality and different lengths of stays. And lastly, their power was calculated based on an estimated 35% 90-day mortality. So I think it's really important to understand who exactly was in this study. So looking at age and gender, those were well balanced. And then going down the APACHE-2 score, the patients were mostly on the lower end of the spectrum, corresponding to a 7 to 15% mortality risk. And moving on, most patients were actually surgical patients, with almost half overall being elective surgery. 20% had sepsis, and about 5% had TBI. Almost half the patients were intubated during their ICU admission, and about 20% had elevated baseline creatinine. And then over the first three days of the study period, on average, patients received 4.1 liters of IV fluids, with 2.9 liters of that being the study fluid. The rest, they did allow sodium bicarbonate, dextrose, albumin, or colloids per the treating team's discretion. And then there was some pre-randomization fluids. So there was actually a significant chunk of patients in each group that did receive the opposite fluid prior to randomization, as you can see. So then moving on to the primary outcome of 90-day mortality, approximately one quarter of patients expired in both groups, and there was no statistical difference found. So then for the secondary and tertiary outcomes of interest, there were no differences in renal replacement needs, aka vent free day, length of stays, or in-hospital mortality. However, there was a slight difference in day 7 SOFA score that they found, which was not found at other assessment intervals. You know, I would really suggest caution with the secondary finding here, as they did have at least 32 variations of these secondary and tertiary outcomes at different measurement intervals. These weren't corrected for multiple hypothesis testing. So the other finding that you might have heard was here in the subgroup results. So the fluid choice did not have an effect on 90-day mortality when it was broken down into patients with AKI specifically, sepsis, or surgery versus non-surgical patients. However, there was a significant difference in patients with TBI, you know, having a higher association with mortality if given balanced solutions. And again, you know, caution, which should be exercised here as well, since there wasn't a correction for multiple hypothesis testing. So it could be a chance finding. However, you know, I would say this finding does have a plausible mechanism as the sodium content and tonicity of the two solutions are different, which could potentially affect the development of cerebral edema. And so the authors concluded from these findings that among ICU patients requiring a fluid challenge, the choice of balanced solution versus normal saline did not affect 90-day mortality. And I actually go further to suggest that the findings don't support the use of plasmolyte 148. So as far as the strengths of the BASICS trial, number one, the study was far more methodologically rigorous than the SMART and SALT since there was blinding at all levels and there was no crossover of patients. Number two, the study assessed many different types of ICU patients and they also analyzed subpopulations of interest like sepsis and TBI. And number three, the study used higher fluid volumes than SMART with the median of 4.1 liters over 72 hours versus in SMART, patients on average got about 1 liter over the entire duration of hospitalization, which was, you know, so this study is more in line with resuscitation volumes that I'm familiar with. And, you know, it would demonstrate a greater effect of the underlying fluid choice. So then number four, you know, I did think the study groups were very well balanced and then lastly the authors did write their statistical plan prior to the study, so I think there's a low risk of interpretive bias here. So then my favorite part, you know, I'm excited to get to this one, is the limitations. And so number one, you might have noticed that sepsis was a minor contribution and number two, half of all the patients were elective surgery. So this might be a little bit different from your ICU practice. Moving on to number three, on average each patient got about 1 liter of non-study fluid in addition to the 3 liters of study fluid. So, you know, you'd expect that to dilute out the results, you know, and a full pun intended here. So then number four, plasmolyte does have some differences from other balanced solutions like lactated ringers since it uses acetate and gluconate as buffers instead of lactate. So, you know, it remains to be proven if any of those differences are significant or if you could really extrapolate this to lactated ringers. And then moving to number five, this trial was actually a bit smaller, if you've noticed, than the SMART trial. So it was about 11,000 versus 16,000 patients. Moving on to number six, the subgroups such as the TBI finding, you know, weren't corrected for multiple hypothesis testing. So interpret at your own risk. Number seven, the power was calculated for a mortality of 35% versus 27% in the actual study. So it might have been a little bit underpowered. And then lastly, the differences in serum chloride were only about three MEQs per liter. So this study doesn't really answer if the development of hyperchloremic acidosis itself is, you know, associated with differences and outcomes there. So, you know, what's the big takeaway message? Overall, the BASICS trial is the most methodologically sound study to date and generally refutes prior findings from the SMART and SALT trials. So, you know, the message I wanted to share to you guys is that this was on the whole a neutral study. So neither fluid was inferior overall. Therefore, you know, I think that each fluid still has a role in therapy and fluid choice, you know, should continue to be individualized. So in my practice, I'm no longer going to strongly, you know, push for balanced solutions unless the patient has established hyperchloremic acidosis. And, you know, some other things based on these results, it would be also reasonable to avoid balanced solutions in favor of non-hypotonic saline solutions in traumatic brain injury. And lastly, some other factors that are going to influence your fluid choice here are going to be obviously cost, any admixture compatibility, and obviously provider preference. And then lastly, the upcoming PLUS trial did recently complete enrollment, and this one's slightly smaller, but it will likely further contribute to the fluid choice discourse. So I'm looking forward to that one too, and I'm looking forward to any meta-analyses that are going to come out of these. All right, so thank you everyone for listening. I will pass the baton. All right, thank you Griffin. And you did mention the amount of study fluid that was administered to the patients here. Can you comment a little bit more on the total amount of study volume that was given to these patients, and is it consistent with what you typically see in your practice as far as the amount of fluid that the patients receive? Yeah, so it, you know, based on what I read in the study, on average patients in the first day got about two liters, and then on the second day it was, you know, half that, and the day after that, you know, a quarter of that. So I think that's similar to how we do things in practices, you know, giving liters up front and then slowly tapering off and de-resuscitating, you know, as the days go on. So I think that was familiar with, you know, what I'm, what I see in practice. I think the volumes were, you know, appropriate. Okay, thank you for that. And the other question I have for you is regarding the overall severity of illness in the study population, would you apply the same sort of thought process in selecting one fluid over another in, say, a patient with septic shock on multiple vasopressors at a high risk of death with multiple risk factors for kidney injury, or would you stay consistent with what you're saying from the overall conclusions from the study? Yeah, so I think that's the golden question here, is obviously these patients weren't super sick on average. So, you know, if you have somebody that has profound renal impairment and has basically lost their compensatory mechanisms for their acid-base balance, you know, I could still see a role for balanced solutions there, especially if, you know, clinically you start to see abnormalities, you know, in their serum chloride or bicarb, things like that. So I still see a role for both balanced and normal saline. Okay, so I think those are all the questions that we have. Thank you so much. That concludes our Q&A session. Thank you, Griffin. All right, thank you. All right, before moving to our next presenter, we would like to ask a brief polling question regarding today's attendance to gain a better understanding of our overall attendance to ensure continued support of the Spotlight on Pharmacy webcast. How many people at your site were watching today's webcast? Now I'd like to introduce our second presenter, Sterling Torian. Hi, thank you for that introduction. Today I'll be presenting on shining a light on fluid resuscitation, which is on the SCOPE-DKA trial. So when you think of DKA, majority of people probably put insulin first on their list. But I think kind of flipping that and thinking of fluids first is probably the best way to think of DKA. And this is really due to the huge amount of fluid losses that are occurring in diabetic ketoacidosis. And this is majority due to hyperglycemia resulting in osmotic diuresis and dehydration. The current guidelines are fairly outdated from over 10 years ago, but they are still recommending sodium chloride over a balanced electrolyte solution. And as you can see here, these large amount of fluid volumes are really required to replace that fluid loss. And as Griffin already hit on a little bit, there are a lot of clinical controversies going on in the realm of critical care. And so this is a very popular topic that is becoming very widely discussed and very widely looked at when it comes to literature. The first controversy is this hyperchloremic metabolic acidosis. And this is pretty widely established at this point and due to the high chloride load of sodium chloride. And you can also get the subsequent acute kidney injury as we've already kind of discussed with this hyperchloric metabolic acidosis and the low pH that is resulted from NS. The hypertensity, we've already discussed a little bit too when it came to TBI. However, thinking about it in realms of DKA is also important. And so when you think about a DKA patient, they're not necessarily extravasating fluid into different compartments. It's more of a actual total fluid deficit. And so the hypertensity actually may play down to a negative effect in a DKA patient. And lastly, and probably one of the more interesting discussion points of today is this ketone production mechanism of action that is proposed with plasmalite. Plasmalite has an acetate component that is primarily metabolized to citrate. However, there is an alternative pathway with acetyl-CoA that can potentially result in acetoacetate production and further ketone production. So my first question that I wanted to pull the audience with is what is your current institution's initial choice of fluid resuscitation in DKA? All right, and I feel that this is pretty consistent with what we see in the literature as well, or what we see in actual practice is that sodium chloride is typically preferred. As far as previous literature, there are three main pieces of literature that I wanted to discuss with you guys today. The first was a study published back in 2011 that looked at adults with moderate to severe DKA. And these patients received either plasmalite or normal saline and had significantly higher increases in chloride with normal saline, which is not something that would be unexpected. The next study is a recent trial that was published in 2020 that primarily focused on children with mild, moderate, or severe DKA and randomized patients to normal saline or plasmalite. There was essentially no difference in the incidence of acute kidney injury or resolution of DKA and therefore kind of showed that equal efficacy in those two groups. And lastly, this most recent study was a subgroup analysis of DKA patients from the SALT-ED and SMART trials that Griffin mentioned earlier. This looked at a subgroup analysis of patients who received normal saline compared to the balanced crystalloids, which was primarily LR. And their conclusion from that subgroup analysis was that there was a shortened time to resolution with balanced crystalloids over normal saline. And so as you can tell, all of this literature is very heterogeneic, some showing benefit, some showing no difference. And so there really was need for this consolidated randomized controlled trial in adult patients. Which brings us to the SCOPE-DKA trial that was recently published in Intensive Care Medicine, in which the study hypothesis was that critically ill patients would have a faster resolution of severe DKA with plasmalite in comparison to sodium chloride. This study was a cluster crossover open label randomized controlled trial that randomized patients from zero to 13 months at each institution. And so for the first six months at that specific institution was randomized to either plasmalite or normal saline. There was a one month washout period, and then that institution received the opposite of what was received in that first six month period. And I've listed the inclusion and exclusion here, criteria here for you to review. But essentially, what's most important is that these patients were mostly severe DKA with a pH of less than 7.25 and a blood glucose of greater than 252. As far as outcomes and analysis, the primary outcome, the primary outcome was change in base excess to greater than negative 3 milliequivalents per liter at 48 hours post ICU admission. Secondary outcomes mainly focused on those clinical questions that we have regarding length of stay and mortality, as well as DKA resolution. And the analysis was a modified intention to treat analysis with patients who had complete outcome data for the primary outcome. The last point I would like to take a second on to talk about is the analysis. So the authors do mention that this analysis was not powered to detect differences. So they did not run a power analysis to determine how many patients would need to be enrolled, and was really more of a hypothesis driving study. And so we're going to talk a little bit more about that in a second. A total of 90 patients were enrolled from seven sites in Australia. And I've posted some of the demographics here for you to review. But the biggest takeaways I'll present to you are that primarily this was a young female type 1 diabetic population. And patients had, there were patients in the plasmolyte group that had more than type 1 diabetes. And the ketones and anion gap were significantly higher in the normal saline group as well. The primary and secondary outcomes, if you remember from earlier, the primary outcome was the base excess of greater than negative 3 at 48 hours. And this was not shown to be statistically different between groups with only a difference of 10%. Where we did see a difference was base excess at 24 hours, where a higher percent of patients at 24 hours had met that threshold, with 69 patients in the plasmolyte group and 36% of patients in the normal saline group. And lastly, I would like to say that they did analyze DKA resolution by the ADA criteria and found no significant differences between the two groups. As far as clinical outcomes and compliance, I posted these here for you to see, but essentially no differences between ICU length of stay, hospital length of stay, mortality, adverse effects. Where we did see a bit of a difference was median compliance to study regimen, with 66% of patients on the normal saline group. And the median compliance to study regimen of patients on a median having compliance to the plasmolyte group versus 100% in the normal saline group. Some other pertinent findings before we get into some of the discussion and conclusion I wanted to present to everyone are the changes in ketones during the first 48 hours. So if you remember earlier, we did have this proposed hypothesis that ketone production may be higher in the plasmolyte group. But as you can see in this graph on the left, that was really not visually demonstrated with the plasmolyte group in this study. The other outcome I wanted to present is the changes in serum chloride, in which you can see the sodium chloride group having an obvious increased risk of having that hyperchloremia from like 6 to 12 hours, and then balancing off towards that 48-hour mark. So the author's conclusions of this study was that when compared to normal saline, plasmolyte 148 may lead to a faster resolution of metabolic acidosis and severe DKA without the increased risk of ketosis. There are a few important things that I think are up for discussion with this trial. And first, I would like to say I do think that this is a very relevant and important topic and debate within the current realm of critical care. As we've already discussed, we looked at the basic trial, which assesses all critically ill patients and looking at the difference there. But I do think DKA is a specific niche where they are requiring these large volumes of fluids, and is this hyperchloremic risk really contributing to our time to DKA resolution? Another important point that I wanted to hit on was the institutional specific protocols. So each group really, each site had their own institution specific protocol that they followed for DKA specifically. And each, the time to change to a D5 fluid and the specific rates of insulin were not mentioned in this study. The next point I wanted to talk about was the choice of balanced electrolyte solution that they chose as a comparator, which was plasmolyte 148 versus plasmolyte A versus lactated ringers as a comparator. And so we've already looked at these contents multiple times throughout this afternoon. But what I wanted to point out to you guys is this difference in pH that is seen with comparison to plasmolyte A versus plasmolyte 148. And it's not exactly described as to why these pHs differ well in the literature. Plasmolyte 148 is more commonly used in Australia. I do not believe it has, it is readily available in the United States. So we're more familiar with plasmolyte A anyway. But it is interesting that they do have these differences in pH, but no difference in content. So it may be just due to the pH testing itself. However, it is an interesting point that plasmolyte 148 may have more of a pH balance similar to normal saline. The hypothesis and primary outcome really didn't align with what the authors had initially mentioned, which if you remember from one of my first slides, the hypothesis was that the plasmolyte group would have a faster time to decay resolution. However, the primary outcome was base excess or percentage of patients that received base excess resolution. And base excess is not a typical outcome that we look at for decay patients specifically. So the misalignment there, I feel like the authors could have picked a different hypothesis or primary outcome to really help align the study and make it more consistent. And the last two points, unequal baseline characteristics. We talked about how that plasmolyte group had a higher anion gap and higher ketone production, which could have led to why they didn't see a faster time to resolution. And the low compliance to the plasmolyte 148 group was that the low compliance could have contributed to overall the poor outcomes of the study. And so clinical interpretation, base excess correlation between decay resolution is not a common association. And then the benefit and risk of these fluids was really not well discussed in the study. So we don't really know if there is this benefit of less hyperchloremia in DKA and if this leads to an AKI risk because this was not mentioned as well. So main takeaway points that I concluded from this study was that balanced electrolyte solutions such as plasmolyte 148 trended towards improvement in surrogate outcomes, but further data would be needed to confirm clinical outcomes such as quicker resolution of DKA or the incidence of AKI. So my last polling question for you guys is, does this trial sway you to consider balanced electrolyte solutions as a primary resuscitation fluid for DKA? And I would agree that for my practice, I think quite, there's not enough evidence quite yet to convince me to use plasma light or a balanced electrolyte solution over sodium chloride. I think it's certainly reasonable and don't think it's wrong by any means. So I think that is appropriate assessment for the study. And with that, I will take any questions you guys have. All right, thank you, Sterling. I wanted to ask you a question based on something you mentioned. One of your critiques was the primary outcome that they used in the study. If you were to design the study again, what specifically would you use as the primary outcome that you would think would give more meaningful results? Yes, I think that is a great question and something I definitely thought about as I was reading through the study with the base excess. So as we already mentioned, it's not your typical outcome with DKA. And so if I were to redesign it, ideally, I think the most clinical question that we have, especially in the realm of DKA, is how quick can we get these patients to a DKA resolution? So if I were to redesign it, I think my primary outcome would be DKA resolution and using the definition that the ADA has for criteria. So blood glucose, less than 200, anion gap closure, all of those things. And I really feel like the authors missed their mark a little bit on the primary outcome. Okay, and thank you for that. And one other question I have is, you mentioned that there were differences in baseline characteristics between the groups. So what specifically do you think could have contributed to any lack of difference that maybe we saw or influence on the results based on the baseline characteristics? Yeah, I think that primarily in this study too, it was relatively small, kind of consistent with all of the previous DKA literature we had seen. So certainly I think it could have been driven by the small population size that was seen as only about 45 patients were in each group, which definitely could contribute to an unequal baseline characteristic population. So I think that was probably the biggest driver in the differences between groups, although there could be some selection bias, although most of these patients were randomized by computer generation. So there shouldn't necessarily be selection bias. Okay, thank you. All right, so that concludes our Q&A session. Thank you, Sterling. All right, thank you. Now I'd like to introduce our final presenter, Kristen Brenninger. Thank you for that introduction. Today I'm going to discuss the COVID steroid two trial, effective 12 milligram versus six milligrams of dexamethasone on the number of days alive without life support in adults with COVID-19 and severe hypoxemia. The current NIH guidelines recommend administering dexamethasone six milligrams IV or PO once daily to hospitalized COVID-19 patients in order to modulate this inflammation mediated lung injury that we see. Additionally, previous trials have demonstrated reduced progression to respiratory failure and death. Of significance is the recovery trial in which a mortality benefit was observed in COVID-19 patients who received that six milligrams dexamethasone once daily. And that's demonstrated by the image on the right. A prospective meta-analysis of seven international trials at evaluating critically ill patients with COVID-19 also found a mortality benefit with the use of steroids when they compared that to usual care or placebo. And that was with a significant odds ratio of 0.66. The trials included, they used a dexamethasone equivalent dose of six to 16 milligrams. However, I should note that the majority of patients included were actually from the recovery trial. As you can see here, the overall results are largely driven by the results of the recovery trial with a 57% weight. Higher doses of dexamethasone have also been reported as beneficial in a randomized trial, including patients without COVID-19 who also had acute respiratory distress syndrome. When higher doses were compared to six to 16, compared to standard treatment, results indicate that higher doses of dexamethasone resulted in significantly fewer ventilator days, suggesting the possibility that higher doses of dexamethasone than that recommended six milligrams per day may benefit patients with COVID-19 who have more severe disease. So why not administer higher doses of dexamethasone to these COVID-19 patients? Well, there are concerns about the side effects with higher doses of steroids, particularly severe fungal infections. Therefore, the COVID Steroid 2 trial was conducted to evaluate the efficacy and safety of a higher dose of dexamethasone in hospitalized adults with COVID-19 and severe hypoxemia. The research question here asks, what is the effect of 12 milligrams versus six milligrams of dexamethasone on the number of days alive without life support at 28 days in patients with COVID-19 and severe hypoxemia? The authors hypothesized that a higher daily dose of dexamethasone would increase the number of days alive without life support at 28 days in these patients. This is an international parallel group stratified and blinded randomized controlled trial, and it takes place or took place in 26 international hospitals. The randomization in the two dose groups occurred in a one-to-one ratio and in block sizes of six or eight. The stratification of the patients was based on trial site, age, and use of mechanical ventilation at screening. And of note, which is always important to consider in COVID research, is the timing that this took place. So between August, 2020 and May, 2021. Eligible patients were 18 years or older, hospitalized with confirmed COVID-19 infection and required supplemental oxygen of at least 10 liters per minute, non-invasive ventilation or continuous positive airway pressure for hypoxemia or invasive mechanical ventilation. Those excluded were patients who were treated with systemic glucocorticoids in doses higher than that of six milligrams of dexamethasone equivalents, and that's for indications other than COVID-19, or had been treated with systemic glucocorticoids for COVID-19 for five days or longer prior to enrollment, had an invasive fungal infection or active tuberculosis, known hypersensitivity to dexamethasone or if they were pregnant. The intervention group consisted of 497 patients in the intention to treat population who received 12 milligrams of dexamethasone daily for a total of 10 days or a maximum of 10 days. While the control group consisted of 485 patients in the intention to treat population who received six milligrams dexamethasone for that 10-day mark. So if patients had used dexamethasone for COVID-19 prior to enrollment, the intervention period was reduced so that patients received dexamethasone for a maximum of 10 days. Therefore, that's why the median days of therapy was seven for that high-dose group or the intervention group and six days for the standard dose or control group. And then all other, just of note, all other interventions were at the discretion of the clinicians and other immunosuppressive agents for COVID-19 were not recommended at that time. The primary outcome was number of days alive without life support at 28 days, which is defined as an use of invasive mechanical ventilation, circulatory support, kidney replacement therapy. And then some notable secondary outcomes that we will discuss include mortality at 28 and 90 days and then serious adverse reactions at 28 days. But as you can see, there were a number of additional secondary outcomes measured. A sample size calculation estimated that 1,000 patients were required for the trial to have 85% power to show a relative reduction of 15% in 28-day mortality. The primary and secondary outcomes were analyzed using the Krieger-Jensen and Lange test and were actually adjusted for stratification variables, the ones we had discussed earlier. Ultimately data for the primary outcome were obtained for 971 patients. So that's only 491 in the 12 milligram group and 480 in the six milligram group. So just under the number of patients required for power. At 28 days after randomization, the median number of days alive without life support was 22 days in the 12 milligram group and 20.5 days in the six milligram group. The results were similar in those pre-planned and post-hoc sensitivity analyses. Additionally, in the predefined subgroup analyses, no statistically significant heterogeneity was found for the effect of high-dose steroids on this primary outcome. The percentages of patients with 28 days alive without life support were 42.6% in the 12 milligram group and 40.2% in the six milligram group. This figure shows the number of days alive without life support as proportions when they are stacked horizontally. And that's in both intervention groups in the intention to treat population, which as I said is 971 patients. The red represents worse outcomes and the purple represents better outcomes. In terms of secondary outcomes, at 28 days, a total of 133 of the 491 patients or 27% in the 12 milligram group and 32% in the dexamethasone six milligram group had died. Additionally, at 28 days, 56 of the 497 patients in the dexamethasone group or 11.3% as compared with 13.2% in the six milligram group. Had one or more serious adverse reactions. And again, no significant differences between comparators. At 90 days, the median number of days alive without life support was 84 days in the 12 milligram dexamethasone group and 80 days in the six milligram dexamethasone group. There are 14 patients who were not followed up for this 90 days and they were included to the last day that they were known to be alive. And so that was seven patients in each group. So loss to follow up was similar in both groups. Again, not statistically significant. At 28 days, 56 of the 497 patients in the 12 milligram group had one or more serious adverse reactions compared with 65 of 485 in the six milligram dexamethasone group. And those adverse events included new episodes of septic shock, invasive fungal infections and clinically important GI bleeds. And none of the patients had an anaphylactic reaction to dexamethasone. Based on these results, the authors concluded that among patients with COVID-19 and severe hypoxemia, 12 milligrams dexamethasone compared to six milligrams dexamethasone did not result in statistically significantly more days alive without life support at 28 days. And so it's important to consider some of the limitations. Firstly, the trial may have been underpowered to identify a significant difference. This is based on assumed number, as we all know. The sample size estimation for the primary outcome was based on expected relative difference of 15% in the mortality at 28 days and 10% in time required in life support. So these differences may have been too large as an assumption to begin with. Secondly, the intervention period was only six days in some patients per protocol. And that's because as described, the trial design allowed up to four days of dexamethasone use before enrollment. And so that may have reduced any effect of the intervention. And then third, the distribution of primary and secondary outcome data was not normal. And so in order to mitigate this, they utilized some newly developed statistical tests that accounted for data sets that had a lot of zero values. And post-hoc analyses were used to further test those results. Use of IL-6 inhibitors was not recommended during the majority of this trial period. And so the results can lack some of that external validity as current recommendations have since changed in terms of standard of care for COVID-19. Moving forward, vaccination status will be an important patient characteristic to pay attention to and account for in any COVID-19 trial. Lastly, this is a non-U.S. trial, and so may not represent U.S. patients, particularly in regards to weight. A lot of those older ARDS literature is actually weight-based steroid dosing. To summarize, the design was a randomized controlled trial, multi-center. They had blinding of staff, patients, and researchers. They conducted multiple sensitivity analyses. They did per-protocol analyses as well as intention-to-treat analyses, and they had pre-specified subgroups. The fairly large sample size, although it didn't meet power, they had minimal loss to follow-up, and the secondary safety outcome is very beneficial. But overall, it's pre-Delta, non-U.S., and the median age was 65, so external validity is something that should be considered here. In patients with severe COVID-19 who were predominantly not treated with tocilizumab, the use of 12 milligrams compared with 6 milligrams dexamethasone resulted in no significant difference in the number of days alive without life support at 28 days or mortality at days 28 and 90. The confidence intervals, though, as we saw, do suggest the potential that there may be a benefit with higher dose dexamethasone, but the trial was ultimately underpowered to determine this. If there is a difference between the two groups, it's much less likely that 6 milligrams would give a benefit compared to 12 milligrams. As there is no significant difference in side effects between treatment groups, it could be argued that we should change to 12 milligrams dosing for dexamethasone. However, there were, as I kind of stated but want to reiterate, there were low numbers of patients treated with tocilizumab in this trial. Following the recovery and remap CAP trial, that's now a standard of care. Additionally, as discussed, pre-Delta, non-U.S. trial, vaccination status has not been accounted for, and so ultimately we don't know the impact that these factors would have on our overall results. So I will start with my first assessment question. Dexamethasone 12 milligrams per day was associated with the greater Days Alive at 28 days but resulted in more adverse effects. Yeah, so there were not, there was no, the Greater Days Alive was a trend. It wasn't statistically significant, and the adverse effects were fairly similar between the two groups. So I'd agree with this assessment. And the next question. The impact of tocilizumab is unclear due to a low number of patients treated in this trial. Yeah, I agree. And that's only one of the things that have changed. COVID is, as we all know, a very quick evolving disease state. And so it's important to consider changes that have occurred since trials have been completed when we're trying to apply these results to our patient populations. All right. Thank you. And now I would love to take any questions. OK, very nice job, Kristen. First question is, this trial did not meet power and did not find statistical significance. What type of error is this trial at risk for and what are the implications of such? Sorry, can you I think you cut out at the very end there. So it's. Sure, I can I can repeat the question. This trial, this trial did not meet power and did not find statistical significance. What type of error is this trial at risk for and what are the implications of this? OK, so this would be a type one error. And so if it's this would mean that that we would be concerned for rejecting the null hypothesis when it's actually true. And so the type one error would mean that, yeah, we would reject the null hypothesis when it's actually true, whereas a type two error means that we are failing to reject the null hypothesis when it's actually false. And so in this case, we would not we would not be able to see a difference if there really was a difference. OK, so it seems like it's more of a type two error, right, because we didn't find a difference between the two groups there. So what do you think some of the implications of that are? Sorry. Yeah, I guess I misspoke there. Yeah, so the null hypothesis would be that there is. I guess I'm confused about the moving forward with the question. OK, there's a couple of other questions that we can move on to for sure. Given that there were no differences in adverse effects, are there any subgroups assessed that you feel may benefit from the higher dose of corticosteroid? Does the use of an alternative immunomodulator affect your decision? Yeah, so one particular subgroup was the ECMO patients. And so these patients actually much, much fewer patients received ECMO in the high dose steroid group. And so that might be that might be a situation or a patient population where I would consider using steroids. The second part of your question was related to. Does the use of an alternate immunomodulator affect your decision? Yeah, I think as of now, we don't have enough evidence to support the use of the high dose steroids in the immunomodulator in the patients who are receiving immunomodulators. And so it might be it might not be the best idea to use the higher dose in the Tocilizumab patients or those receiving other immunomodulators. I think we would need additional studies and focus on the patient population that that is receiving immunomodulators. OK, thank you. And third and final question that we have is, will your institution be implementing 12 milligrams of dexamethasone for critically ill COVID-19 patients? If so, will there be any specific subsets that will be included or excluded from this practice? Yeah, as of now, there are no plans to initiate the 12 milligram dexamethasone dosing regimen. There have been some providers who have started to talk about it. And I believe a couple of patients have received that higher dose of dexamethasone, but it's not a standard at this time. OK. All right. Thank you. And thank you so much, Kristen. That concludes our Q&A session. Thank you. So, in conclusion, thank you to our presenters today and to the audience for attending. Please join us on the third Friday of the month from 2 to 3 p.m. Eastern Standard Time for the next Journal Club Spotlight on Pharmacy. That concludes our presentation today. Have a great weekend.

Journal Club Spotlight on Pharmacy

BASICS trial

balanced solutions

normal saline

critically ill patients

SCOPE-DKA trial

dexamethasone

diabetic ketoacidosis

COVID Steroid 2 trial