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 Ashley Selby, Assistant Professor of Pharmacy Practice at Texas Tech University Health Sciences Center, Jerry H. Hodge School of Pharmacy, and Cardiopulmonary ICU Clinical Pharmacy Specialist at the VA North Texas Healthcare System in Dallas, Texas. I will be moderating today's webcast. A recording of this webcast will be available to registered attendees. Log in to 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. 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 Kelsey Dobbins, a PGY-2 critical care resident at Wake Med Health and Hospitals in Raleigh, North Carolina. Our second presenter is Teresa Benitez, PGY-2 critical care resident at WellSpan York Hospital in York, Pennsylvania. And our third presenter is Joseph Morata, PGY-2 critical care resident at Banner University Medical Center, Tucson, in Tucson, Arizona. And now I'll turn things over to our first presenter. Good afternoon, everyone. Good afternoon, everyone. Today I'll be presenting the risk of overcorrection and rapid intermittent bolus versus slow continuous infusion therapies of hypertonic saline for patients with symptomatic hyponatremia, otherwise known as the SALSA trial. Hyponatremia is the most common electrolyte abnormality in hospitalized patients and if it's left untreated is associated with significant morbidity and even mortality. However, treatment of hyponatremia also carries risk. As if serum sodium is corrected too quickly, it can lead to severe adverse events such as osmotic demyelination syndrome. The safest and most effective treatment for hyponatremia has been a longstanding debate. The 2013 American Expert Panel and the 2014 European Guidelines both recommend treating severe symptomatic hyponatremia with rapid intermittent boluses of 3% sodium chloride. However, these two guidelines differ in their recommendations for mild or moderate symptoms with the American guidelines recommending a slow continuous infusion and the European guidelines recommending to continue with rapid intermittent bolus. These recommendations are largely based off of small trials, case reports, and expert opinion. And what we've seen in practice is that slow continuous infusion still tends to be preferred despite guideline recommendations. The data comparing these two treatment strategies head to head is very limited. So before I get into the SALSA trial, I wanted to ask which strategy your institution predominantly uses to manage symptomatic hyponatremia. All right, so as expected, it looks like most institutions are using slow continuous infusion. Although there is a pretty good percentage that are using both treatment strategies. So, the purpose of the salsa trial was to compare these 2 strategies head to head. The investigators hypothesized that rapid intermittent boluses would carry a lower risk of overcorrection compared with slow, continuous infusion when used to treat symptomatic hyponatremia. This was a prospective multi center, randomized open label study conducted in 3 hospitals in the Republic of Korea. Patients were included if they were adults with a glucose corrected serum sodium less than or equal to 125 and if they had moderate or severe symptoms. Moderate symptoms were defined as nausea, headache, drowsiness, weakness, and severe symptoms were defined as vomiting, stupor, seizures, or coma, which was defined as a GCS less than or equal to 8. Some key exclusion criteria were patients with primary polydipsia patients with liver disease defined as LFT is greater than 3 times the upper limit of normal or decompensated cirrhosis or patients that had a history of cerebral trauma or an increased intracranial pressure within the previous 3 months. After patients were screened for inclusion, they were randomized 1 to 1 to receive either rapid intermittent bolus or slow, continuous infusion. Patients in the rapid intermittent bolus group with moderately severe symptoms received a single bolus of 3% sodium chloride at 2 milliliters per kilogram. If these patients had severe symptoms, they received that same bolus times 2. In the slow, continuous infusion group, if patients had moderately severe symptoms, they received an infusion at 0.5 milliliters per kilogram per hour and if they had severe symptoms, that infusion rate was increased to 1 milliliter per kilogram per hour. Both treatment arms had additional treatment criteria based off of the rate of increase in serum sodium and the timeframe after treatment initiation, and then re-lowering treatment was initiated if the serum sodium increased by 10 or more at hours 1 through 24 or by 18 or more at hours 30 through 48. The primary end point was the incidence of overcorrection, which they defined as an increase in serum sodium of 12 or more in 24 hours or 18 or more in 48 hours. Key secondary end points were the presence of symptoms at 24 and 48 hours, the incidence of target correction rate, which was defined as an increase in serum sodium of 5 to 9 in 24 hours or 10 to 17 in 48 hours, the need for additional treatment, the need for re-lowering treatment, and the incidence of osmotic demyelination syndrome, or ODS. The authors performed both an intention to treat and per protocol analysis, and they calculated that they needed 89 patients in each group to achieve 80% power to detect a significant difference based on a predicted overcorrection rate of 5% in the rapid intermittent bolus group and 20% in the slow continuous infusion group. They also assumed a dropout rate of 15%. Key baseline characteristics are listed in this table. You can see that patients with a variety of different causes of hyponatremia were included, but the two most common were thiazide use and SIADH. Three-fourths of the patients had moderate symptoms of hyponatremia, and the mean baseline serum sodium was 118 in both groups. The majority of patients were initiated on treatment in the emergency department, with only one patient being initiated on treatment in the ICU, and the dropout rate was 17.2% in the rapid intermittent bolus group and 19.8% in the slow continuous infusion group. When the investigators analyzed the primary endpoint incidence of overcorrection, they did not find a significant difference in either the intention to treat or the per protocol group, although you can see that the incidence of overcorrection was higher in both groups than what was predicted based off of their statistical analysis, and if you look at the curve, slow continuous infusion tended to have a numerically higher incidence of overcorrection than rapid intermittent bolus at pretty much every treatment group. There was also no significant difference in either the intention to treat or per protocol analysis in presence of symptoms at 24 or 48 hours, or the incidence of target correction rate. However, patients that received therapy with rapid intermittent bolus tended to need additional treatment more often, but needed re-lowering treatment less frequently than patients that receive slow continuous infusion. There were no incidents of ODS at all in the study. The total amount of 3% sodium chloride was similar between groups after 48 hours, with a little over 500 milliliters being administered. The investigators also allowed additional fluids to be administered throughout the study timeframe, and similar amounts of additional fluids were administered in both groups at both the 24 hour and the 24 to 48 hour mark. The investigators did perform a post hoc analysis, and they found that significantly more patients in the rapid intermittent bolus group achieved a target correction rate within the first hour compared to the slow continuous infusion group, with a number needed to treat of seven. However, when they performed the same analysis in the per protocol population, they did not find a significant difference. One of the major strengths of this study is that it is the first randomized multi-center study that was performed looking at these two treatment strategies head to head. Although, due to issues with blinding these treatments, this was an open label study, and there were some changes in treatment protocol as the study progressed. So, for the first 2 years of the study timeframe, they only included patients in the emergency department. But after the 2 year mark, they expanded to include patients that had hospital acquired hyponatremia, and also included an additional hospital to increase enrollment. So, this could have potentially introduced some confounders. The baseline characteristics were generally well balanced across both study groups. However, the authors did not specify the percentage of patients that were ultimately admitted to a hospital ward or ICU, so we don't have a lot of information about the need for increasing levels of care. They also excluded patients with cerebral edema and liver disease, which are two patient populations that may receive these treatment strategies. So, these results cannot be generalized to those patients. And they had a pretty high dropout rate that was higher than even what they predicted, which could have potentially underpowered the study to detect a difference in the results. Overall, the interventions chosen were appropriate based off of guideline recommendations. However, they did allow a high utilization of additional fluids, which could have potentially confounded the results. The types of additional fluids and re-lowering treatments were left to provider discretion. And they had a different threshold for initiating re-lowering treatment than the definition of overcorrection in the primary endpoint. So, re-lowering treatment was initiated if sodium increased by 10 or more within the first 24 hours, but overcorrection was defined as an increase by 12 or more in the primary endpoint. So, this can explain why so many more patients received re-lowering treatment compared to the patients that were actually documented as experiencing overcorrection. Ultimately, the study was appropriate to evaluate the safety and efficacy of the interventions, and it was clinically relevant because it addressed important evidence gaps. However, they did have to use overcorrection as a surrogate endpoint, since the incidence of osmotic demyelination syndrome is so rare, and it's hard to say if overcorrection has a strong clinical correlation with that severe adverse event. The authors ultimately concluded that there's no difference in the overall safety or efficacy of 3% sodium chloride given as a rapid intermittent bolus or slow continuous infusion for the treatment of symptomatic hyponatremia. But they did mention that utilizing rapid intermittent bolus therapy may limit the need for re-lowering treatment. So, looking to apply this to our clinical practice, it may be reasonable to utilize 3% saline bolus treatment over continuous infusions in patients with symptomatic hyponatremia, not only to potentially limit the need for re-lowering treatment, but also because boluses are easier to dose and they're consistent with guideline recommendations. However, the results of this study are most applicable to patients with moderate hyponatremia symptoms, secondary to thiazide use or SIADH. So, caution will need to be used before applying these results to other populations. In the future, we need additional adequately powered studies to confirm the results of the SASA trial, and we also need studies in sicker patients with life-threatening hyponatremia since these patients had more moderate symptoms and were primarily in the emergency department or on a hospital ward. We also need studies in excluded populations like traumatic brain injury or cirrhosis. So, after seeing this information, would you recommend changing your institution's treatment protocol based on the results of the SASA trial? That's pretty much what I expected. I agree, we do need more data, but hopefully this study will be hypothesis generating for a much larger adequately powered study. I'd like to thank Kelly Rumball for her help with this presentation. And at this time, I'll take any questions. So for questions and answers, again, remember you can put those in the questions box. First, Kelsey, let's just start with kind of what are your thoughts on why in general the slow continuous infusion is still primarily preferred in practice? I think in general it's something that we're more comfortable with. And there's just not a lot of data for either treatment strategy, but we do tend to have more concrete data on the overcorrection rate for slow continuous infusion. So I think that just having those more concrete numbers makes people more comfortable going that route versus giving patients boluses that aren't necessarily based off of calculations. Awesome, thank you. Next question looks like, would the duration of hyponatremia change your treatment strategy? So this study actually did include patients with both chronic and acute hyponatremia. It was about 40% acute and about 60% chronic. So if you're looking at this study, you can kind of apply these results to either group. Okay, thank you. What role do you think 23.4% saline has in the context of this study? I think it's hard to say. Although you can get similar sodium amounts into patients with different calculations with either 23.4% or 23.4%, I think it's hard to say. Amounts into patients with different calculations with either 23.4% or 3%. These patients were a lot less sick than patients that would typically be getting the 23.4%. Most administration of the 3% was peripheral. These patients didn't have central lines. So I think I would use caution in applying this study to that higher concentration of hypertonic saline. Thank you. So that will conclude our Q&A session for this presenter. Again, thank you, Kelsey Dobbins, for that presentation and review. Now I'd like to introduce you to our second presenter, Teresa Benitez. Hi, thank you very much for the introduction. Good afternoon, everyone. I'm going to start off with a poll that reads, oops, sorry, at your institution, do you have a protocol for the administration of tranexamic acid or TXA in patients who present with trauma? Yes, I do. So I'm going to start off with a poll that reads, at your institution, Okay. Wow. That's great. At our institution, we actually do not have a formalized policy in place for using PXA in trauma, so it's interesting to see that there is a larger majority that does. Okay. Moving on. So, quick background. So, one of the most devastating features of TBI, or traumatic brain injury, is intracranial hemorrhage, which is associated with trauma-induced coagulopathy. This contributes to worse outcomes due to the depletion of clotting factors, hyperfibrinolysis, and overall uncontrolled bleeding. PXA and antifibrinolytic has been shown to decrease mortality in trauma patients with severe hemorrhage, and this was shown in the CRASH-3 trial, the largest trial of PXA in trauma. PXA has also shown some benefits in TBI in the CRASH-3 trial, which I'll get into shortly. However, the use of prehospital PXA for TBI has remained unclear. Here are two recent studies that involve PXA in TBI. The first row is the CRASH-3 that I mentioned briefly that looked into the effects of PXA on death, disability, and adverse drug events in TBI. Their primary outcome was head injury-related death after receiving PXA within 3 hours, and what they found was that PXA did not significantly reduce head injury-related death compared to placebo in the overall population. There was a statistically significant reduction in head injury-related death in the TXA arm for patients with mild to moderate TBI, however. And the second study is by Rowell and colleagues published last year that looked to determine if TXA-initiated prehospital within 2 hours improved neurologic outcomes in moderate to severe TBI, and what they found is that TXA-administered prehospital within 2 hours of TBI did not significantly improve 6-month neurologic outcomes, and they also found that there was no difference in 28-day mortality. So there's very scarce guidelines mentioning the use of TXA in TBI, let alone regarding prehospital use. In the 2016 Brain Trauma Guidelines for the Management of Severe TBI, there's no mention of TXA whatsoever. The other 2 guidelines are expert opinion is directed towards trauma and not TBI, but I figured it was worth mentioning. So, to the right was a guidance document regarding prehospital use of TXA for injured patients. They recommended that prehospital TXA should be administered, but only in patients with non-compressible bleeding. And on the bottom is the European Guideline on the Management of Major Bleeding After Trauma, which recommends TXA be administered as soon as possible if the patient is bleeding or at risk for significant hemorrhage. So, with that introduction, I'm going to present my article on the use of prehospital TXA in outcomes of severe TBI. So, hypothesis of the study is that prophylaxis prehospital administration of TXA in severe TBI may reduce the burden of trauma-associated coagulopathy and thus improve outcomes. For methods, they retrospectively gather data from a study called BrainProtect, and the BrainProtect study was a prospective observational study to investigate the relationship and outcome between prehospital treatment for patients with suspected severe TBI in the Netherlands. So, between February 2012 to December 2019, they collected patients that were treated via physician-staffed helicopter emergency medical services within nine participating trauma centers, and they followed these patients for one year. So, I have another audience poll, and I found it interesting that physicians were staffed in the helicopters for this trial. So, the poll reads, does your institution have physician-staffed ambulance or helicopter emergency medical services? Okay, so this is exactly what I thought. I'm not sure of the practicality or reality of having physician-staffed during medical transportation. So, same thing in our institution, we do not have this as well. Okay, so inclusion with suspected TBI with transportation to one of the nine participating trauma centers, and exclusion was, understandably, patients undergoing trauma arrest. Primary outcome was 30-day mortality. Secondary outcomes, one-year mortality. Secondary outcome, which was defined by the Glasgow Outcome Scale, which is a scale that is out of five. One is death, and five is good recovery. And then, the other secondary outcome is hospital length of stay. Moving on to statistical analyses, they used several t-tests to note the differences between those who received TXA versus those who did not. There were three planned subgroup analyses that consisted of the full cohort, confirmed TBI, and then isolated TBI, and they differentiated between confirmed TBI and isolated TBI based on the abbreviated injury scale. So, confirmed and isolated TBI both had a head-abbreviated injury scale of 3 or greater, but for isolated TBI, the abbreviated injury scale for other areas of the body had to be 2 or less. Because there were three subgroups to prevent Type 1 error of rejecting a true null hypothesis, they used the Bonferroni correction by dividing the standard alpha of 0.05 by 3 to give a significant threshold of 0.017. For missing data, they performed multiple imputations to prevent bias and loss of power. Results were divided into three analyses, unadjusted analysis with unadjusted logistic regression, confounder-adjusted analysis using the original data, and then confounder-adjusted analysis after multiple imputations due to missing data. They also did a multivariable model to account for potential confounders, such as patient demographics, pre-injury, medical condition, medications in which they only mentioned anticoagulants, severity of injury using the injury severity scale, the Glasgow Coma scale, operational characteristics, and then initial vitals at arrival to the helicopter. So, for results starting with demographics, a total 1,827 patients were analyzed in which most did not receive TXA. For demographics, the following was squared, which was deemed statistically significant in the trial. So, by statistically significant means, patients who received TXA were older, had a higher injury severity score, had a lower Glasgow Coma score, and had a higher heart rate. What this means in terms of clinical significance, however, may be minimal, as these values are not that substantially different from each other. Other values I include in the chart is the oxygen saturation and systolic blood pressure, as these two values of less than 90 have been shown to increase mortality in CVI. Moving on to results for the primary endpoints, for patients who received TXA versus those who did not, death at 30 days was higher in the TXA arm at 37% than in the non-TXA arm at 30%. This had a p-value of .005, and there was about 6.2% of missing data. For the unadjusted analysis using unadjusted logistics regression, all three cohorts had a statistically significant increased risk of 30-day mortality after receiving pre-hospital TXA. After adjusting for confounders, but still using the original data set, statistical significance for increased 30-day mortality was lost in the full and confirmed TBI cohort, but still remained only in the isolated TBI cohorts. And the same trend occurred for the adjusted analysis after multiple invitations were done for missing data. So, only isolated TBI cohorts showed statistical significance for increased 30-day mortality. For sensitivity analysis, similar from the previous slide, 30-day mortality was increased only in the isolated TBI cohort after using confounder-adjusted survival analysis with original data set and after multiple invitations were put in for missing data. For sensitivity analysis, similar from the previous slide, 30-day mortality was increased only in the isolated TBI cohort after using confounder-adjusted survival analysis with original data set and after multiple invitations were put in for missing data. So, overall, for the primary endpoints, the unadjusted analysis showed statistically significant higher 30-day mortality in all three cohorts, but when you adjusted this data for confounders and also imputations for missing data, you only saw statistically significant increased risk of mortality in just isolated TBI. Authors also conducted a post-hoc analysis on whether there was an association of TXA with mortality and the use of anticoagulant medications prior to injury, and they found no evidence of an interaction. So, this is just a visual representation of the survival analysis where we see no difference in the full cohort or patients with confirmed TBI, but looking at graph C in the isolated TBI, you can start to see the diversion of survival in the first initial days. Okay, for secondary endpoint involving one-year mortality, similar to the primary endpoint results, all cohorts show significant increase, but when you adjust it for confounders using original data and then also doing multiple implications from missing data, we only see this applied for isolated TBI only. For neurologic outcome measured by the Glasgow Outcome Scale at discharge, unadjusted analyses showed worsening neurologic outcome for all three cohorts, but after adjusting for confounders and missing data, there was no difference. For the third secondary endpoint, isolated TBI had an increased hospital length of stay after adjusting for confounders with original data only. So in the author's conclusion, pre-hospital TXA administration was associated with increased mortality in patients with isolated severe TBI and should be avoided in such patients. They also did note that the data did not suggest abandoning the current practice of using TXA in patients with extracranial injuries and substantial blood loss. Moving on to critique, for strengths, I appreciated that the authors performed many statistical tests to limit confounder influence, as there are a number of things that can impact mortality and TBI. Another strength is that they involved nine trauma centers, so there was some variability in physician practice. For weaknesses, I have a few. One big one was that it was not randomized. Another was that it was limited to one geographical location in the Netherlands, which is a rather small but well-developed country with relatively short distances to trauma centers. Next, the results required multiple imputations due to the amount of missing data, which could have influenced true values. Next, despite their aggressive attempts at statistical tests to limit confounders, there were still probable residual confounders, such as treatment variability, neurosurgical intervention, et cetera. Another thing was that authors did not assess adverse drug reactions related to TXA, such as VTE events, and also seizures, which severity of TBI can cause in and of itself as well. They did not measure pertinent labs to assess efficacy, such as hemoglobin or fibrinogen level. Another is that they didn't mention other interventions that could have impacted mortality, such as neurosurgical interventions, steroid use, and TBI. And lastly, they relied on the GLAD-Calcoma score to define severity and no other physical assessments, such as pupil reactivity. So for takeaway points, this was a hypothesis-generating study. The unadjusted analyses results contained confounding variables that could have influenced the data, so I would take these results with a grain of salt. Increased mortality in isolated TBI after TXA administration was shown only after extensive manipulation to the data due to the considerable amount of missing data. Also, it's important to know that despite aggressive resuscitative efforts, severe TBI still carries a high mortality rate. In this trial, the median GLAD-Calcoma score ranged from 3, which is the lowest possible score, to 6, with a median of 4. Lastly, pre-hospital TXA is not widely adopted in many trauma centers. To implement, patient factors strongly need to be considered when formulating a protocol, such as time of arrival to trauma center, as delayed time management to treat TBI increases mortality, feasibility of administration, and most importantly, which patient would likely benefit. As we saw in this trial and also previous trials, TXA in severe TBI has not shown a benefit. So, in general, TXA should be considered in those on concomitant head injury and hemorrhagic shock. Trials such as the CRASH-3, for example, concluded possible benefits in TBI with minimal adverse effects. Thank you all very much for listening in, and I will now take some questions. Thank you, Teresa. So, the first question we have is, can you comment on the administration of TXA? Was it given by bolus or infusion, and was it compounded by the first responders, or was it premixed? Do we know any of that? So, they used a 1-gram bolus, but there were some patients in this trial that did have 2 grams. So, that was also one of the limitations to the trial, was that they didn't assess the efficacy of 1-gram versus 2-grams, so it actually varied. And there was no mention of who actually prepared the TXA as well, so that was one of the points I made at the end, one of the take-home points of feasibility of administration, who's actually going to prepare it. Excellent. Excellent. Did the study investigators report or split out the analyses by blunt versus penetrating TBIs by chance? They did not. No. They did have a demographics chart that showed the cause of injury, and the most was blunt trauma, just from what I looked at, such as motor vehicle injury, pedestrian injury, fall from height. But other than that, they didn't actually do a separate analysis of blunt versus penetrating trauma. Do you think this potentially had any impact on what their study found? I think so. Penetrating trauma is pretty devastating. It carries a pretty high mortality rate. I believe in the trial, about only 3% had penetrating trauma, a stab wound or whatnot, or firearm. So, I think this could have definitely impacted some results for sure. And then, I think this will be our last question. If anyone has any others, let me know. One of the other questions was, you mentioned that in the Netherlands, for example, there's likely shorter distances to trauma centers. Did the authors report the time that was anticipated to have been from the time of injury to receiving the TXA? Not that I saw, but I did read in the supplemental that the average distance from the site of injury to a trauma center was an average of 18 miles. Oh, wow. Okay. So, that will conclude our Q&A session. Thank you again, Teresa Benitez, for that presentation. Now, I'd like to introduce our final presenter, Joseph Morata. Hi, everyone. My name is Joseph Morata. I'm the Director of the National Center for Traumatic Stress Disorder at the National Center for Traumatic Stress Disorder. And I'm here to talk to you about trauma-related trauma-related trauma-related trauma-related So looking at our intention to treat population, which was everybody who was included in the trial, there was no significant differences in any baseline characteristics between this group. If we look at our primary outcome of in-hospital mortality, we do see a favorable benefit with beta blockers with a mortality rate of 8.1% versus 16.7% in our control group. However, this was not statistically significant. With regard to that primary outcome, the authors additionally state that after adjusting for all clinical variables between the groups, there was no difference for this in-hospital mortality. They presented an adjusted incidence rate ratio of 0.06 favoring beta blockers, but again, not statistically significant. When we look at both of the functional outcomes, so the GOSE score greater than or equal to five at both discharge and at six months, again, we see a favorable benefit with beta blockers. However, neither of these are statistically significant either. And additionally, these secondary outcomes were also not statistically significant after adjustment for all clinical variables as well. In a pre-specified subgroup analysis, only those patients with isolated severe TBI were included for further analysis, and this included a total of 154 patients. And so similar to Therese's trial, in order to qualify as isolated severe TBI, these patients had to meet all the same inclusion criteria as the intention to treat group, but here patients also had to have all extracranial injuries with a score of less than or equal to two. Again, there were no significant differences in baseline characteristics between the groups in this subgroup. And we see with the in-hospital mortality in the primary outcome, there is a statistically significant benefit with beta blockers with 4.4% mortality rate versus 18.6% mortality rate in the control group. Looking at the secondary outcomes, there is no statistical difference with the GOSE score at discharge. However, at six months, we do see a favorable benefit with beta blockers, 92.3% versus 79.1% in the control group. So the authors concluded that the use of early propranolol in patients suffering isolated severe TBI led to improved survival as well as better functional outcomes up to six months following injury. And the authors further conclude that this trial provides support for the routine administration of beta blocker therapy as part of a standardized neurointensive care protocol. As we critique this trial, the things I want to highlight from a strength standpoint are that it addressed an important clinical question that is relevant to clinical practice. Prior to the study, there is increasing evidence that beta blockers and propranolol specifically may improve patient outcomes in our TBI patient population. Additionally, it didn't have a pragmatic design that was realistic and sensible to actual clinical practice. They looked at relevant clinical outcomes, not only in-hospital mortality, but possibly even more importantly, long-term functional outcomes, and in this case, six-month functional outcomes. And finally, relatively few patients were lost to adverse effects and or follow-up. Specifically, three patients in the beta blocker group were removed secondary to bradycardia, and three patients were removed or lost to follow-up. On the flip side, from a weakness standpoint, it was single center, so this is going to limit the external validity in and of itself, but it was also performed in IRAN only, so this is going to further limit that external validity. Furthermore, it was a relatively small trial, limited number of patients, so it's going to further limit our external validity. It was not blinded, and there was no placebo. Initially, the ethics committee did consent to the use of blinding and placebo for this trial. However, due to insufficient funding, the decision was made to continue the study without blinding or the use of a placebo. Just wanted to make that known. Additionally, the authors made some continuous variables dichotomous, which is not typically how we analyze those variables, so that needs to be taken into account as we analyze these results. The definition of severe intracranial injury in this trial, again, was that AIS score of 3 to 5, and most major TBI studies define a severe injury with a GCS score rather than an AIS score, and specifically a GCS score of 3 to 8, so this really isn't in line with how we typically define severe injury. Additionally, we may have been looking at a potentially less severe TBI patient population. Both the mortality and the functional outcomes seemed too good to be true in this trial. Even in our isolated severe TBI subgroup, in the control group specifically, the mortality rate was 18.6%. In the trial that was just presented before this, for example, the mortality rate was 30% in the control group. Additionally, the results were only statistically significant after subgroup analysis in that isolated severe TBI patient population, and finally, other treatment modalities for TBI were not documented, and this can certainly have an effect on our outcomes. Looking at our number needed to treat for both in-hospital mortality as well as our functional outcome at 6 months, we see that in the primary group as a whole, we need 12 patients in the number needed to treat, whereas in our isolated severe TBI patient population, that number is 8 for both in-hospital mortality as well as the GOSE score greater than or equal to 5 at 6 months. And unfortunately, no number needed to harm with regard to hypotension or bradycardia could be calculated since these patients were removed from analysis, which is yet another weakness of the trial. So take-home points, there is increasing evidence that propranolol may improve patient outcomes in our TBI patient population. Unfortunately, the limited patient population, some methodological concerns, and statistical significance only after subgroup analysis in this trial limits the applicability of the findings. If we are going to take to heart the results of this trial, propranolol 20 milligrams every 12 hours for 10 days can be considered, but it has to be for hemodynamically stable patients, which is part of the main inclusion criteria for this trial, and again, only in our isolated severe TBI patients where we saw a statistically significant difference. However, I would not make this the standard of care or any standardized protocol for all TBI patients. In order for that to be the case, further larger studies and or prospective multicenter RCTs are going to be needed before they can be used in all TBI patients. Furthermore, additional considerations need to be addressed, including what's the optimal agent, what's the optimal dose, how long do we treat these patients, and specifically which patient population is going to show the most benefit. Is it going to be our more severe TBI patients or possibly our less severe TBI patients? All of these questions remain unanswered at this point. So that concludes my analysis of this trial. I'd like to conclude by two polling questions to gauge current practice trends out there. First question is, does your institution currently include beta blockers as part of standard protocol for TBI patients? Yeah. OK, so 70% answered no. And that's kind of in line with what I would expect. It's not anywhere in the guidelines. And up until this point, this was the only prospective RCT. So that is not surprising to me. And regardless of what your answer was to the first question, which we saw was majority no, the second question, what percentage of your TBI patients currently receive beta blockers as part of their initial care? 0% to 25%, 26% to 50%, 51% to 75%, or 76% to 100%. OK, so the majority are answering 0% to 25%, and then followed by, well, besides not sure, 26% to 50%. So not a lot of hospitals seem to be using beta blockers as part of the initial care, which makes sense to me. All right, that is the end of my presentation. I'll be happy to answer any questions at this time. Thank you, Joseph. So the first question we have is, what are your thoughts on the dose of propranolol that was used in this study compared to other doses that have been utilized? So the propranolol 20 milligrams, Q12, for a duration of 10 days seems to be in line with at least trials that have shown benefit with propranolol. I will be honest, I can't remember specifically the dosage of some of the previous trials, but I do remember it being around 20 milligrams. I believe 10 milligrams was used in some of the trials as well, but I apologize. Sure. Another question we have is, so similar to some of the other studies, of course, patients were excluded who were already on beta blocker therapy. So of course, in real life, we're going to get those patients admitted to us that would meet criteria otherwise. How do you manage those patients that are either already requiring beta blocker therapy either prior to admission or during their admission? How do you manage those ones? Yeah, so we don't have any data to really guide that decision. So I don't think it'd be out of the realm to change. It's also going to be dependent on what agent they were on prior to admission. Not a lot of patients, in my experience, are on propranolol as an outpatient, at least not chronically. So those that are on, let's say, a more common metoprolol, carbadolol, et cetera, we can consider switching to an agent that has shown benefits, such as propranolol. I don't think that's out of the question. But again, in this specific study, those patients were excluded in this trial. But if we are going to practice what's been studied in the trials, we would want to use the agents that have shown the most benefit. Thank you. Another question, how would you define early in terms of initiation? I would say, in this trial, of course, they looked within 24 hours after admission is when they started or they enrolled these patients. I think the earlier, the better. When we see that hyperadrenergic storm, kind of attenuating that process as soon as possible, in theory, would have the most benefit if we do that sooner rather than later. So with regards to this trial here, again, they enrolled patients at 24 hours after admission. But I think you could see other trials enrolling patients even before that cutoff point. I think maybe our last question for today, unless any stragglers come in, so you mentioned that they reported out, they kind of summarized the patients in terms of AIS instead of GCS. Is there kind of a way, like right there, we're talking about apples and oranges, essentially. Is there a way to kind of give us an idea in terms of GCS, what patient population we were looking at in this study? So that was one of the variables that they dichotomized. So they had a cutoff score of less than or equal to 8. And as a whole, about 37% of patients were with a GCS less than or equal to 8. So again, probably a less severe patient population was analyzed in this trial. Excellent. Thank you so much. This does conclude our Q&A session. Thank you again, Joseph Morata, for presenting today. And so thank you overall to all of our presenters today and for the audience for attending. Please join us the third Friday of the month. So our next session will be March 19th from 2 to 3 PM Eastern Standard Time for our next Journal Club Spotlight on Pharmacy. And this concludes our presentation today. Have a great day.

journal club spotlight

SALSA trial

hyponatremia

TXA

traumatic brain injury

neurologic outcomes

mortality

functional outcomes

beta blockers