SCCM Resource Library-September Journal Club: Spotlight on Pharmacy (2022)

Video Transcription

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 Jill Starkiewicz, and I'm a clinical pharmacy specialist in the Medical and Cardiac Intensive Care Unit at Advocate Lutheran General Hospital in Park Ridge, Illinois.  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 in 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 sccmcppjc and PharmICU.  Please note the disclaimer stating that the content to follow is for educational purposes only. 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 Dimitar Zelenkov, who is a PGY-2 critical care pharmacy resident  at Tufts Medical Center in Boston, Massachusetts. Our second presenter is Savannah Gross.  She's a PGY-2 critical care pharmacy resident at UCHealth Memorial Hospital Central in Colorado Springs, Colorado. And our third presenter today is Brittany Curran.  She's a PGY-2 critical care pharmacy resident at the University of Cincinnati Medical Center in Cincinnati, Ohio. And now I'll turn things over to our first presenter.  Good afternoon, everyone, and thank you for the introduction. The article that I'll be presenting today is titled, The Association of Vancomycin Plus  Piperazole and Tazobactam with Early Changes in Creatinine vs. Cystatin C in Critically Ill Adults.  One of the most common regimens chosen for empiric therapy for critically ill patients with sepsis is vancomycin plus piperazole and tazobactam.  This combination provides the necessary coverage of MRSA and pseudomonas to influence overall survival, but its use has been associated with concern for acute kidney injury.  In a recently published meta-analysis of nearly 50 observational studies in 60,000 patients,  vancomycin plus piperazole and tazobactam increases AKI risk twofold compared to vancomycin monotherapy.  All of the literature supporting the AKI in humans, though, is based on creatinine-defined AKI. As a reminder, serum creatinine, which is your standard kidney function biomarker, is  subject to proximal tubular secretion. In light of this evidence, health institutions have implemented large-scale initiatives to avoid the combination.  Now, this could be problematic because the avoidance of the combination could limit treatment  of life-threatening infections, expose patients to toxicity from alternative antibiotics, and worsen antimicrobial resistance patterns.  The mechanism of this potential interaction is unknown. We are aware that vancomycin can cause acute tubular necrosis through oxidative stress  and formation of obstructive tubular casts. In contrast, we have little evidence linking piperazole and tazobactam to nephrotoxicity  other than very few cases of acute interstitial nephritis. As you can see in the slide, piperazole and tazobactam inhibits OAT1 and OAT3, which are  the organic anion transporters 1 and 3, leading to a reduced secretion of creatinine into the urine. Thus, the drug causes an increase in serum creatinine.  This phenomenon of pseudonephrotoxicity appears to be a common property of drugs that inhibit this system, such as probenazid and trimethoprim.  Vancomycin suppresses OAT1 and OAT3 expression, so in the end, you have this competitive inhibition  of creatinine secretion and vancomycin-mediated transporter suppression as a plausible mechanism of an increase in creatinine concentration.  Kidney function is generally closely related with clinical outcomes, such as mortality and length of hospital stay.  A true nephrotoxin should, therefore, be expected to affect patient-centered outcomes, such as hemodialysis, hospital length of stay, and mortality.  A pseudonephrotoxin, on the other hand, will not.  So, in this recent meta-analysis of studies comparing vancomycin plus meropenem versus vancomycin plus piperacillin-tazobactam, of about 15,000 patients, you can see, again,  a robust increase in AKI in patients treated with vancomycin plus piperacillin-tazobactam. So, let's examine the patient-centered outcomes, such as mortality and hemodialysis.  So, as you can see here on these four spots, there was no increase in either.  So, this shows that elevated creatinine levels are dissociated from patient-centered outcomes.  So, this is the article that we'll be discussing today, which was recently published in the  Journal of Intensive Care Medicine, and this is the first polling question for the public. So, do you have cystatin C monitoring available at your institution? Is it in-house?  If yes, is it in-house, or is it a send-out lab, or no?  All right. Thank you for responding. So, it looks like half of the participants do not currently have the statin C monitoring available at their institutions. About a third have  it, but it's a send-out lab, and about 20 percent monitor in-house. So, thank you.  So, the authors of the study hypothesized that vancomycin plus piperacillin-tazobactam is associated with changes in creatinine, but not with changes in statin C, need for  dialysis, or mortality. Therefore, these changes in creatinine do not reflect changes in kidney  function or underlying tissue injury. You may ask yourself, though, why did they choose statin C? Well, statin C is an alternative to serum creatinine. It's a validated kidney  function biomarker that does not undergo tubular secretion. It's a small protein that's actually produced by all cells in the body that have a nucleus, and it's easily filtered  and broken down at a constant rate. So, statin C also has a quicker half-life than creatinine and has been shown to identify AKI earlier than creatinine in patients with sepsis.  So, this trial was part of the MESI study, which is an ongoing prospective observational study that enrolls patients admitted to the ICU with severe sepsis or septic shock. Patients  were included in the antibiotic cohort if they were treated with vancomycin plus piperacillin-tazobactam  or vancomycin plus sepapine for 48 or more hours, with each drug initiated within 48 hours of ICU admission. The antibiotic cohort had 739 patients, and sepapine was chosen  as a comparator because it's commonly used for empiric sepsis treatment and then combines when vancomycin has been associated with a lower AKI risk compared to vancomycin and  piperacillin-tazobactam. Patients were then further included in the statin C subcohort if their timing of their statin C blood draws aligned with a priori defined windows for  baseline and follow-up statin C levels, and there was a total of 192 patients in this subcohort. Patients were excluded from the study and the analysis if they had end-stage  renal disease, dialysis within 14 days before the index date, which is when both antibiotics  were initiated or a baseline AKI. Primary outcomes were creatinine and statin C concentrations  measured at index in day 2 for 48 to 72 hours. The statin C was not able to be obtained in all patients, though, so B1 was examined as a secondary biomarker of GFR, and it's also  not subject to tubular secretion. Secondary outcomes included creatinine-defined AKI through day 14, dialysis initiated within 7 days of AKI events, and 30-day mortality.  In terms of statistics, baseline covariate balance was examined with standardized mean differences and an absolute SNP values of greater than 0.1 were indicative of a balance.  Investigators also adjusted for condounding variables using propensity scores and inverse  probability of treatment weight, and they also examined the change in biomarker concentrations at day 2 as dichotomous variables, so they compared the incidence of 50% or greater  increases in biomarker concentrations baseline to day 2 with regression models.  I mentioned that only 192 patients had statin C levels drawn out of the 739 patients, but both groups, so those who had statin C and those who did not, had similar baseline kidney  function, severity of illness scores, and need for mechanical ventilation. Shifting gears to  baseline characteristics, so what did the average patient look like? Well, they were admitted to the  emergency department with sepsis, and they were transferred directly to the ICU on high-dose  pressors, fluid resuscitation, and it was a general MIECHV population, so there was a high  prevalence of patients with cancer and cirrhosis. So, shown in the table in red are variables with  standardized mean differences greater than 0.1, which is a reminder of points to imbalances. So, it's important to note that before weighing, patients on vancomycin plus  piperacillin-tazobactam had higher severity of illness scores, lower baseline GFR, higher lactate concentrations, and more frequent diabetes, cirrhosis, and metastatic solid cancer.  As you can see from this slide, patients were also on a variety of different medications during this study period. Note how most of these can actually be nephrotoxic as well.  So, vancomycin plus piperacillin-tazobactam was associated with a significantly higher average  creatinine concentration and a higher frequency of creatinine increases of 50% or greater.  In contrast, vancomycin plus piperacillin-tazobactam was associated with non-significantly  lower average C, cystatin-C concentrations, and a similar frequency of cystatin-C increases of 50% or greater compared to vancomycin and cefepime. BU-1 measures did not differ  significantly between the two groups at day 2, nor did the rate of increases through day 14.  And also, both cystatin-C and BU-1 to creatinine ratios were significantly lower in the vancomycin plus piperacillin-tazobactam group versus the vancomycin and cefepime group,  which means that the creatinine increased to a significantly greater extent than either cystatin or BU-1 by day 2. We also saw that vancomycin plus piperacillin-tazobactam was  associated with a significantly higher rate of creatinine-defined AKI in day 14. The association  was attenuated in non-significance, though, when restricted to stage 2 or higher events.  Similarly, vancomycin plus piperacillin-tazobactam was not associated with an increased rate of dialysis or mortality after weighing confounders. And as you can see here,  the rate ratio for mortality in the blue box is after weighing for confounders.  Also to mention, results were minimally changed by propensity score trimming, and they were also similar after excluding corticosteroid-exposed patients.  And as you can see here on the Kaplan-Meier analysis, vancomycin plus piperacillin-tazobactam  was associated with an excess of creatinine-defined AKI events, but it was primarily during the first  two to three days of treatment, and this is in line with the significantly elevated average creatinine concentrations observed on day 2. So authors concluded that vancomycin  and piperacillin-tazobactam was associated with an increased risk of creatinine-defined AKI, but not with changes in alternative kidney function biomarkers or clinical outcomes  downstream from true AKI. So this supported the hypothesis that creatinine-defined AKI with combination may actually represent pseudotoxicity.  So what makes this study different and relevant? Well, it was the first study of vancomycin plus  piperacillin-tazobactam to employ cystatin C as an alternative to serum creatinine. We discussed  that cystatin C is a validated kidney biomarker that is not affected by creatinine secretion.  So I believe that utilizing cystatin C as a primary outcome in this study is an appropriate way to determine that vancomycin plus piperacillin-tazobactam-associated creatinine  changes do not actually reflect changes in kidney function or underlying tissue injury.  I also think that looking at patient-centered outcomes is what is truly important at the end  of the day. I'm not as concerned if my patient develops a bump in their creatinine, but I'm more  concerned if they end up on dialysis after a few days of treatment. Authors also noted that most  AKIs were stage 1 and that the association was attenuated when they only lived at stage 2 AKI or  higher. We saw that only 25% of patients in the antibiotic cohort had a cystatin C drawn,  but there was no difference in the characteristics of the patients that had a cystatin C drawn and  those that did not. The authors also looked at changes in B1 to supplement their analysis,  but the use of B1 to assess kidney function has limitations of itself.  Investigators also performed extensive confounding adjustments, and I think the study design is particularly vulnerable to confounding. The authors tried to evaluate  AKI due to drug toxicity in critically ill patients, and we know that these patients have  other factors that can be causative of AKI, but I believe they used appropriate statistical methods  to test the average treatment effects with propensity score matching and inverse probability  of treatment weighing. These methods have become increasingly popular to address confounding in  real-world evidence studies, especially when comparing two possibly different populations. We also saw that their results were minimally changed by propensity score trimming, which  showed that confounding was minimal. And for missing data, patient data authors mitigated data bias with multiple amputation, allowing us to control for covariates such as albumin,  laxate, and bilirubin. In terms of limitations, the cystatin C could be a very useful lab to assess  kidney function, but as we saw from the poll, it may not be available at other institutions as  it's relatively new. It's a send-out lab here at Tufts Medical Center and takes about 48 hours  for the results to come back, and the test also costs around $40. Also, cystatin C concentrations  can be affected by several factors such as BMI, corticosteroid, thyroid activity, cancer,  diabetes, and organ transplants. Many of these are common comorbidities and are critically ill  patients with sepsis, so we must be cautious interpreting these results. The population that  was studied, which was critically ill patients with sepsis, mostly presenting from the ED or the  ward, was also appropriate for the hypothesis. You know, this is the population that we give empiric antibiotics to, but baseline characteristics show that the vancomycin plus  piperacil and tazobactin group was thicker with higher rates of comorbidities and higher Apache  scores. Additionally, we know that patients with sepsis receive aggressive fluid resuscitation,  which could have had some dilutional effects. However, dilution is expected to impact creatinine,  cystatin C, and B lens similarly. Unfortunately, there was no urine output data that was collected,  but I think this is often difficult to accurately collect unless the patient has a catheter.  We also saw that a good amount of these patients received medications that could be nephrotoxic,  which could have impacted our results. And lastly, the study only included critically ill patients from a single center, which may limit generalizability. However, we did see  replication of the association between vancomycin and piperacil and tazobactin and the creatinine rise, which suggests that these findings are not unique to the population.  So, despite the limitations, the study stands out to me as a well-designed and unique experiment  to test whether vancomycin plus piperacil and tazobactin combination is nephrotoxic or merely  only responsible for increase in creatinine. Based on this data, we can accept the fact that  our patients will have a rise in their creatinine, but won't necessarily develop nephrotoxicity down  the line. And this is similar to other medications such as Bactrim. Although it is important to be  aware of it, the rise in the creatinine shouldn't be a reason to avoid the combination. I also  wouldn't send cystatin C levels in my patients that have a rise in their creatinine, as a single  level won't tell me much. Patients will need to have baseline cystatin C levels so we could  trend the fluctuation of these levels, which is not always entirely feasible. I believe we should  pay attention to other biomarkers such as BUN in urine output and utilize therapeutic drug monitoring for medications such as vancomycins and cephalosporins if able to. But overall,  I believe the study will bring some clarity to the vancomycin and piperacil and tazobactin controversy that has been investigated for the last decade. I think we should anticipate  large and randomized clinical trials for a definitive answer on the question, so stay tuned for the ACORN trial that's estimated to be completed in 2024.  Here at Tufts, cefepime is our preferred anti-pseudomono agent due to higher susceptibility  compared to piperacil and tazobactin, but I do believe this data impacts my clinical practice. I will feel more comfortable using this combination in certain situations,  for example, in my critically ill patients with an intra-abdominal infection, and I've also had attending providers and consulting services be reluctant to use  vancomycin plus piperacil and tazobactin due to the nephrotoxicity risk in the past where  that actually would have been appropriate based on cultures. So based on the results of the study, I will feel more comfortable providing education to my teams.  And I have one last question for the audience. What is the preferred anti-pseudomono agent at your institution, cefepime or piperacil and tazobactin or another agent?  So it looks like 70% of the participants, their go-to is cefepime for the anti-pseudomono agents.  While 30% use piperacil and tazobactin. So that's interesting to know. Thank you.  So thank you for inviting me to share this article with you today. I will now take any questions you may have.  Thank you, Dr. Zelinkov. There are some questions coming in in the poll,  and I will go ahead and get started. The first question is, did this study indicate the dose of vasopressors that the patients were receiving?  That is an excellent question. I did take a look at the supplementary  pending for the study, and I do not believe that there was any indication of the doses of vasopressors that the patients are on. So that is a question that I think is appropriate  to be answered. Our next question is, did this study look at vancomycin levels as we know that  higher vancomycin troughs are associated with nephrotoxicity or AKI? That's a great question  as well. So all antibiotics were actually dosed by critical care pharmacists in this study,  and I do believe that they monitored vancomycin levels in this study as well. And I believe the  average concentrations, the average doses that they used for vancomycin was about 30 milligrams  per kilogram per day. And then our final question is, if serum creatinine does rise with vancomycin, plus peptazo, do you dose adjust at your institution, even though it may not be  an accurate reflection of their renal clearance?  So that's a good question. I don't believe that I have dose adjusted this combination before, although I have seen the rise in the serum creatinine. Like I mentioned before, going  forward, I will be paying attention to, you know, other, you know, biomarkers such as B1 in urine output. But I don't believe that we currently empirically dose adjust  starting patients on this combination when we see a rise in creatinine.  Okay, that concludes our Q&A session. Thank you very much, Dr. Zelenkov. 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.  If you all wouldn't mind just briefly responding, how many attendees are you viewing this webinar with? Okay, now we will move on and I'd like to introduce our  second presenter, Dr. Savannah Gross. Thank you for the introduction. Today I'll be presenting  the journal article, Andexanet Alpha versus Four-Factor Prothrombin Complex Concentrate for the Reversal of Apixaban or Rivaroxaban-Associated Intracranial Hemorrhage,  a Propensity Score Overlap Weighted Analysis.  Before we get started today, I did just want to poll everyone to see if at your institution we currently use Andexanet Alpha for emergent bleeding reversal.  I'd imagine every institution is a little bit different here, but maybe for some this is more routine and for others this may or may not be the case.  So it looks like we had about 76% of people saying they weren't using it quite as much,  about less than 10%. And that is pretty much what I was expecting just based on the practices that I've seen and I think it goes along with a lot of evidence that we've had for the  years leading up to this point. So thank you for your participation.  So, I wanted to start off just by providing some background information to discuss how we've gotten to the point of the study we'll be discussing today.  So, while we do have alternative options, AndexNet-ALSA is currently the only specific factor Xa selective reversal agent that is currently available.  It was originally FDA approved in 2018 based off of two phase three trials, those being the Anexa-A and the Anexa-R trials, which compared reversal of apixaban  and ribaroxaban by assessing the change in anti-TinA activity from baseline. So, these trials found statistically significant reductions in anti-TinA activity in both  of the groups who received Andexa compared to the groups that received the placebo. Since it was approved a couple years ago,  the American Stroke Association guidelines have updated the recommendations to prefer Andexa for reversal of factor TinA-associated intracranial hemorrhages.  And they also still make note that four-factor PCC is still recommended as an alternative if and when Andexa is unavailable.  What we have here is a trial of just a few studies that have come out since Andexa was initially approved, the first of which being the Connelly  and colleagues trial, which is more commonly known as the Anexa-IV trial. And this has become one of the most referenced Andexa trials over the last few years.  It was a single cohort study assessing acute major bleeding within 18 hours of a factor TinA inhibitor after being administered Andexabolism infusion.  For this trial, the dosing was based on the package insert guidance, with doses being based on the timing of the last dose of the anticoagulant.  This study noted 82% of patients achieved excellent or good hemostasis, as defined by a less than 35% increase in hematoma size,  but found no significant relationship between hemostatic efficacy and an overall reduction of anti-TinA activity during treatment.  Of the 34 patients who had a thrombotic event, 11 of them had occurred by day five,  while 11 patients had a thrombotic event from day six through 14,  and the remaining 12 patients had a thrombotic event following that with day 15 through 30. Thirty-five of the 49 patient deaths were related to cardiovascular causes,  but overall for this trial, the biggest limitation, which was noted in a study, is that there was no randomized comparison with a control group at this point in time.  The second trial that we have is the Amar and colleagues trial, who conducted a single center study to assess intracranial hemorrhage stability  for patients who received either Andexa or four-factor PCC, and found no difference in the CT scans at six or 24 hours following reversal,  and they also noted similar neurologic outcomes as assessed by a modified Rankin scale.  Lastly, in the bottom row, we have the Pham and colleagues study, which published a retrospective trial, and it actually came out just a month prior  to the one we'll be discussing today, and it similarly compares Andexa to four-factor PCC in patients with a factor TinA-associated intracranial hemorrhage.  They found no significant difference in hemostasis achieved, no difference in thromboembolic events, or inpatient mortality between the two agencies for reversal, and of note,  they only found one statistically significant difference, and that would be the cost of the reversal, as Andexa costs over $23,000, whereas four-factor PCC costs just under $6,700.  So this brings us to the study we'll be discussing today, which was a two-cohort indirect comparison study that assessed a subset of patients  who were initially enrolled in the Andexa IV trial a couple years ago and compared them to a synthetic control arm of patients  who received four-factor PCC during a similar timeframe from the Hartford Health Care System.  Similar to our previous studies, the inclusion criteria were patients 18 years or older  who had CT evidence of an intracranial hemorrhage,  or who had previously taken either a Pixaban or a Baroxaban less than 24 hours prior to their bleed.  Exclusion criteria matched that of the Andexa IV trial, which excluded patients with a GCS less than 7, a hematoma volume greater than 60 milliliters,  and patients with surgery scheduled within a 12-hour timeframe.  The study compared the Andexa bolus plus infusion compared to the four-factor PCC weight-based infusion, so that being 25 to 50 units per kilo infusion.  Coprimary outcomes were hemostatic effectiveness, and again, that was defined as a less than 35% increase in hematoma volume, and all-cause mortality at 30 days.  The secondary outcome was thrombotic events during the first five days after the reversal agent was administered.  I recognize this is a bit of a lengthy slide, but I did want to speak a little bit more to the statistical analysis of this study, as it really does impact the outcomes  that we're observing in this trial quite a bit.  So, for our baseline characteristics, an absolute standard difference was calculated to illustrate what was considered a clinically meaningful difference  between the two treatment groups, with a value of 0.1 correlating with a relevant difference, and I'll discuss this a bit more on our next slide whenever we talk  about baseline demographics.  A sensitivity analysis was conducted to assess hemostatic effectiveness as deemed by the clinical judgment of separate independent investigators,  and a subgroup analysis was also done to assess the change in hematoma volume for patients with an intracranial hemorrhage or an intraventricular hemorrhage.  A propensity score overlap weighted binomial generalized estimated equations to calculate odds ratios with a 95% confidence interval, and this weighted analysis was used  to determine the mean differences between the INDEXA and the four-factor PCC group.  So, for those of you who are less statistically savvy, such as myself trying to read this trial, this method ultimately assesses the differences  in our patient characteristics between the two treatment groups, and it's adjusting them through the overlap weighted analysis to account for confounding of the two groups.  Ultimately, this is done to mimic the attributes of a randomized clinical trial, and it is attempting to overcome the limitations of a non-randomized trial  by assessing the weights to each patient characteristics that are proportionally equivalent to the counterpart of the opposite treatment group.  Which brings us to our baseline characteristics. So, for those of you who noticed, in the article itself, there were actually three tables  of baseline characteristics listed, and I have the first one shown here, which represents the data prior to the weighted overlap analysis.  So, these are the true baseline characteristics for each group. As I previously mentioned, an absolute standard difference of 0.1  or greater was deemed a clinically relevant difference between characteristics in our treatment group.  And of note, all of the baseline characteristics that are listed in this table, aside from gender and baseline GCS, had a value greater than 0.1,  meaning virtually all of the baseline characteristics had statistically significant differences.  Overall, most of the patients were males in their 70s with a baseline GCS of 14. Presenting with an intrathebral or intraventricular bleed.  A couple other things I wanted to point out at the bottom of this chart is that the vast majority of patients who received Andexa got the lower dose,  so the 400 milligram bolus plus a 440 milligram infusion.  And likewise, about three-quarters of patients in the four-factor PCC group also received a lower end of the dosing with that 25 units per kilo infusion.  So for the result of this trial, our co-primary endpoints are listed across the first two rows at the top.  As you can see, there was a trend towards Andexa achieving greater hemostatic efficacy compared to the four-factor PCC group, as well as favorable outcomes related  to all-cause mortality for the Andexa group compared to the four-factor PCC group.  Of note, there were two patients who received Andexa and seven of those who received four-factor PCC who did die of worsening intracranial hemorrhage in the study.  There were no significant differences noted in hemostatic efficacy in the sensitivity analysis or in the subgroup analysis assessing those  with an intracranial or an intraventricular hemorrhage.  And lastly, across the bottom, we do see no significant differences noted in the thrombotic events, nor was there any statistical analysis performed for this particular parameter.  The authors concluded that Andexa is associated with better hemostatic effectiveness and a reduction in all-cause mortality at 30 days compared to four-factor PCC.  In accordance with our new updated guidelines, they do recommend Andexa over four-factor PCC for the management of a PIXIVAN or a RIVAROXIVAN-associated bleed.  My conclusions after evaluating this article are that we still desperately need prospective randomized clinical trials to really better assess the efficacy of Andexa  without requiring a synthetic control arm like what was used in this study. I do believe Andexa may be a feasible option for certain patients  when specific criteria is met, such as what was presented in this trial.  But overarchingly, I do feel that regardless of which reversal agent we're utilizing,  there's an inherent risk associated with reversing anticoagulation. And due to this, we do have patient-specific consideration.  And I think that's always warranted on a case-by-case basis when we're assessing patients with a clinically significant bleed.  A few thoughts of my general interpretation overall for this study is that there are some strengths to the study, and that being  that we do have a slightly larger comparison of our two reversal agents compared to some of our previous trials that have been published.  So, this does add to our knowledge of outcomes that were reported in this particular trial.  Majority of patients in both arms did receive the lower doses of reversal agents, which does make this a bit more of a homogenous data point to be able  to compare our two treatment groups and their respective outcomes. And a propensity score was utilized in this trial to minimize confounding  between our two treatment arms, which again, allows for a more randomized comparison between the two groups. Now, some of the limitations, by contrast, do stem from the design  of this indirect comparison study as it compares with the synthetic control arm. So, while the statistical methods were conducted to allow for the purpose  of mitigating those confounding variables, this ultimately is removing those baseline differences that potentially could have impacted our clinical outcomes,  making the synthetic control arm a bit more difficult to interpret at base value.  The hemostatic effectiveness was determined by independent clinical investigators. And while this was not well-defined in the study, it may have potentially imparted bias.  This particular outcome.  Thrombotic events were only assessed through day five following reversal, which doesn't necessarily account for the thrombotic events that may have occurred beyond this window.  And if we think back to the Anexa 4 trial, that could be as much as two-thirds of patients that were unaccounted for by not extending that follow-up period through about 30 days.  There were fairly restricted exclusion criteria, which again, did parallel that of the Anexa 4 study. But it restricts criteria to those with a GCS greater than eight  and a hematoma volume less than 60 milliliters, which overall, this limits applicability of these outcomes to our real-life patients who are presenting with a bleed.  And lastly, the study was funded by Olexian Pharmaceuticals, which is the drug company who makes Anexa. And that could have potentially imposed some level of bias as well.  So, for future directions and what's on the horizon for Anexa, the Anexa 1 trial is currently in progress and has an estimated completion date of 2024.  So, this is a prospective randomized trial that is aiming to really bridge the gap to provide the perspective we've been lacking thus far for a truly randomized trial  that compares Anexa to a standard of care. So, I think this will be really interesting to follow to see how the outcomes compare to what's already been published and potentially see  if this impacts, you know, our practices down the road.  The last question I wanted to ask is, does your institution currently have a protocol for specific criteria for use when considering using an XNA-alpha?  So, our institution does have some specific use criteria, so I thought it just might be interesting to see the different practices  across different institutions with regards to this medication. All right, so it looks like we have about 67% of participants are saying no at this point,  which, again, I think goes to show the differences between facilities and the routine use of different medications depending on where you practice. So, I think it's interesting.  Thank you for your participation.  All right, and with that, here are my references, and I really appreciate your time. I'd be happy to answer any questions you may have.  Thank you very much for your presentation, Dr. Gross. I'll go ahead and we can start our Q&A session. We have some questions coming in in the poll.  The first is, do you have any insight into why they did not perform a statistical analysis for thrombotic events, knowing that previous studies have shown an increase  in thrombotic events utilizing N-doxin and alpha?  Absolutely. I think that is a really good question, and I think that was one of my limitations, too, is looking at at least a follow-up period for thrombotic events.  I don't know why they didn't specifically do a statistical analysis, other than that the reports, at least for the five-day follow-up period, were so low.  I think at that point, they may have considered it to be negligible, but I certainly was considering that interesting, as well,  that they didn't have any statistical analysis for that specific data point.  And our next question here is, how do you think the dose of PCC affected the outcomes in this trial?  That was something we talked a lot about, too, as we were looking through the trial. I thought it was really interesting in terms of, again,  that both of our treatment groups were using a lower end of the dosing, so a bit more conservative from both respects in our treatment groups.  So I certainly think it would have been much more interesting, and perhaps in the future, to be looking at the upper dose of the 50 units per kilo of the four-factor PCC to be able  to compare that to a more conservative dose of Andexa, or even a higher dose to have a better representation of what the different dosing regimens might have as a clinical impact.  So I certainly think that may have posed some limitations and potentially impacted the outcomes that were observed in this trial, as well.  Okay, that concludes our Q&A session. Thank you very much, Dr. Goss. And now I will introduce our final presenter, Dr. Brittany Cran. Thank you so much for that introduction.  And today, I'm going to be presenting on a pre-hospital synergy, looking at the combination of transexamic acid and blood transfusions in patients at risk for hemorrhage.  So just briefly, an introduction. So patients who present acutely after their trauma, there are a lot of factors  that predispose them for, you know, potential short and long-term morbidity, mortality, one of those being coagulopathy.  So, of course, there's two kind of fields of thought with coagulopathy, both that thrombotic end, but then also the hemorrhagic end of that.  And so hemorrhage is one of the leading causes of survivable death and trauma. So there have been a handful of studies that have looked at, you know,  a variety of potential interventions to potentially prevent that, especially in the pre-hospital time period, so on transport from the scene to the hospital.  And one of those interventions has been transexamic acid, or TXA. And so it primarily works by preventing the conversion of plasminogen to plasmin,  which would then prevent that breakdown of that fibrin clot, which would hopefully prevent further bleeding and help to stabilize some  of those clots your patient is forming in that acute time period. And there have been, as I mentioned, a handful of studies examining the use of TXA in this patient population.  So the CRASH-2 study being one of these initial studies looking at patients presenting within eight hours of their initial injury, and they found that the administration  of TXA was most beneficial when given within three hours from injury. When you get outside of that window, so greater than three hours  from their initial injury, there's actually an increased risk of mortality. And that is what guided the MATTERS trial, which was primarily combat-related injuries,  but again, looking at the administration of TXA within three hours of that initial injury, and they found that patients who received TXA within three hours  of initial injury had increased battlefield survival, and then they also found that in the subgroup of patients who were objectively sicker, more hemodynamically unstable,  they were also demonstrated greater mortality benefits with the administration of TXA.  And then the CRASH-3 and Rowell and colleagues looked at the administration of TXA specifically in patients suffering from a primarily head-related injury, so TBI,  and they did not demonstrate necessarily the same mortality benefit associated with TXA at that point in time.  And so over the course, you know, since 2010, we've had multiple other single-center studies and meta-analyses as well as secondary analyses that have kind of guided our use  of TXA in this acute trauma population, and we've seen that the greatest benefit occurs when TXA is given closest to the injury, so within three hours from primary injury,  ideally within one hour, and then patients who also benefit most are those who are in more severe shock, so patients who are more hemodynamically unstable,  have higher injury severity scores seem to benefit the most. Unfortunately, dosing is a little less definite, as there have been a variety  of dosing strategies used across these trials and the ones that I'll mention, ranging from boluses out of hospital, followed by in-hospital boluses  or in-hospital continuous infusions, and then there is also the theoretical concern for thrombosis as well, just given the mechanism of TXA.  And so my first question is, just broadly, trying to find out, does your institution or hospital have EMS protocols which incorporate the use of pre-hospital TXA  in your trauma patients at risk for hemorrhage?  And it looks like it's relatively mixed. You know, 39% said no. Looks like another third said yes, and then another third said uncertain, so I can't say,  you know, I was necessarily certain of what our hospital protocols were beforehand, but we also here at UCMC do recommend the incorporation of TXA pre-hospital  in our at-risk trauma patients.  All righty, and then so this brings me to my study. So it's actually a secondary analysis of this parent stamp trial that came out in 2021,  looking at pre-hospital TXA in our at-risk trauma patients, and there have been two secondary analyses of this trial since then. One looking at early pre-hospital transexamic acid.  Again, we see that patients administered TXA within an hour of injury experienced the greatest mortality benefit.  And then my analysis is looking specifically at the combination of pre-hospital TXA plus additional packed red blood cell transfusion and looking at their outcomes.  So briefly, I just want to touch on the design of, of course, the parent-stamped trial. So this was a double-blind placebo-controlled trial in which patients could be randomized  into one of four groups, three of which being various TXA dosing strategies. So all patients received that one gram bolus on transport and then they could receive additional  infusions or boluses from there. So patients could receive a total of one gram, two grams, or three grams total of TXA. This was all compared to placebo.  And then patients were eligible for inclusion if they were transferred within two hours of injury.  So again, seeing that timeframe kind of falling into play here, and then they had to be considered at risk for hemorrhage.  So one episode of hypotension defined as systolic blood pressure less than 90 or tachycardia with a heart rate greater than 110 beats per minute.  And so the authors weren't able to demonstrate a 30-day mortality benefit in their overall groups.  So if we break down to the subgroups, that's where we see kind of this repeat demonstration of where our true mortality benefit seems to lie. So again, you see time from injury.  Patients who received TXA within one hour seem to benefit the most, as well as our patients who are more severely or acutely ill.  So patients with systolic blood pressures initially less than 70 millimeters of mercury  that also favors the administration of transexamic acid and favors their longer-term mortality outcomes.  Authors did also note that repeat boluses, so the group that received a total of 3 grams of TXA, did also note an improved mortality rate as well.  So dosing starting to come into play with this as well. And then you'll see they also analyzed safety outcomes, so DBT, so that theoretical risk  of developing thrombosis after the administration of TXA, not found to be statistically significant across this patient population. And now that brings us to the secondary analysis.  So in the STAMP trial, they included and collected on patients who received out-of-hospital pactored  blood cells or additional resuscitation with product, but they didn't analyze that. So this is an interesting question because there have been other studies who have looked  at the use of pre-hospital pactored blood cells by themselves, and they've been able to demonstrate some level of mortality benefit and improval in 24- and 30-day reduction in  mortality in patients who receive pactored blood cells by themselves. So then that stems the question of what is the combination of pactored blood cells and  TXA do for short- and long-term mortality. And that's where the concept for this secondary analysis came along.  And so the secondary analysis of STAMP, so again, very similar inclusion-exclusion criteria of the parent trial.  Some of the key differences, though, is that there were 12 of 24 total EMS bases, all air transport-based, who had the ability to provide pre-hospital pactored blood cells.  So not everyone had the ability to provide them, but of those who did receive them, you can see the breakdown across the groups below.  So patients who received primarily just TXA, a combination of TXA and pactored alone, or they did not receive either intervention.  You can see the criteria required for patients to receive pactored blood cells. So they would need to have received one liter of crystalloid, penetrating injury, evidence  of active bleed. And then you can see this list of hemodynamic and also lab components that would also lead  to them requiring or meeting criteria for that pactored blood cell infusion. So automatically, these patients are objectively sicker than your patients who might not have  received the pactoreds from or initially. And then it's important to note there were 127 patients from the parent STAMP trial who  were not able to be included in this analysis because they were transferred before they arrived at their primary trauma center and they were unable to collect the necessary  data for them. So their analysis, their outcomes were both short and long-term mortality. And the statistical analysis.  So since this was a secondary analysis of STAMP, the authors got a little creative with their statistics to try to overcome the selection bias that they experienced.  So using this Cox proportional hazard regression, using that placebo group as the reference. So patients who did not receive either TXA or pactored blood cells before arrival at  the hospital. And they did adjust those models for other factors, including ICU admission, what dose they received of TXA in hospital to review as well.  And then this propensity score is what they used to reflect the likelihood of receiving pre-hospital pactored blood cells.  So as I mentioned, the patients who received the pre-hospital pactored were objectively sicker.  So they use this propensity score to try to account for that and make their analysis a little less biased.  And you can see the included components below, including age, looking at a lot of their pre-hospital  vitals, how long it took them to arrive at the hospital if they were intubated, and then looking at abbreviated injury scores for these patients as well.  Looking at the baseline characteristics, so this breaks down the four total groups that patients could be in. I think it's most helpful to look at column three and four.  So of the patients who received pactored blood cells, regardless of whether or not they received  TXA, it's interesting because when you look at their baseline characteristics, that pre-hospital  systolic blood pressure stands out to you much lower than the TXA alone or the placebo group.  So looking at systolic in the 80s versus systolic in the 120s, dropping down to that second to last column, injury severity score, the patients who received pactored blood cells  had a much higher injury severity score when compared to the other groups. And there were also a higher percentage of patients who met the definition for multi-organ  failure as well across the groups who received pactored blood cells as opposed to that placebo or TXA only group.  So again, just at baseline, these patients look a lot, lot sicker than our other groups. And then looking at additional resuscitative requirements, you can see listed the pactoreds,  total plasma, and then crystalloid volumes received across the groups within the first 24 hours.  Again, noting that the patients who received pactored blood cells with or without TXA again required higher volume of crystalloid resuscitation within that first 24 hours.  And then dropping down to this graph of mortality across time points. So this was an unadjusted look first at mortality.  And you can see at each time point, three hours, six hours, 24 hours, and at 30 days, that gray bar representing TXA and pactoreds, and that light blue bar representing pactored  blood cells is visually has a lower mortality at each time point when compared to your TXA alone, which is interesting enough. And then this jumps us into our primary analysis.  So again, this was the cost proportional hazard regression at both 24 hours and 30 days. And you can see that placebo only group was our reference group.  And just jumping across, so 24-hour mortality, you can see the TXA and pactored blood cells  group, while not statistically significant, there was an association of decreased mortality rates in that group.  But then in that same group, when you move into that 30-day mortality, you do see that 35% decrease in the rate of or risk of mortality.  So potentially seeing some level of benefit in mortality in the longer term with that combination.  And then looking at pactored blood cells alone, you can see that at 24 hours, there was a 61% reduction in that hazard of mortality.  But then as you move into that 30-day mortality, you lose that significance and potential benefit of pactoreds alone.  Also just interesting to call out the TXA dose, so this was the dose that they received in hospital, doesn't appear that additional in-hospital dosing affected either short or  longer-term mortality outcomes. And so the authors concluded, based on this, that they would consider administering TXA  in any patient who met criteria for pre-hospital pactored blood cells. So again, pushing for that use of additional multimodal pre-hospital resuscitative efforts.  So in combination with what some might consider a little more standard of therapy being that TXA, and now starting to add on the pactored blood cells.  So my next polling question is based off of your local EMS protocol. So does your ED health system, pharmacists, physicians play a role in the development of local EMS protocol?  And so I think, you know, the reason I'm asking this one is to find out, you know, if there were studies coming out and you wanted to change local protocol, how would you approach  that? Is there a possibility to do that? And it looks like a majority of participants said yes.  It looks like a similar majority to the question before said they were about uncertain, which kind of rings true, but it is nice to know that a fair amount of our health systems do  play a role in our development of EMS protocols.  And then just jumping into the analysis of this study. So there are a variety of strengths and weaknesses based on the way the authors wrote this study.  So the research question, definitely a great question is the more we can do to prevent or to reduce mortality is optimal.  So it's interesting taking some of these individual components that have shown mortality benefit and putting it together.  An additional question I would have, though, is, you know, what about other potential resuscitative products?  So like plasma, as other trials, including the PAMPER and COMBAT trial, have shown that benefit of plasma on, or pre-hospital plasma on reducing mortality outcomes.  Inclusion, or excuse me, methods and designs, while mostly based off the parent STAMP trial, I thought they were all reasonable.  Their outcomes are very patient-specific, things that patients would care about. Again, 24-hour and 30-day mortality.  Again, with the design being a post hoc analysis, it is limited to those, you know, traditional weaknesses. So again, this is more hypothesis generating than anything.  And again, the number of our included patients was relatively low in the two groups who received pactorate blood cells alone or in combination with TXA.  For statistical design, I think the authors did the best they could to eliminate that selection bias.  So using that propensity score, that hazard regression, but truly unable to eliminate that altogether.  And even when they did their interaction testing, they still found there were a variety of patient  components, patient hemodynamics, lab values that contributed to interaction testing across the group.  So of course, the patients who received pactorate blood cells had higher lactates, lower systolic  blood pressures, things like that, which would definitely confound some of the interpretation. And then generalizability. These results are relatively generalizable.  I think one of the biggest components to this is going to be facility capability to administer pactorate blood cells.  So a lot of facilities are unable to house pactorate blood cells on their ambulances or might not have the support to be able to do that.  So then finally, just running through risks and benefits. So those risks, again, from the parent stamp trial, no difference in primary VTE outcomes,  but we did see in the secondary analysis that decreased mortality at 30 days, which would lead me to clinical applicability.  So it does seem reasonable to incorporate the use of both early TXA as well as packed red blood cells, if that is feasible for a facility for the pre-hospital resuscitation  of trauma patients. And again, the benefit is mostly seen in patients with higher injury severity scores, lower pre-hospital systolic blood pressures.  So again, these patients are more sick, more hemodynamically unstable on presentation than some of the other patients.  Just touching a bit on our current practice here at UCMC, our mobile care, air care is able to administer pactorate blood cells, but again, our ground transport is not.  We do play a role in the development of our Southwest Ohio EMS protocols, which we do include the recommendations for the use of pre-hospital TXA, but at this point in time,  do not recommend pre-hospital pactorate blood cells. And moving into next steps, so continuing to expand on multimodal pre-hospital resuscitation.  So outside of pactorate blood cells, could we use other resuscitative components like plasma as well, and what benefit lies there?  Looking at the patient population, there is a patch trial that's currently ongoing. It's a randomized placebo-controlled trial looking at more high-risk patients to evaluate  the use of TXA compared to standard of care in patients presenting within three hours of injury. And again, that dosing question still remains in this patient population.  So what is the ideal TXA dose across the patient, this patient population? And with that, I will go ahead and take any questions. Thank you very much, Dr. Cran.  If anyone else has questions, you can put them in the question box. We have some coming in. We can go ahead and get started with the first one.  The first question is, does your institution utilize whole blood in combination with TXA? Have you come across any studies evaluating the combination of whole blood and TXA?  Yeah, so that's a good question. So at this point in time, for pre-hospital management of our patients, you know, our recommendations are primarily focusing on the use of TXA alone.  We don't have recommendations for combination of TXA with either whole blood, pactoid blood cells, so primarily TXA.  And in my review, I did not find any studies that looked at the combination of TXA with whole blood.  So still, most of the data seems to be kind of extrapolations from larger trials like this. But yeah, no combinations with whole blood at this point in time.  Next question that was put in is, were there any data in this post hoc analysis regarding thrombosis? Yeah, so that's a great question.  And I think, you know, that's one of the questions most of the studies bring up with each study looking at transexamic acid.  Now this post hoc analysis specifically did not look at rates of DVT, PE, VTE across the different groups, unfortunately.  So we extrapolated that data primarily from that parent trial. So the STAMP trial overall did not show any significant difference in VTE rates, but yeah,  this secondary analysis did not analyze that across the four different sub-treatment groups. And our final question, why do you think there was a difference between TXA plus PRBC  versus PRBC outcomes, specifically, or specifically that the TXA plus PRBC received PRBC in their intervention? Sorry, do you mind repeating that?  Is that looking at the difference between TXA plus PRBC versus the PRBC group or versus the TXA group?  To clarify which group, it just said, why do you think there was a difference between the TXA plus PRBC versus PRBC outcomes?  I'm assuming this is in the TXA plus PRBC given that they all got PRBCs in the intervention. Okay. Yeah, so I mean, it's definitely interesting.  It's hard to tell which group had the greatest effect on outcomes. So again, yeah, that TXA and PRBC group, just visually, those mortality outcomes were lower at each time point.  When compared to either the TXA group alone or the PACTRED blood cell group alone, I think  the PACTRED blood cell group at baseline reflected the TXA and PRBC combined group a little more  similarly, as these groups were both pretty severely ill, the more hemodynamically unstable at baseline. So I think the benefit there was definitely seen.  I think there is potentially greater benefit in that combination, TXA and PRBC group. But then when you compare the TXA plus PRBC group to your TXA group alone, I think that  gets a little more confusing, because again, their baseline characteristics were a little more different. The TXA group was objectively less acutely ill.  So I think that might potentially be a confounding factor when you're trying to analyze those between group differences.  And then when the authors analyzed it, it was mostly, you know, compared to that, neither group.  So visually, I think it's interesting to see the differences, but it's hard to say for certain what exactly led to those, or what you could chalk those between group differences  up to.  Okay, and I think there's one more quick question. If we have time, does your institution nebulize TXA for airway hemorrhage?  Yeah, so that, I'm not actually sure what our protocols are, but we have utilized TXA nebulized for airway hemorrhage.  Yes, yeah, I have seen that done both in the ED and then in some of our ICU level patients as well. Okay, that concludes our Q&A session for this presentation.  Thank you very much, Dr. Cran. And thank you to our presenters today and 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.  Thank you for joining us.

Video Summary

In this Journal Club Spotlight on Pharmacy webcast, three presenters discussed different articles. The first presenter discussed a study on the association of vancomycin plus piperacillin/tazobactam with changes in creatinine versus cystatin C in critically ill adults. The study found that the combination of vancomycin plus piperacillin/tazobactam increased acute kidney injury risk but did not affect other kidney function biomarkers or clinical outcomes such as mortality and need for dialysis. The presenter concluded that the rise in creatinine levels may not necessarily indicate nephrotoxicity and that the combination can be used in specific situations. The second presenter discussed a study on the use of andaxanet alpha versus four-factor prothrombin complex concentrate for the reversal of apixaban or rivaroxaban-associated intracranial hemorrhage. The study found that andaxanet alpha was associated with better hemostatic effectiveness and a reduction in all-cause mortality at 30 days compared to four-factor prothrombin complex concentrate. The presenter recommended the use of andaxanet alpha over four-factor prothrombin complex concentrate for the management of factor 10a-associated intracranial hemorrhage. The third presenter discussed a study on the combination of tranexamic acid and blood transfusions in patients at risk for hemorrhage. The study found that the combination of tranexamic acid and packed red blood cell transfusions was associated with decreased mortality at 30 days compared to tranexamic acid alone or packed red blood cell transfusions alone. The presenter concluded that it was reasonable to incorporate the use of both tranexamic acid and packed red blood cell transfusions for pre-hospital resuscitation of trauma patients. Overall, the studies highlighted the importance of individualizing treatment based on patient characteristics and clinical judgment. The presenters acknowledged the limitations of the studies and the need for further research.

Asset Subtitle

Pharmacology, Neuroscience, Trauma, 2022

Asset Caption

"The Journal Club: Spotlight on Pharmacy webcast series focuses on pharmacy topics. This event is held on the third Friday of each month and features lively discussion and in-depth presentations on the latest research.

Dimitar Zelenkov, PharmD (Twitter handle: @zelenkov_rx)
Miano TA, Hennessy S, Yang W, et al. Association of vancomycin plus piperacillin-tazobactam with early changes in creatinine versus cystatin C in critically ill adults: a prospective cohort study. Intensive Care Med. 2022 Jul 14. doi: 10.1007/s00134-022-06811-0. Online ahead of print.

Savannah Gross, PharmD
Costa OS, Connolly SJ, Sharma M, et al. Andexanet alfa versus four-factor prothrombin complex concentrate for the reversal of apixaban- or rivaroxaban-associated intracranial hemorrhage: a propensity score-overlap weighted analysis. Crit Care. 2022 Jun 16;26(1):180.

Brittany Curran, PharmD
Deeb AP, Hoteit L, Li S, et al. Prehospital synergy: tranexamic acid and blood transfusion in patients at high risk for hemorrhage. J Trauma Acute Care Surg. 2022 Jul 1;93(1):52-58.

Follow the conversation at #SCCMCPPJC."

Meta Tag

Content Type Webcast

Knowledge Area Pharmacology

Knowledge Area Neuroscience

Knowledge Area Trauma

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Tag Antibiotics

Tag Cerebral Blood Flow

Tag Hemorrhage

Year 2022

Keywords

Journal Club Spotlight on Pharmacy

vancomycin plus piperacillin/tazobactam

creatinine

cystatin C

acute kidney injury

andaxanet alpha

intracranial hemorrhage

tranexamic acid

blood transfusions

resuscitation

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.