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

Video Transcription

Hello, everyone, and welcome to today's Journal Club Spotlight on Pharmacy webcast, which is supported by the Society of Critical Care Medicine CPP section.  My name is Corey Goodwin, clinical pharmacy specialist in critical care and PGY-2 critical  care pharmacy residency program at Carilion Roanoke Memorial Hospital in Roanoke, Virginia. Today I'll be moderating today's webcast.  And remember, a recording of this webcast will be available to registered attendees. You can get that information on mysucm.org and navigate to the My Learning tab to access  this information. Once again, thank you for joining us. A few housekeeping items before we get started. There will be a Q&A after each of today's speakers.  To submit a question, just type into the question box located on your control panel, please. You'll also have the opportunity to participate in several interactive polls.  When you see a poll, please just submit or click on the bubble next to your choice and click Submit.  And then please note the disclaimer of this, stating that the content is a follow-up for educational purposes only.  Okay, and for today's speakers, each of our speakers will be giving 15 minutes, followed by the question and answer session.  Our first presenter today is Brittany Harnisher, who's our PGY-2 emergency medicine resident at Inner Mountain Medical Center in Murray, Utah.  And today, she will be talking about intravenous connected place compared with alteplase for acute ischemic stroke in Canada.  Our second presenter is Emerina Most, who is a critical care pharmacy resident at Michigan Medicine, University of Michigan in Ann Arbor.  And today, she will be talking about the effect of therapeutic drug monitoring-based dose optimization of ciptazo on sepsis-related organ dysfunction in patients with sepsis.  And our third presenter today will be Emily Duff, another critical care pharmacy resident. She's at Bay State Medical Center in Springfield, Massachusetts.  Today, she'll be talking about an article, Propofol versus midazolam sedation in patients with cardiogenic shock, an observational propensity-matched study.  And now, I'll turn it over to our first presenter. Thank you, Brittany. Thank you, Corey. So, yes, I'm Brittany Harnisher, and I'm from Inner Mountain Medical Center.  I'm a PGY-2 in emergency medicine. And today, I will be presenting on IV-connected place compared with alteplase for acute ischemic stroke in Canada.  And this is also known as the NIN or the ACT trial.  So, there are other guidelines available for acute ischemic stroke, but for the purpose of this presentation, we'll note the following two.  The 2019 guidelines or the 2019 update to the 2018 American Heart Association and American  Stroke Association guidelines recommend alteplase in patients presenting with neurological deficits  who present within at least four and a half hours of symptom onset or are shown to be eligible per the wake-up protocol and who do not have contraindications to fibrinolysis.  These guidelines mention that tenecteplase may be an alternative to alteplase in patients who are candidates for mechanical thrombectomy, who have minor deficits, and without major  occlusion. The Canadian stroke guidelines from 2018 will also reflect the same alteplase recommendations.  However, they do not specifically recommend tenecteplase as an alternative due to the lack of evidence.  Just keep in mind that these guidelines, again, were published in 2018 and updated in 2019. And since then, about three trials have been published.  And so, these guidelines do not reflect those recent publications.  So, acute ischemic stroke is when there is a disruption to blood flow to the parts of the brain.  It's typically caused by thrombus or a ruptured plaque and is usually treated with fibrinolysis, such as alteplase or tenecteplase, and or mechanical thrombectomy.  Alteplase is our current gold standard for the treatment of acute ischemic stroke. It does require reconstitution and it's administered as a bolus followed by an infusion, whereas  tenecteplase is currently the preferred agent in acute myocardial infarction without percutaneous coronary intervention.  It's actually a modified variant of alteplase with greater specificity towards fibrin and a longer half-life. And it's actually administered as a single bolus.  And so, these properties have put a lot of research into determining the safety and efficacy of tenecteplase used in acute ischemic stroke.  And so, for the purpose of this journal club, I'm not going to cover all of the literature in detail.  But as a quick review, the NINDS trial showed benefit in alteplase versus placebo in patients presenting within at least three hours of symptom onset, while the ECAS-3 trial showed  benefit up to four and a half hours. There are some data looking at tenecteplase versus alteplase, the first being the NOR test trial.  It looked at specifically tenecteplase dosed at 0.4 milligrams per kilogram in mild stroke patients.  And between that and alteplase, they found no difference in MRS scores, which is modifying brink and scale, at three months.  And they did end up showing an increase in bleed because of the high dose of tenecteplase likely.  The NOR test 2A trial, which is shown on the slide, reflected the same intervention. So, tenecteplase dosed at 0.4 milligrams per kilogram, as well as our standard alteplase dosing.  This studied more moderate to severe stroke. And this trial was actually stopped due to an increase in symptomatic intracranial hemorrhage  due to tenecteplase, and again, likely related to that higher dose. Because of this, there is an ongoing trial assessing a lower dose of tenecteplase at  0.25 milligrams per kilogram. And this is the NOR test 2B trial. Lastly, we have the EXTEND IAT and K trial that found that tenecteplase was associated  with a higher reperfusion rate and lower modified brink and scale score at 90 days and found no difference in safety outcomes.  So the ACT trial was written by ACT trial investigators. It was published recently in the Lancet in June of 2022, and it was funded by a few research  organizations, the Canadian Institute of Health Research and the Alberta Strategy for Patient-Oriented Research Support Unit.  As for their methods, this was a phase 3 non-inferiority trial. It was multi-centred, taking place at 22 Canadian stroke centers. It was parallel, and it was an open-label design.  However, the outcomes assessment was blinded. Patients included adults who had acute ischemic stroke who presented within at least four  and a half hours of symptom onset, and who had a neurological deficit as well. It did exclude patients with the typical contraindications to fibrinolysis as well as pregnancy.  So patients were randomized one-to-one to either tenecteplase given as an IV push at 0.25 milligrams per kilogram with a maximum dose of 25 milligrams, or they were randomized  to alteplase, dosed as a total of 0.9 milligrams per kilogram with a maximum of 90 milligrams.  And this is given as a bolus followed by an infusion, and again it does require that reconstitution.  The primary outcome was the modified Rankin scale score of 0 to 1, meaning mild disability symptoms or no disability symptoms, and this was at 90 to 120 days.  They had various secondary outcomes that will be listed on a graph later in this presentation.  Safety outcomes did include symptomatic intracerebral hemorrhage, other bleeding, as well as angioedema.  The authors did use a 5% non-inferiority margin to test for their statistics, and an alpha of 0.025 to test for superiority.  They did estimate that 1,600 patients were needed to reach a 90% power, and they used Kaplan-Meier to assess for their 90-day mortality.  Authors did include an intention-to-treat analysis and a protocol analysis that can be found in this supplementary index. So overall, the baseline characteristics were well-matched.  There was 806 patients randomized to tenecteplase and 771 to alteplase. There was no difference in respect to the median age or sex.  The median age was anywhere from 73 to 74, and just over 50% of their population was male. The intercutile range baseline for the NIHS score was 9 in tenecteplase group and 10 in  the alteplase group. Further, large vessel occlusions were similar between the two groups with about 35% of patients having clots in the M1 and M2 middle cerebral arteries.  The average Jordan needle time was just above 35 minutes in both groups, but otherwise their workflow times and other characteristics were similar.  And so the primary outcome of an MRS score of 0 to 1 occurred in 36.9% of patients in the tenecteplase group and 34.8% of patients in the alteplase group.  They found no difference, and it did meet their pre-specified non-inferiority margin of negative 5% with theirs being negative 2.6%.  They also found similar results in their secondary outcomes, which again assessed an MRS score  of 0 to 1 and 0 to 2, and both of these were adjusted for age, sex, stroke severity, stroke symptom onset to needle time, and source registry.  And again, they did not find any statistical difference. The actual MRS score between both of the groups was 2, and so again no difference seen.  And then the use of thrombectomy and the length of hospital stay were similar between the two groups as well.  And so this first figure does show the difference in our modified Rankin scale groups between  tenecteplase and alteplase, with alteplase shown on top. The x-axis is actually the percentage of patients within those defined MRS scores on a scale of 0 to 6.  Zero is basically zero disability symptoms, whereas five is severe disability symptoms, and six is death.  And you can basically see that there are more patients between the scores of 0 to 2 in the tenecteplase group versus alteplase group.  However, there's more of a significant difference between the 0 and 1 scores. The second figure is a forest plot that shows, again, no difference in outcomes when adjusted  for baseline characteristics, but overall you do see a trend favoring tenecteplase, but again no statistical significance. The outcomes were very similar between the two groups.  Death occurred in just over 15% in both groups. Symptomatic intracranial hemorrhage occurred in just over 3% of the total population, whereas  intracranial hemorrhage that was identified through imaging or possibly asymptomatic occurred in just about 20% of the total population.  Similar serious adverse events occurred in about 9 to 10% of patients total. So overall, the authors concluded that IV tenecteplase was non-inferior to alteplase  for their primary outcome of disability measured by an MRS score of 0 to 1, but it was not superior to alteplase.  They did not find any differences in safety outcomes or 90-day mortality or other secondary outcomes as well.  So for my critique, first, I think that the title and the abstract were appropriate without any perceived bias.  However, authors did mention that their aim in the abstract was to see if tenecteplase increased reperfusion versus alteplase.  However, their primary outcome was the MRS score or modified Rankin Scale score. Further, they did not specifically state their objective in their introduction, but otherwise  outcomes and the abstract were appropriate. Authors were also credible, and I did not perceive any pertinent financial bias from them.  And this was also published in the Lancet, which is a peer-reviewed journal with an impact factor of 202.7.  I think that their multicenter approach to their study design is a strength. However, the open label aspect can add some potential bias to their results.  But due to the time-sensitive nature of the intervention, I can totally understand why these authors did not want to use a placebo in like a double-blinded, double-dummy design.  But it could be considered in future studies where strategic planning is definitely built in because we all know that time is brain.  As for the population, the inclusion and exclusion criteria reflected those of other pivotal  trials, as well as prescribing recommendations for fibrinolytics that are currently used. So this was a strength. And their baseline characteristics were pretty well matched.  The one thing that they did not do is they did not collect information such as stroke history or comorbidities and other information that was collected during previous pivotal  trials. And so this can limit its generalizability to the entire population. As for their statistics, I think that the methods that they used were appropriate given  the trial design of non-inferiority and their goal to prove that alteplase or tenecteplase is just as good as alteplase.  As for their results, again, their aim in their abstract was for reperfusion, but they did not assess or evaluate reperfusion in this study.  So again, I think that was just something that was missed. And then overall, outcomes were assessed through a telephone interview on days 90 up to 120 days.  So there technically is a risk for interviewer bias, but a strength is that they blinded those interviewers to the intervention that the patient did receive.  And so hopefully that'll help mitigate that. As for their conclusion, based on the results that we read and looked at, I came to the  same conclusion as these authors, which is that tenecteplase is as effective and as safe as alteplase in treating acute ischemic stroke, especially in patients who are presenting  within four and a half hours of symptom onset, who have neurological deficits, and who do not have any contraindications to fibrinolysis.  Personally, I would love to see tenecteplase used in practice as an alternative to alteplase,  considering we have some data out there that shows no difference in outcomes, and the ease of administration as well makes it ideal.  And then further, most of the time you see that tenecteplase is cheaper than alteplase in most practice sites. And so here is my first polling question.  Where do you primarily practice? We'll give you guys a moment to answer this question on our poll.  Perfect, so it looks like most people do practice at the Comprehensive Stroke Centers. That is awesome. Thank you guys for entering in those answers.  And then our next polling question is, does your institution use Tenecteplase to treat acute ischemic stroke?  So the answers are Tenecteplase can be used as an alternative, or Tenecteplase is used as sole thrombolytic for acute ischemic stroke.  Tenecteplase or Alteplase can be used depending on the stroke type or severity. And then no, Alteplase is still being used, and you may also select that if you're at  a referral center. Perfect. So yes, I work at a Comprehensive Stroke Center, and we are currently also using Alteplase right now. Okay.  And so here are my references, and I am happy to answer any questions that there might be.  Okay, Brittany, one of the questions was, what do you think about the author's comment  regarding a trend toward superiority of Tinecteplase after adjustment for age, sex, and stroke?  I think that, like, honestly, with the trend towards, I think that there should be additional  baseline characteristics that we assess, especially when we're adjusting for some of those, but I think that even though there's a trend, it's still not statistically significant,  and so I think that, overall, my assessment is that both of them are just as good as each other with the current population that's being studied. It's a good question, thank you.  One more, given the easier preparation of T and K, did the authors further elaborate on why there was similar door-to-needle times, you know, for both of the arms?  I did not see anything that they elaborated the differences. Just thinking about typical presentation in the hospitals, I know that with at least studies,  sometimes it can take a moment to get people on board, I guess. One thing that I will mention, though, is that these did use an informed consent that  could have been done at least seven days after the administration of alteplase, so I don't  even know if that would have been the contributing factor is, you know, obtaining consent for the study itself, but that is also a great observation, especially because you would  think that with the quick administration time that it potentially might bring that time down with our TenectePlace group, our door-to-needle time.  Great, well, that concludes our question-and-answer session. Thank you, Dr. Hornischer. Thank you.  And before moving on to our next presenter, there will be another quick polling question  for today's attendees to gain a better understanding of our overall attendance and our support of SECM and the webcast.  Oh, let me ask this other question. Brittany, real quick. So, what do you think are some barriers to institutions switching over to TNK entirely? And does it make sense to even  have both options on formulary? So, that's a really good question. I've actually discussed it a little bit at my institution with a couple of preceptors.  And it will be a lot of work to switch to TNK if that's what we end up doing.  But I think that both fiber analytics possibly need to be kept on board,  considering the different indications that we use, especially with massive PEs and using alteplase. And it might be difficult for rural communities or rural hospitals or clinics  to really keep both of them. So, that's going to be another barrier,  because it might not make sense for those smaller institutions to have  multiple fiber analytics on board. But further, there's a lot of guidelines, documentations. At our institution, we use what's called power plans. And so,  that's like an order set that a doctor can easily place an order on a patient for alteplase.  And it comes with everything already on it, like the monitoring parameters and other labs and  things that we need to obtain for those patients. And so, it's going to be a lot of work to institute  TNKplace into a system. But overall, if it's as safe and as effective and it's even cheaper, I can see the benefits of actually switching to TNKplace if that makes sense.  Any other questions?  That's great. Thank you, Dr. Hornker.  Thank you. For those in attendance, if you could answer this audience poll, how many attendings are viewing  with you? So, if you could answer just me, two to five people, five to 10, or greater than 10, please.  All right. And thank you. We'll now turn things over to our second presenter, Emerina Most. Thank you, Dr. Goodwin.  I am happy to be here presenting Effective Therapeutic Drug Monitoring-Based Dose Optimization  of Piperacillin-Tazobactam on Sepsis-Related Organ Dysfunction in Patients with Sepsis. And we'll jump right into the background before discussing the actual article itself.  So a little bit for our background information. Patients with sepsis are often failed to meet pharmacokinetic and pharmacodynamic PKPD targets.  And it's been shown in the literature that they often have worse clinical outcomes, such as lower mortality and shorter hospital length of stay.  And it's also been demonstrated in the literature that up to 50% of critically ill patients do fail to reach target exposure with our current recommended beta-lactam antibiotic  dosing strategies. And this can be due to altered volume of distribution and drug clearance that we see in our critically  ill patients that may prohibit them from reaching the targeted historically 60 to 70% of free  time above MIC for gram-negative organisms, but also more recently a push towards attaining 100% free time over MIC.  And because of these different issues with attaining proper PKPD targets with beta-lactam  antibiotics, there have been various approaches that have been studied to help optimize this  with our beta-lactam antibiotics, such as continuous infusion administration or therapeutic drug monitoring TDM-guided therapy.  Before the article that we'll be discussing, there have been two previously published prospective studies evaluating TDM for piperacillin-tazobactam administration.  Diwali et al. reported that TDM led to higher attainment of PKPD targets for piperacillin-tazobactam, which was administered as an extended interval infusion.  And then additionally, Simi and colleagues reported similar findings with TDM and optimizing PKPD targets, and they were utilizing intermittent infusion of piperacillin-tazobactam.  However, it's important to note that up until this point, neither study had reported differences in clinical outcomes.  This leads us to the study hypothesis that we'll be evaluating. The authors hypothesize the question, does TDM-based dose optimization compared to fixed  dosing improve clinical outcomes in patients with sepsis treated with piperacillin-tazobactam as a continuous infusion?  Now diving into the methodology of the study, this was a randomized controlled patient-blinded trial.  Patients were randomized one-to-one to receive either TDM or no TDM, and both groups were administered piperacillin-tazobactam as a continuous infusion.  Piperacillin-tazobactam was first given as a loading dose of 4.5 grams, followed by the continuous infusion, corresponding to either 13.5 grams over the day, or they dose adjusted  it to 9 grams over 24 hours in patients with an EGFR of less than 20. Their target plasma concentration of free piperacillin concentration was defined as  four times the minimum inhibitory concentration of the pathogen, and if the pathogen was unknown then they utilized four times MIC of pseudomonas, MIC of 16.  This was a multicenter study that occurred in 13 sites in Germany, and for their sample size calculation, they calculated 234 patients needed to be enrolled to detect a difference  in their primary outcome of immune SOFA score, which they set as an alpha of 0.05 and a power of 80%, and this was an intention-to-treat study.  The primary outcome of this study was mean of daily total sequential organ failure assessment or SOFA score up to day 10.  Key secondary outcomes that we'll be discussing include all-cause mortality at 28 days, ICU and hospital length of stay, microbiological treatment response, and PK-PD indices.  And then moving on to our inclusion criteria for this study. The inclusion criteria included sepsis onset and start of piperacillin-tazobactam therapy  that occurred less than or equal to 24 hours prior to randomization, and it included adult patients.  Other exclusion criteria included contraindications to piperacillin-tazobactam, impaired liver  function defined as child PUC, and additionally, patients that did not have one measurement of piperacillin concentration within 24 hours after randomization.  Now moving into the results of this study. For our primary outcome, which as previously mentioned is our mean SOFA score over 10 days,  they found for our TDM group, including 125 patients, a mean SOFA score of 7.9 compared to our non-TDM guided group, including 124 patients of 8.2, which was not statistically  significant. Looking closer at our key secondary outcomes, for a 28-day all-cause mortality, the all-cause  mortality was lower in our TDM group, 27% compared to our no-TDM group of 32%. However, this was not statistically significant.  ICU length of stay was shorter in our TDM group, again, not statistically significant, and hospital length of stay trended towards being shorter but not statistically significant.  Microbiological cure, which was defined as either documented or suspected microbiological eradication, was higher in our TDM group, but as all of our other primary and secondary  outcomes, this was not statistically significant within the group.  And looking a little bit more closely at our piperacillin-tazobactam concentrations from day one through day five and attainment of our targeted PK-PD parameters, they found  that attainment of piperacillin concentration up to day five after randomization was more likely in our TDM guided group, as denoted on the right side of this picture.  Here we can see that on the left side, patients who were not utilizing TDM were more likely  to either be super therapeutic or sub-therapeutic all the way through day five, whereas our TDM guided group were more likely to be in range for our PK-PD indices.  Additionally, looking at individual days for attainment of target concentration, we were  more likely to achieve this with TDM, 37.3% versus 14.6%, which was statistically significant. And depicted here, it shows that utilizing TDM does help target our PK-PD indices that  we want to achieve with piperacillin-tazobactam that the authors defined for this study.  The authors concluded that overall a significant benefit in mean total SOFA score was not observed  for TDM guided therapy in patients with sepsis and piperacillin-tazobactam that was administered as a continuous infusion.  And they also noted that they will need studies with a higher enrollment of patients to confirm  the observed benefit in our 28-day mortality, given that the study was not powered for this secondary outcome.  Additionally, the authors noted that future studies should focus on certain patient populations that may specifically benefit from TDM guided therapy.  And moving on to discussion. So first and foremost, I would like to focus on the primary outcome that these authors chose to utilize for their endpoint.  Looking at mean SOFA score over 10 days does have the potential for various confounding variables and utilizing that as your mean for your primary outcome does run the risk  of alternative variables altering why there was not a difference that was seen between the two groups.  Additionally, the study design utilized piperacillin-tazobactam administered as a continuous infusion for PK-PD  target attainments versus traditional dosing that is typically administered as intermittent or extended interval for dosing strategies.  Also pointing out that authors decided to exclude patients with impaired liver dysfunction  despite this is not a manufacturer recommendation that we need to dose-adjust piperacillin-tazobactam based on hepatic dysfunction.  Additionally, the authors empirically chose to target plasma concentrations either four times the MIC of the causative agent or four times the MIC of pseudomonas, which is 16  if a pathogen was not isolated, which may be targeting higher plasma concentrations than we typically associate with our beta-lactams when we're thinking about just needing to  stay above our MIC for optimal time-dependent killing. Overall, the study does have limited generalizability given the administration method of piperacillin-tazobactam  as continuous infusion, and also the study was only in Germany, which does have low prevalence  of gram-negative resistance and a very different resistance pattern compared to what we see here in the United States.  Authors were not blinded to the study groups, and also the study was performed with a sepsis-II definition of SOFA. The authors did comment on that.  However, when adjusted for the new sepsis definition, there is a lower percentage of patients that did meet definition of sepsis and septic shock.  Number needed to treat for 28-day all-cause mortality was 24. Another thing to point out was, although they were utilizing TDM-guided therapy for  the piperacillin-tazobactam dosage adjustments, they did not have a specified dosage algorithm or software for consistent dose adjustments.  Additionally, our standard of care of de-escalating antibiotics was not performed in this study, and although PKPD target attainment was more likely to be achieved with TDM-guided  therapy, there was no significant difference in our median piperacillin concentrations observed.  Overall takeaway points is that TDM-guided piperacillin-tazobactam administration does increase our PKPD target attainment, however, clinical impact is still unclear, and regarding  this study specifically, there is limited clinical applicability given it was administered as a continuous infusion and geographical microbiological resistance patterns.  At present, this study will not be influencing my practice given my institution does not  offer piperacillin-tazobactam laboratory draws, and we do have limited beta-lactam antibiotic  administration options at my institution. Overall, further exploration of microbiological and clinical cure rates as a primary outcome  for TDM-guided therapy for beta-lactam antibiotics is needed to really help drive if TDM-guided  therapy is the way of the future and something that we should be implementing at our institutions.  And now I'll be moving on to the first audience polling question. How does your institution administer beta-lactam to critically ill patients?  Continuous infusion, extended interval, intermittent, depends on the beta-lactam or depends on access and other medications that are being given.  All right. So, as expected, no answers for continuous infusion, which I do agree is one of the biggest  limitations of this study was the administration method of piperacillin-tazobactam. And additionally, at my institution, we do utilize extended interval and intermittent,  depending upon the beta-lactam.  And moving on to question number two. Does your institution offer TDM for piperacillin-tazobactam administration? Yes, in-house lab. Yes, send-out lab. No, or unknown.  And 84% of participants answered no, which is the same here at my institution. We do not offer this for piperacillin-tazobactam monitoring.  And with that, I will take any questions. Great. The first question we have is, can you comment about for patients that have renal impairment?  You sort of touched on them excluding liver impairment, but could you elaborate more on how they incorporated patients with renal impairment? Yeah, absolutely.  So, for participants with renal impairment, they did empirically use a lower total daily dose of piperacillin-tazobactam.  And they used an EGFR cutoff of 20 to utilize 9 grams if they fell under the 20 for their EGFR.  Their total daily dose that they used for over a GFR of 20 was 13.5 grams, which is lower than the total daily dosing strategies that are utilized at my institution.  So, I was interested in those renal cutoffs that they utilized and the 20 EGFR. Thank you. Thanks.  And you sort of commented a little bit about this in your discussion as well, but could you elaborate more on the difference between the continuous and the extended interval?  And do you think there would be a difference in the TDM for those two groups, or if they used extended infusion instead of the continuous?  Yeah, I think that's a really good question. And I don't think from a clinical outcomes perspective, if we do start seeing a difference with TDM-guided  therapy, that we would see a difference with our continuous infusion or extended interval.  However, from a nursing administration and practicality standpoint, it is more useful to use extended interval given access, and access is usually a large issue.  So, when we're thinking about our beta-lactams and target attainment, likely with TDM-guided therapy,  we'd be able to target the PK-PD indices that we need just with our extended interval.  However, I think that is something that needs to be explored since that is a major limitation of this study. Thanks.  And the last question was related with patients who also receive vancomycin as part of their up-front therapy.  Do you have any thoughts of how that might sort of change their AKI since you're focusing on the TDM so much? Yeah, I think that's a really good question.  So, historically, we've always thought of piperacil and tezobactam and vancomycin being administered together as nephrotoxic.  However, new literature has been coming out showcasing that this might not be the case.  And while they didn't comment in the article directly about patients who concomitantly received  vancomycin and effect on renal function, I think that this is something that we are starting to have  a new conversation around regarding the nephrotoxicity of both of those agents being utilized together.  Great. Thank you. And I think you sort of answered one of the other questions. Well, the last one. Oh, here's another one.  What are your thoughts on the choice of the primary endpoint? I think you sort of commented on it already, but maybe just elaborate just a little more, please. Yeah, definitely.  I think the primary endpoint is something that investigators commonly utilize. However, there are quite a few limitations given all the confounding variables.  And more importantly, since they utilize the mean over 10 days, we know that up front, getting  antibiotics on in patients with sepsis and septic shock is really important within the first three hours and the first 24 hours.  So, utilizing the mean over 10 days, I think it's difficult to extrapolate to the critically ill patient population.  So, I would have been more interested if they used the mean over the first 24 hours or additionally  using other endpoints, such as eradication of infection, vasopressor requirements. I do think there are just too many confounding variables with the primary outcome and  specifically looking at it as a mean over 10 days.  Great. Well, that concludes our Q&A session. Thank you, Dr. Motz. And our final presenter today is Dr. Duff, who will be talking to us about sedation and cardiogenic shock.  Thank you. Thank you very much for that introduction. As said, I'll be talking about propofol versus midazolam sedation in patients with cardiogenic shock.  And as I titled my presentation, Shocking News. So, just to review the objectives for this journal club presentation, we will review  recommended sedation strategies in patients with cardiogenic shock. We will examine a retrospective study investigating the hemodynamic effects of  propofol versus midazolam sedation in patients with cardiogenic shock. And then we will discuss the results of the study and its implication for clinical practice.  To start off, I do have an audience polling question. Does the potential cardio-depressive effects of propofol influence your sedative choice? Yes, no, or depends on the situation?  So it looks like 63% voted on depends on the situation and 24% voted yes. Thank you for those answers. That will be helpful kind of as I go through the presentation.  So to provide some background, patients with cardiogenic shock are an understudied ICO cohort concerning the choice of sedative options.  And there really isn't much information supporting one agent versus another. And there are little recommendations as well.  The 2016 European Society of Cardiology Heart Failure Guidelines are the only guidelines that actually give a recommendation for sedation.  And here they prefer midazolam over propofol in patients with cardiogenic shock. However, they do not cite any studies to support this recommendation.  The wording that they use in their guidelines is that propofol can induce hypotension and have cardio-depressive side effects, and that midazolam may have fewer cardiac side effects  and thus is preferred with patients with heart failure or cardiogenic shock. The 2022 American Heart Association American College of Cardiology Heart Failure Guidelines  do not address sedation recommendations in cardiogenic shock. And the 2018 Society for Critical Care Medicine PADIS Guidelines, which we utilize as our main guidelines for sedation,  also do not address sedation recommendations in patients with cardiogenic shock. In general, these guidelines recommend using non-benzo agents over benzos.  And this is due to the improved short-term outcomes such as shorter ICU length of stay, less delirium, and shorter duration of mechanical ventilation.  The one thing that these guidelines do mention is that they suggest propofol over benzodiazepines for sedation in patients after cardiac surgery.  And while this is not looking specifically at patients in cardiogenic shock,  here they do say that the desirable consequences of propofol probably outweigh the undesirable consequences. And propofol has been shown to have a shorter time to light sedation  and shorter time to extubations in patients who underwent cardiac surgery. Like mentioned, patients with cardiogenic shock are understudied,  and there are very little studies that look at this. One study that does kind of address patients with cardiac issues  is patients who are undergoing therapeutic hypothermia after cardiac arrest. This trial here looked at propofol and remifentanil versus midazolam and fentanyl  for sedation during therapeutic hypothermia after cardiac arrest. It was a small, randomized, controlled trial of 59 patients  and was published in 2012 in the Intensive Care Medicine Journal. Here they found that time to offset of sedation was significantly lower  in patients given propofol in remifentanil versus midazolam. And they also found that patients given propofol and remifentanil  needed norepi infusions twice as often compared to the midazolam group. So they found here that patients that received propofol actually did require more pressers.  Moving on to our second polling question. What is your preferred sedative agent when managing an intubated patient with cardiogenic shock?  Is it propofol, midazolam, do you not have a preference, or do you prefer another agent?  All right, so it looks like 39% voted for propofol and 17% voted for midazolam or no preference, or 27% said other. Thank you very much for that.  So moving on, there is a new study that was published to help build on current literature looking at sedations in this understudied cohort of cardiogenic shock patients.  Propofol versus midazolam sedation in patients with cardiogenic shock was a retrospective observational propensity match study  that was published in the Journal of Critical Care in April of this year. This journal was funded by the German Research Foundation and the hospital that it was conducted at.  The objective of this study was to investigate the efficacy of propofol versus midazolam sedation with regard to catecholamine requirements.  It had a hypothesis of that using propofol for sedation in cardiogenic shock patients does not lead to more catecholamine requirements compared to midazolam.  Inclusion criteria included patients with cardiogenic shock treated in the cardiac ICU of the LMU University Hospital in Germany between the year of 2010 to 2021,  and they defined cardiogenic shock based on the ESC guidelines. These patients also required mechanical ventilation and were sedated primarily with either propofol or midazolam.  For exclusion criteria, they excluded patients who were receiving palliative care sedation, patients intubated after day 7 of ICU admission,  and patients who were sedated with other agents other than propofol and midazolam. For the study design, there were two groups that patients were put into.  There was the propofol group and the midazolam group. Groups were based on which sedation was used more often during the first 7 days. So for example, the propofol group,  patients who underwent propofol sedation lasting longer than the midazolam sedation during the first 7 days of ICU admission were put into the propofol group.  The goal RAS of these patients was negative 2 to 0. All patients received sufentanil for analgesia, and catecholamines were administered to patients to reach a MAP of at least 65.  The primary endpoint was mean time-weighted catecholamine dose at days 1 through 7 of the ICU stay.  Catecholamine doses were calculated by using a formula that added up the total dose of dobutamine, epi, and norepi.  And the secondary outcomes were patients with 30-day all-cause mortality, greater than or equal to BARC3 bleeding events, which is the Bleeding Academic Research Consortium,  and it's a way to standardize bleeding definitions. It has a score of 0 to 5, with 5 being fatal bleeding.  So here they were looking at a score of 3 or more, which indicated major bleeding. And they also looked at ischemic events 30 days after ICU admission, which included MI and CVA.  In regard to statistical analysis, appropriate tests for variables were conducted in this study, and p-values less than 0.05 were considered significant.  The study also did propensity matching, propensity score matching, with a one-to-one nearest neighbor algorithm with a caliper of 0.1.  The goal with propensity score matching is to basically balance potential confounders in each group  and make the patients alike in all factors except for the exposure to the medication. It's used for observational retrospective non-randomized trials.  And by doing this, it helps make this observational retrospective trial as reliable as possible by helping reduce bias, since it is non-randomized.  The caliper mentioned here is the max permitted difference between the subject groups. And for this trial, it was appropriate.  And they matched for potential mortality confounders a priori. And this included things such as age, lactate level, etiology of cardiogenic shock, GFR, and more.  For enrollment, they looked at the LMU shock registry, which included over a total of 1,000 patients.  And after excluding patients who were not mechanically ventilated or had missing information, that was brought down into their propofol sedation and uromidazolam sedation groups.  After doing the propensity score matching to make sure that the patients were as balanced as possible,  in the one-to-one method, both groups ended up having a total of 174 patients that were matched.  So baseline characteristics are listed in Table 1 of the study. Overall, the baseline conditions and cardiovascular risk factors were evenly distributed after matching.  And the only significant difference noted between the groups was in BMI, with 26.9 in the propofol group versus 25.8 in the midazolam group.  They also reported on ICU characteristics. The majority of ICU characteristics were similar throughout, with the only significant differences being duration of ICU stay.  Here we see that propofol had a longer duration of stay of nine days compared to midazolam at 7.1.  The average systolic blood pressure was higher in the propofol group at 111.8 versus 106.3 in the midazolam group.  And the use of balloon pumps in Tirofibin was higher in the midazolam group compared to the propofol group.  Otherwise, all the other ICU characteristics were similar, including serum creatinine, lactate levels, PCI intervention, and pellet treatment and ECMO treatment.  In the supplemental index, it reported the mean doses of propofol and midazolam used in the trial. So they are both here displayed in milligrams per hour.  I converted propofol doses to microgram per kilo per minute, as that is how we reported at our institution here.  So just for some reference, on day one, the average dose in microgram per kilo per minute was 26.8. It was 25.7 on day two, 16.6 day three, 14.2 on day four, and 11.6 on day five.  And these are comparable propofol doses that I have seen in clinical practice,  and in my opinion, a decent dose enough to see if there really was a difference between the two types of sedation.  For the midazolam doses, here on day one, the reported mean dose was 22.7 milligrams per hour. These are doses that are much higher than I've seen in clinical practice,  and I feel like this is important to note and keep in mind when we are evaluating the results of this trial.  Additionally, the mean sedation score between the two groups was not statistically different.  So looking at our primary endpoint, the time-weighted median catecholamine doses were significantly lower in the propofol group  compared to the midazolam groups on days one through four. At admission, the mean doses were similar between the two groups,  and then on days five through seven, there was not a statistical difference.  So looking at day one here, the mean dose in the midazolam group of catecholamines was 110 compared to 80 in the propofol group.  To kind of look at the results more closely, I standardized these mean doses to a 70 kilo average patient  and looking strictly at the norepi doses in microgram per kilo per minute using the equation that was given in the trial.  Based on this calculation, for the midazolam group, the mean dose was 0.26 micrograms per kilo per minute of norepi compared to 0.2 micrograms per kilo per minute of catecholamines.  Compared to 0.2 micrograms per kilo per minute in the propofol group. And the difference between the groups was a difference of 0.06 microgram per kilo per minute.  So while this result was found to be statistically significant, it's important to consider if you believe that this is a clinical significant difference in catecholamine doses.  However, what this does kind of tell us is that propofol did not require more catecholamine doses compared to midazolam in our cardiogenic shock patients.  Here is a graph that depicts the difference in catecholamine doses. As we can see on day 1, 2, 3, and 4, there was a significant difference depicted by the stars here.  Moving on to our secondary endpoints, mortality rate was 38% in the propofol group compared to 52% in the midazolam group after 30 days. So there was a lower mortality rate.  And there was a hazard ratio of 1.7, as you can see here, depicted on the mortality curve.  And then for the rate of greater than three BARK3 bleeding, it was significantly lower in the propofol group compared to the midazolam group.  There is thought that differences in bleeding could be due to differences in antiplatelet therapy.  But after looking at the baseline characteristics between the patients, the combined proportion of antiplatelets and the Tirofibin drug that we talked about earlier were evenly distributed between the two groups.  But the use of only the Tirofibin treatment was higher in the midazolam group.  And then looking at the rate of myocardial infarction and stroke was similar in between both groups.  So the authors concluded that they proposed that propofol sedation for patients with cardiogenic shock should be primarily considered over midazolam.  However, there is a need for a randomized trial to confirm these findings.  There are some strengths to the study. It includes it being the largest investigation to date, looking specifically at sedation in patients with cardiogenic shock.  They also use propensity score matching to help limit confounding factors that really help make this observational retrospective non-randomized trial as reliable as possible.  Baseline characteristics were matched and they were similar between two groups. They had two distinct treatment groups. The study talks about how there was a practice change.  Originally, they were treating patients with cardiogenic shock with midazolam. And then that kind of switched over to propofol over the last few years.  So they really did have two distinct treatment groups that they were able to kind of separate into our groups here looking at our study.  They also mentioned that there was no relevant new treatment introduced during the time period. And if it was, it was made sure to be evenly distributed between the two groups.  And lastly, sedatives were used at reasonable doses that we would be able to detect a difference in vasopressor requirements, in my opinion.  And then looking at limitations, it is a retrospective observational design.  That was not randomized. The indication for sedation was up to the physician's discretion, even though the study talks about there was a practice change.  Providers may have felt that they wanted to use midazolam in patients who were sicker, just because that was what the previous thought was.  Also, generalizability is limited as this was conducted at one center in Germany.  And the higher doses of midazolam compared to the US practice may also kind of limit the generalizability.  So my conclusions is that this study definitely challenges previous beliefs that propofol cannot be used in patients with cardiogenic shock.  And I feel more comfortable and confident to recommend propofol in these patients.  Because from this study, I was able to see that propofol did not require more pressors compared to midazolam.  But it would be nice to have a more robust trial, randomized clinical trial, that has more data and potentially allowing for some guideline recommendations on sedation choices.  This slide here shows my references.  I thank you for listening to my presentation, and I will take any questions now.  All right, thank you. Our first question is, given the lower mortality in the propofol group, would this signify that patients were not well matched using propensity scores?  And do you think this inequality in groups may have led to the increased catecholamine dose rather than the midazolam itself?  I think that's a really good point. As mentioned, propensity score matching, and especially with it being a retrospective observational design, it's not going to be perfect.  And I think they really did do a good job trying to match for particular mortality confounders. However, you really can't remove all of them.  And then when considering the differences in mortality, there likely probably could have been some other confounders as well. That makes it a little tricky to kind of evaluate.  However, it's helpful that if maybe there would be another clinical trial that would look more specifically, whether it was a prospective kind of trial, we would be able to see a true difference in mortality.  In this study, I feel that the results relating to that propofol didn't require more catecholamines is a bigger takeaway than I believe that the mortality since it is an observational retrospective trial.  Yeah, I agree. I think that's a good answer.  Second question, did they mention anything about performing spontaneous breathing trials? And did they comment on any pneumonia rates along with that?  That's definitely a great question since that is kind of our standard of practice.  I did not see anything mentioned about doing those specifically, but they did talk about how they did complete what the guideline recommendations were.  So I would assume that. However, I can't exactly say if they did that and what kind of how that reflected dosing.  But I would hope that they were practicing similar to how we practice here with that.  Agreed. And the last question for you is, so do you feel that or why do you feel that the midazolam group required more catecholamine therapy?  Do you think it's really due to some pharmacologic difference in the drugs themselves or something else that you noticed as you went through the trial?  Another great question. My first thought definitely was that's quite interesting that midazolam would require more catecholamines because that's different than everything I was taught and kind of seen before.  Once again, this was a retrospective observational propensity match trial.  So there was likely things that weren't matched appropriately or not thought of that could have related to the differences.  And one limitation I mentioned before is that the sedation was up to the physician's discretion.  So there could be some bias relating to the sicker patients being given midazolam versus propofol.  Another interesting point was the doses, the mean doses of midazolam given with the mean dose on day one being 22.6.  Now, I'm not sure if that has anything to do with mortality, but it is something to kind of consider when thinking about the increased mortality rate.  I think likely that has to do with it being an observational study rather than differences in pharmacology.  However, that would be something else good to investigate further.  All right. Thank you, Dr. Duff. And thank you for all of our other presenters today and every member of the audience for attending.  And please remember and mark your calendars for the next Journal Club, which is the third Friday of every month. And for September, that'll be September 16th at 2 p.m. Eastern.  Thank you again for joining.  That concludes our presentation.  Thank you.

Video Summary

In this video transcript, three presenters discuss different topics related to pharmacy. The first presenter discusses a study comparing the effectiveness of intravenous tenecteplase and alteplase for acute ischemic stroke in Canada. The study concluded that tenecteplase was non-inferior to alteplase in terms of disability outcomes. The second presenter discusses a study on the effect of therapeutic drug monitoring (TDM) based dose optimization of piperacillin-tazobactam on sepsis-related organ dysfunction in patients with sepsis. The study found that TDM-based dose optimization led to higher attainment of pharmacokinetic and pharmacodynamic targets. The third presenter discusses a study comparing propofol and midazolam sedation in patients with cardiogenic shock. The study found that propofol sedation did not lead to more catecholamine requirements compared to midazolam sedation. The study also found that propofol sedation was associated with lower mortality rates and less bleeding events. Overall, the presenters highlighted the potential benefits of these interventions and the need for further research in these areas.

Asset Subtitle

Pharmacology, Neuroscience, Sepsis, 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.

Brittany Harnicher, PharmD
Menon BK, Buck BH, Singh N, et al; AcT Trial Investigators. Intravenous tenecteplase compared with alteplase for acute ischaemic stroke in Canada (AcT): a pragmatic, multicentre, open-label, registry-linked, randomised, controlled, non-inferiority trial. Lancet. 2022 Jul 16;400(10347):161-169.

Amoreena Most, PharmD
Hagel S, Bach F, Brenner T, et al; TARGET Trial Investigators. Effect of therapeutic drug monitoring-based dose optimization of piperacillin/tazobactam on sepsis-related organ dysfunction in patients with sepsis: a randomized controlled trial. Intensive Care Med. 2022 Mar;48(3):311-321.

Emily Duff, PharmD
Scherer C, Kleeberger J, Kellnar A, et al. Propofol versus midazolam sedation in patients with cardiogenic shock: an observational propensity-matched study. J Crit Care. 2022 May 5;71:154051.

Follow the conversation at #SCCMCPPJC."

Meta Tag

Content Type Webcast

Knowledge Area Pharmacology

Knowledge Area Neuroscience

Knowledge Area Sepsis

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Tag Cerebral Ischemia

Tag Antibiotics

Tag Shock

Year 2022

Keywords

pharmacy

intravenous tenecteplase

alteplase

acute ischemic stroke

disability outcomes

therapeutic drug monitoring

piperacillin-tazobactam

sepsis-related organ dysfunction

propofol sedation

midazolam sedation

	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.