SCCM Resource Library-June Journal Club Webcast: 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 CPP section. My name is Kelly Rumball,  Critical Care Pharmacy Specialist in the Surgical ICU at Vanderbilt University Medical Center in Nashville, Tennessee. I will be moderating today's webcast.  A recording of this webcast will be available to registered attendees. Log into mysccm.org and navigate to the My Learning tab to access the recording.  A few housekeeping items before we get started. There will be a Q&A after each of today's speakers. To submit questions throughout the presentation,  type into the question box located on your control panel. You will also have the opportunity to participate in several interactive polls. When you see a poll,  simply click on the bubble next to your choice. You may also follow and participate in live discussion on Twitter, following hashtag SCCMCPPJC and hashtag 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 Tyler Browse,  PGY2 Critical Care Resident at Baylor University Medical Center in Dallas, Texas. Our second presenter is Michael Young, PGY2 Critical Care Resident  at Ohio Health Riverside Methodist Hospital in Columbus, Ohio. And our third presenter is Claire Sizz, PGY2 Critical Care Resident at Beth Israel Deaconess Medical Center  in Boston, Massachusetts. And now I'll turn things over to our first presenter.  Thank you for the nice introduction, Kelly. Like mentioned, my name is Tyler Browse. I am a PGY2 Critical Care Pharmacy Resident  at Baylor University Medical Center in Dallas, Texas. And today I will be presenting an article titled hemodynamic effects of ketamine compared with propofol or dexmedetomidine  as continuous ICU sedation. The 2018 PADIS guidelines emphasize the importance of inaugural sedation in a good practice statement where the authors wrote that the management of pain  should be guided by routine assessment and should be treated before a sedative agent is considered. To go along with this statement, the guidelines do suggest low dose ketamine  as an adjunct to opioid therapy in post-surgical adults that are admitted to the ICU. In the sedation section of the guidelines, the authors mentioned that non-benzodiazepine  sedatives, which include propofol or dexmedetomidine, are preferable to benzodiazepines for sedation in critically ill mechanically ventilated adults.  However, in practice, ketamine is occasionally used for inaugural sedation given its analgesic properties and its ability to sedate patients at higher doses.  But still, there is no mention of ketamine in the recommendations under sedation in the 2018 PADIS guidelines.  Let's quickly review and compare the two guidelines.  First, let's compare these three sedative options. Ketamine itself is an NMDA antagonist that may be used for sedation in mechanically ventilated adults.  When used, it has been noted that it can cause hypertension and tachycardia. It's also been found that this is mostly seen in patients who have adequate catecholamine reserves.  In patients who are catecholamine depleted, such as in septic shock or acute decompensated heart failure, ketamine may cause bradycardia and hypotension,  partly due to its negative inotropic effect. Are other sedatives used such as propofol, which is a GABA agonist, and dexmedetomidine, which is an alpha-2 agonist,  have both been associated with hypotension and bradycardia. These adverse effects make ketamine an attractive choice for sedation in critically ill patients.  On the bottom there are the non-hemodynamic adverse effects for your reference.  There is a paucity of literature describing the hemodynamic effects of ketamine when used alone for sedation in the ICU. The many trials that have looked at  its hemodynamic effects have used it in conjunction with other agents, therefore confounding the findings. Additionally, the literature that is published  is limited to retrospective studies at this time. However, in the recent years, the hemodynamic effects of ketamine have been of increased interest,  and many more studies have been coming out. In 2019, Buchheit and colleagues published a retrospective observational study looking at 40 patients who received ketamine  on a median dose of 0.3 mg per kick per hour, obviously for adjunctive pain control. The primary outcome of this study was the change in phenylephrine equivalent requirements  between the first and sixth hour of the ketamine infusion. And what the authors found in this trial is that the addition of ketamine resulted in a decrease of 30 milligrams per hour  in phenylephrine equivalents, which was a statistically significant finding. Additionally, Groth and colleagues published a multicenter retrospective observational study  this year that looked at almost 400 patients who were receiving a ketamine infusion for either adjunctive pain control sedation or analogous sedation as a whole.  The authors measured the incidence of hypertension, hypotension, tachycardia, and bradycardia in the first 48 hours after the ketamine was started.  The only difference was in the incidence of hypertension, where an increase from 24% of patients having an episode at four hours to 40% of patients at 48 hours  was found to be statistically significant. Given the lack of differences in the other hemodynamic parameters, the authors concluded that ketamine does not significantly alter  hemodynamic parameters in their patients.  This is a good stopping point for our first question.  At your institution, how frequently is continuous infusion ketamine used for sedation? Less than 25%, 25 to 50%, 50 to 75%, or greater than 75% of the time?  Okay, it looks like the majority of you, as expected, chose less than 25% for ketamine being used as a sedation. That mirrors what I have seen in practice  as well throughout our ICUs. We mainly use it for adjunctive pain control.  So this brings us to our current article that we will review. Ashley and colleagues published a trial this year in Inalza Pharmacotherapy,  examining the hemodynamic effects of ketamine compared with patients who had received propofol or dexmedetomidine as continuous ICU sedation.  This was a single-center, retrospective observational study that aimed to compare the development of clinically significant hypotension or bradycardia between critically ill adults  who received sedation with ketamine and either propofol or dexmedetomidine. Adult mechanically ventilated patients who received sedation with one of these agents  for at least 24 hours were included. 24 hours were chosen as a time period to ensure that sedation was used outside of procedural purposes in the study. And patients were excluded  if they received any of the study medications concurrently, if they were transferred from an outside hospital with sedation already on board, or if they received ketamine  for any indication other than sedation.  Out of the 833 patients that were screened, 234 were included in the analysis, which excluded 599. The primary reasons for exclusion here  were duration of sedation less than 24 hours or the receipt of propofol and dexmedetomidine concurrently. So in the end, there were 78 patients in the ketamine group  and 156 patients in the propofol or dexmedetomidine group. Of note, there was no distinct breakdown of how many patients received propofol  and how many patients received dexmedetomidine.  The primary endpoint was the incidence of clinically significant negative hemodynamic effects.  This endpoint could have been met by the patient having significant hypotension, defined as either a decrease in MAP by 20% or more, or a decrease in systolic blood pressure  by 30 millimeters of mercury or more. The response to these had to have been a fluid administration, a 20% or more increase in vasopressors, or a decrease in  or discontinuation of the sedative itself. The patient could also have met  the primary endpoint by having significant bradycardia, and this was defined as a 30 beats per minute or more decrease in the baseline heart rate,  or an absolute heart rate of less than 50 beats per minute. Additionally, secondary endpoints included ICU and hospital length of stay, duration of mechanical ventilation,  and mortality less than or equal to 12 hours after the sedation was discontinued.  Based on a two-in-one enrollment to the propofol, dexamethasomidine, and ketamine group, they calculated that 234 patients were required to provide 80% power to detect a difference  of 17% or more in the incidence of a negative hemodynamic event.  The student's T-test was used to analyze continuous data, the Mann-Whitney U-test for nonparametric data, and the Chi-square or Fisher's exact test for categorical data.  Additionally, the authors performed a multivariate logistic regression with variables that were thought to have an impact on the primary outcome, including baseline heart rate,  receipt of propofol or dexamethasomidine, duration of mechanical ventilation, and receipt of fentanyl infusions.  The average patient was 58 years old, and the majority of the male patients were in the propofol or dexamethasomidine group, but the overall population was split evenly  between male and female participants. Ketamine patients had significantly less coronary artery disease and were more frequently diagnosed with severe acute respiratory syndrome,  coronavirus-2, or SARS-CoV-2. Additionally, patients in the ketamine group had lower median hemodynamic parameters prior to sedative initiation,  and that includes systolic blood pressure, mean arterial pressure, and heart rate compared to the patients in the propofol or dexamethasomidine group.  Additionally, the authors recorded potential confounders of hemodynamics in these patients. There was a significantly longer amount of time  between ICU admission to the initiation of ketamine compared to the initiation of propofol or dexamethasomidine.  More patients in the ketamine group were on concomitant fentanyl and midazolam infusions, and despite the two groups having no difference in the overall SOFA score,  it did seem to me that the patients in the ketamine group were of higher system SOFA scores or higher acuity, with more having a diagnosis of septic shock,  more being on streptosteroids, on one or more vasopressors, and on renal replacement therapy.  For the results, the primary outcome of either clinically significant hypotension or bradycardia occurred in significantly fewer patients receiving ketamine,  and this result translated into the individual components of the outcome, which were hypotension and bradycardia separately.  Patients in the ketamine group received sedation for about 33 hours prior to developing a negative hemodynamic event compared with about 17 hours  receiving propofol or dexamethasomidine.  And then looking at the second data set, and then looking at the secondary endpoints, it was found that patients in the ketamine group had a significantly longer ICU length of stay,  but there was no difference in overall hospital length of stay. And they also had significantly longer durations of mechanical ventilation,  and a significantly higher rate of mortality, 12 hours or less, after the end of sedation.  Looking at the supplementary information of this trial, the authors offered additional insight into expected confounders in the study.  What they noted was that significantly more patients in the ketamine group were on vasopressors, including norepinephrine, epinephrine, and vasopressin during the research period.  However, a similar proportion of patients in both groups received vasopressor therapy prior to the negative hemodynamic event that they experienced.  The mean norepinephrine equivalent dose was significantly higher in the ketamine group, and as stated previously, significantly more patients in the ketamine group  were on concurrent fentanyl and midazolam infusions.  So what the authors concluded from these results is that ketamine was associated with less clinically relevant hypotension or bradycardia  when compared to propofol or dexmedetomidine when using these medications in critically ill mechanically ventilated adults that required sedation.  Ketamine also led to a smaller absolute decrease in hemodynamic parameters over the course of sedation, and resulted in significantly less hypertensive episodes  than propofol or dexmedetomidine.  Additionally, the authors noted that out of the four variables that were included in the multivariate logistic regression analysis, which were baseline heart rate,  receipt of propofol or dexmedetomidine, duration of mechanical ventilation, and receipt of fentanyl infusions, the only factor associated with negative hemodynamic events  in the multivariate regression model was the receipt of propofol or dexmedetomidine.  The strengths of this study include it being the first to compare the hemodynamic effects of these agents against each other, and the fact that the authors recorded  and reported potential confounding factors. Additionally, the study did utilize the multivariate logistic regression model, including those four variables  that we previously spoke about.  Limitations do include its retrospective single-center design, which introduces selection bias and relies on accurate EHR documentation.  Additionally, there were significant differences between the groups, including more patients in the ketamine group being on concurrent fentanyl and midazolam drips.  However, this might reflect the deeper sedation targets in patients with SARS-CoV-2, which there were significantly more in the ketamine group.  Fewer patients in the propofol and dexmedetomidine groups had a diagnosis of septic shock, and therefore were on less-stressed-dose steroids,  which were both significantly different between the groups.  And lastly, there was a significantly longer amount of time between ICU admission to the initiation of ketamine compared to propofol or dexmedetomidine,  which obviously could influence the hemodynamic effects of ketamine, given that it was initiated further in a patient's ICU progression.  And lastly, the study included patients from multiple ICUs and did not compare outcomes between ICU types in a subgroup analysis.  My interpretation of this study is that while patients in the ketamine group did have significantly less hypotension and bradycardia and seemed to be of higher acuity,  they did require significantly more vasoactive medications, which may complicate this conclusion. A great takeaway from the study, I think, is that even though we do expect ketamine  to cause bradycardia and hypotension in patients who present and are catecholamine depleted, the primary outcome overcame the significantly higher amount of patients  who did have a diagnosis of septic shock in the ketamine group.  Future directions of this trial include a prospective randomized controlled trial examining these agents when used alone for sedation, and a propensity-matched analysis  to account for pretty insurmountable confounders that are expected when performing a study looking at ICU sedation.  And that brings us to our final question, which is what is the most common dose-limiting adverse effect of ketamine that you have observed when utilizing ketamine as an infusion?  Is it dysphoria, bronchorrhea, tachycardia, or hypertension?  If it's something else, feel free not to vote.  Okay, pretty much what I expected, a good amount of diversity in our answer choices. I think tachycardia is obviously a big one that we come across,  depending on the patient population that ketamine is used in.  And that is all I have for you today. I will happily take any questions at this time.  If you do have a question, you may use the question box to type in your questions.  Tyler, did the article discuss how many patients received dexmedetomidine versus how many received propofol?  It did not, as far as I could tell, break down how many patients. They grouped those patients together.  Unless I missed it in the article, I don't remember them describing how many patients actually received either of those agents.  Thank you. Also, did they discuss, I guess, if a patient had multiple instances of the primary outcome, how they approached that or how they analyzed that?  You know, that was a self-reported limitation of their study that they talked about in the discussion is that they utilized, I guess, an easy way to put it  would be the most severe episode of hypotension or bradycardia over a certain time period.  So, as far as including multiple events under the same patient, it's not something that they did, as far as I could tell.  And then, I guess, another question would be, it looked like in the ketamine infusion group, there was a significant number of patients who were switched from either propofol  or dexmedetomidine to ketamine. Do you think that could have impacted their results in any way?  Yeah, yeah. In the patients that they included, if they had been on dexmedetomidine and propofol and then were switched to ketamine, they were still included in the ketamine group  if they had received 24 hours or more. I do think that, especially when thinking about dexmedetomidine, it takes a little bit longer to clear the system  than something like propofol. So, in the very early stages of a ketamine infusion, if being switched from dexmedetomidine, you could definitely have some hemodynamic effects  of dexmedetomidine still hanging around. So, I think that could definitely affect the hemodynamic effects of ketamine as it would have been reported in the study.  Okay, then that concludes our Q&A session. Thank you very much, Tyler. Thank you.  Before moving on 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 this Spotlight on Pharmacy webcast.  How many attendees are in attendance? How many attendees are you viewing this webinar with? And please go ahead and answer that.  Okay, now I'd like to introduce our second presenter, Michael Young.  Thanks for the introduction, Kelly. As mentioned previously, my name is Michael Young, and I'm a current second year  critical care pharmacy practice resident at Ohio Health Riverside Methodist Hospital. And today, my journal club is going to be focusing on the article, The Evaluation of Outcomes  Following a Hospital-Wide Implementation of a Subcutaneous Insulin Protocol for Diabetic Ketoacidosis, or DKA.  There are going to be a few abbreviations that I'll be using throughout the presentation today. So, I'll pause here to make yourself familiar with a few of these.  And before moving into the article, I'll give a brief background over DKA in general. DKA is an acute hyperglycemic emergency, generally centered around three characteristics  comprised of acidosis, ketosis, and hyperglycemia. As most of us are familiar with, laboratory values to indicate this disease state generally include an RTR pH less than 7.3,  glucose greater than 250, and then moderate ketoneuria or ketonemia.  As far as the epidemiology of DKA goes, DKA is associated with significant morbidity, accounting for greater than 500,000 of hospital days per year.  There's an overall average hospital length of stay of 3.24 days as well. And DKA also contributes towards substantial economic burden in the United States,  with an estimated annual cost of 5.1 billion US dollars, and overall $26,000 per admission. And this is only expected to increase over time,  as we can see here by the following chart.  So if we look at the following bar chart, where we see 140,000 DKA admissions in 2009, and then fast forward to 2017, we almost see a doubling in that rate.  When we focus on the therapeutic management, we're generally guided by the American Diabetes Association or ADA, as well as the American Association  of Clinical Endocrinology or AACE. And there's generally three cornerstones of therapy that are including fluid resuscitation,  and intravenous insulin. The recommendation for IV insulin has been carried over from the 1977 New England Journal of Medicine article by Fisher and colleagues,  which showed significant quicker correction of plasma glucose and ketone bodies compared to sub-Q or IM insulin.  Before going any further, this article does focus on quite a bit of medication safety. So I'd like to point out a few specifics on insulin and medication safety in general.  Insulin products are classified as high alert medications by the Institute of Safe Medication Practice or ISMP. And they're actually the leading cause of harmful medication errors  with approximately 16% of errors attributed towards its use. Additionally, according to a 2014 survey by pharmacists and nurses, subcutaneous insulin specifically is listed last  regarding confidence and effectiveness of hospital-wide precaution to prevent medication errors.  Insulin errors can occur in many ways. There are multiple safety concerns that exist within an insulin-based protocol. These includes improper transitions of care points,  inappropriate timing, dose submissions. And I'd like to keep those in mind as we move throughout this study as well. There's also specific safety concerns,  especially existing within subcutaneous administration in critically ill patients. So a lot of our ICU patients may be placed on vasopressors, have abnormally high body habitus,  or may also be receiving some other medications which may raise or lower glucose in general.  There is an increasing amount of evidence for subcutaneous administration in relation to DKA. The first thing that I'll go through is by Umpierrez and colleagues,  which was published in 2004, which is a single center randomized controlled trial comprising 45 uncomplicated DKA patients. They were set to receive Q1-hour subcutaneous insulin  as compared to Q2-hour as compared to regular IV insulin, which is our standard of care within the ICU setting. Overall, authors in this study found  that the hospital length of stay was significantly reduced with subcutaneous insulin administration as compared to IV regular insulin. Additionally, there was no differences  in therapy duration until DKA resolved, as well as the amount of insulin that was used in total. Authors in this study overall concluded  that subcutaneous insulin was safe and effective in these patient populations. The next study is characterized by Mohamed and colleagues, which was a retrospective cohort study  on 380 patients who were admitted to the ICU for DKA. They were looking at any associations between time to insulin glargine administration and IV insulin in general.  And what they found is that for every six-hour delay to insulin glargine administration, there was a 26-minute increase in time to DKA resolve,  as well as a 3.2-hour increase in insulin infusion duration and comprised of 6.5-hour increase in ICU length of stay. So authors in this study ultimately concluded  that early administration of insulin glargine subcutaneously is safe and associated with better outcomes.  So this brings me to my first audience polling question for you guys is, does your institution have a subcutaneous insulin protocol for DKA management? And select yes or no here.  Perfect. This is exactly what I was sort of expecting. Here at Ohio Health Riverside,  we do not have a subcutaneous protocol for DKA management, but I know of some other institutions  that do. Moving on to our next audience polling question is whether or not your institution utilizes subcutaneous insulin in addition to the IV insulin for DKA management. And  this could be through the addition of basal insulin to IV management. So select a yes or no here as well.  Perfect. Thank you for your responses. This is also what I was expecting in that in my practice I don't usually see subcutaneous insulin management in conjunction with IV  insulin as well for DKA. So this will bring us to our study in question. Overall, the study purpose and objective of our study was to assess outcomes post a subcutaneous insulin  protocol for DKA management. Authors in this study overall hypothesized that there would be no differences in outcomes between the intervention group, those receiving subcutaneous  insulin, as compared to the standard of care practice receiving IV insulin. So looking at the study design, this was a retrospective, multi-center, pre- and post-implementation  cohort study across many Kaiser Permanente Northern California institutions. They were comprised of an intervention site as well as a standard of care site. At the intervention  site, patients were receiving subcutaneous insulin, the vast majority, as well as in  the standard of care site, patients were receiving IV insulin here. There was a pre- and post-implementation  as well. The pre-implementation occurred from January 1st of 2010 through December 31st of 2015, where the post-implementation phase occurred January 1st of 2017 through December  31st of 2019. At the intervention site, those receiving subcutaneous insulin were included if they were of greater than or equal to 18 years of age and meeting ICD-9 and 10 code  criteria for DKA. Of note, this would include mild, moderate, and severe DKA. They excluded  patients who were pregnant, had a GCS less than eight, and those who required other ICU needs.  In regards to the standard of care sites, patients were included if they were of adult age  and meeting ICD-9 and 10 code criteria for DKA. So in looking at the subcutaneous insulin protocol  that author used in this study, this is a very busy slide, but I can break it down in between  the different phases that patients were in. So phase one occurred in the emergency department  where patients were stratified to receive a lactated ringer bolus followed by an infusion  for a total of five liters. Dextrose-containing fluids were accompanied to this lactated ringer infusion and dependent on potassium level. Long-acting insulin was administered via  insulin glargine at a weight-based dose or the patient's home dose if they were previously on  insulin glargine. Rapid-acting insulin was again a weight-based dose, as you can see there,  where repeat doses were provided within four hours if the glucose was still above 250 milligrams per  deciliter. Phase two happened when the patient was admitted to the hospital in a medical ward  and receiving their care there. The initial fluids, dextrose-containing fluids, are pretty  much the same as the ED floors and the long-acting insulin was continued from the previous dose or  at the patient's own home dose. Rapid-acting insulin was again via insulin Lispro at a  dose-reduced weight-based dose until the glucose was less than 250. When the glucose value was less  than 250, providers were consulted for a sliding scale administration. Discharge management was  governed by the type of diabetes patient had. So in type 1 diabetics, endocrinology was consulted for further care, and in type 2, the primary care physician was referred.  As far as outcomes, authors sought to characterize the time to blood glucose less than 250, anion gap closure, hypoglycemic events, 30-day hospital readmissions,  as well as hospital length of stay, 30-day mortality, and overall ICU admission.  So when looking at the statistics, they included an unadjusted pre- and post-implementation comparisons using the following statistical tests. They concluded alpha at p-value less  than 0.05, and overall, in bold here, a power analysis was not completed for the study. And  we can talk a little bit about that in the limitation section. So in looking at the baseline  characteristics of our patient population, there's a few things that I wanted you to focus on.  Overall, our patient age was predominantly in the 30s to 40s age. Regarding the diabetic diagnosis,  most patients included in the study were type 1 diabetics. And then looking at the DKA severity,  most of our patients were seen to be either moderate or severe DKA. And then looking at  the first serum blood glucose and the first anion gap values, as well as the first arterial pH, there's no differences between the intervention site or standard of care site,  as well as the pre- and post-implementation groups.  Looking at the protocol-based results, we can conclude that prior to the implementation of the subcutaneous insulin protocol, there was very little use of subcutaneous insulin  as the first treatment route. As compared to when this protocol was initiated, there was a very large increase in usage of subcutaneous insulin as the first treatment route.  When we look at the observed clinical outcomes in time to glucose less than 250, time to anion gap closure, hypoglycemic events, and even rebound hyperglycemia,  we can see that there's no differences between the intervention site, standard of care sites, as well as the pre- and post-implementation groups.  And then when looking at those results that were statistically analyzed, I want to draw your attention to those outlined in red, which are ICU admission  via direct ICU admission, as well as 30-day hospital admissions. From an ICU admission standpoint, we can find that those in the pre-implementation phase had a much higher  rate of ICU admission as compared to those in the post-implementation phase who were receiving subcutaneous insulin, with a statistical significance of a p-value less than 0.001.  The same can also be said for 30-day hospital readmission rates, where patients had much lower 30-day hospital readmission rates post-implementation and receipt of subcutaneous  insulin protocols. And when looking at these results a little bit further, based upon an adjusted rate ratio of outcomes between intervention and standard of care, as well as  pre- and post-implementations, authors ultimately concluded that there was a relative 57% reduction  in ICU admissions during DKA when using a subcutaneous insulin protocol, and overall 50%  reductions in readmissions within 30 days of DKA. As such, authors ultimately made the following  conclusions from the study, that the use of a subcutaneous insulin DKA protocol appeared safe and moderate to severe DKA. Additionally, authors concluded that a subcutaneous insulin  protocol can be used effectively and moderate to severe DKA, which hasn't been published previously.  And then authors overall concluded that prospective studies were needed to validate these results further. So in evaluating some of the strengths and limitations of this study,  the strengths included a very large sample size, appropriate statistical tests were used,  and the biggest thing is that this study is overall reproducible with a published subcutaneous insulin protocol that these authors have. Some of the limitations that I concluded,  the first one being that a power analysis was not completed, and as such, any association of clinical outcomes like mortality are kind of limited in this setting. Comparison groups  and trial design are another limitation of this study. So they were not well-matched, and comparisons were made based on a pre versus post implementation analysis,  whereas a simplified approach could have just been subcutaneous insulin administration as compared to IV insulin administration. And then again, being a single institution site,  there's always going to be limited external validity when validating these protocols at other large hospital institutions as well. So the takeaway for practice changes in future  directions. I believe that overall there is significant opportunity in the right patient cohort, and this is going to be related to medication safety as well. So in the severe  diabetic ketoacidosis patient receiving vasopressors, we know that some of these types of patient populations are overall not the ideal population to administer insulin in via  the subcutaneous route, but there are a few patient populations that I think would be an  ideal candidate to receive a subcutaneous insulin protocol for DKA. And this includes a potential type 1 diabetic with insulin noncompliance precipitating DKA, maybe an insulin pump  failure resulting in DKA, or maybe even that COVID-19 patient who has DKA and we want to  lab consolidate rather than Q1 checks on an IV insulin protocol. In relation to medication safety,  there should always be proper implementations when focusing on an EHR protocol. So referring  to IS and P's safe insulin use practices, as well as rigorous patient inclusion and exclusion  criteria, analyzing some of those properties that may affect subcutaneous insulin administration.  From a future direction standpoint, again, it would be nice to see a study that included a  power analysis with an appropriate sample size, focusing on single comparator groups of the  subcutaneous versus an IV insulin group alone. And then also from a medication safety standpoint,  looking at reported medication errors and overall cost analysis between those two groups.  So this takes me to my references, and I can certainly take any questions that you guys may have. Thank you so much. A hypothesis or an explanation for why the decreased readmission  rates that they saw in the study from using the subcutaneous or the post-implementation phase of  the study? Yeah, thanks for the question, Kelly. I think that's a good call out. I think due to the  multiple phases and having that final phase of discharge recommendations provided by endocrinology  or the patient's primary care provider made follow-up a little bit easier for these patients.  So I think that may have led to some of the decreased readmissions that we saw with those groups.  Okay, well, that will conclude our Q&A session. Thank you, Michael.  Thank you.  Now I'd like to introduce our final presenter, Claire Sizz.  Thank you for the introduction and thank you to my co-presenters, Michael and Tyler, both very interesting articles. So I'll be wrapping up today's Journal Club webinar  with a study titled, Direct Oral Anticoagulants Versus Warfarin in the Treatment of 3-Row Venous Thrombosis.  I want to start off with some brief background information regarding 3-Row Venous Thrombosis, or CVT, which is how I'll refer to it from this point forward in the presentation.  CVT, we know, is when a clot forms in the venous sinuses of the brain, which are pictured here in this diagram for your reference. In general, CVT is a type of stroke  that more commonly affects a younger subset of populations and it is associated with a relatively good neurological outcome in the majority of cases.  But there are several complications of CVT that we may be concerned about and have to keep in mind. These include hemorrhagic transformation, development of pulmonary embolism,  dural AV fistula formation, seizures, even progressing to status and coma. There aren't full-fledged guidelines available  for this patient population, but the AHA and ASA have expert opinion recommendations. And the recommendations specifically pertaining to anticoagulation are shown here for you.  They recommend anticoagulation in the acute setting, even in the presence of a pre-treatment intracranial hemorrhage. And as far as longer-term anticoagulation,  they do recommend to initiate maintenance anticoagulation  to prevent the recurrence, as well as the development of other thrombotic complications with specifically a vitamin K antagonist. And the duration varies depending on the patient's  risk factors at baseline, as well as if the event was provoked or unprovoked. So similar to how we would approach other VTE events duration.  And the minimum duration that they do recommend is three months.  They do not make a recommendation regarding the use of DOACs. However, this has been an area of interest in more recent years. And some people have extrapolated other DOAC data  for use in this patient population.  There are several observational smaller studies available, but I did want to highlight this one study shown here. It's a study from 2019 titled RESPECT-CBT.  This was a prospective randomized trial and included patients with CBT that was confirmed on imaging. And they're randomized to Dabigatran 150 milligrams twice daily or Warfarin,  which was dosed for an INR goal of two to three. Their primary outcome was a composite outcome looking at major bleeding or VTE.  And they also looked at a number of secondary outcomes listed there for you.  They found that the primary outcome was met in 1.7% of the Dabigatran patients versus 3.3% of the Warfarin patients. Just to note though,  that there was no VTE events overall in either group. So this composite endpoint really just turned into major bleeding outcome with the absence of VTE overall.  Conclusion from this study was that overall, the risk of recurrent VTE in patients with CBT who are initiated on anticoagulation is very low, as was the risk of major bleeding.  They were unable to demonstrate non-inferiority or superiority due to their limited sample size, but offered that Dabigatran may be a safe alternative  to Warfarin in this patient population.  Now getting into our study of interest today, ACTION-CBT.  The objectives of the study was to compare the safety and efficacy of direct oral anticoagulants or DOACs to Warfarin in patients with CBT.  Getting into some of the methods. So this was a multi-center and international retrospective observational study. It took place from January 2015 through December of 2020.  They included patients with CBT that was confirmed on imaging and excluded patients not being treated with oral anticoagulation and those patients  in which a specific anticoagulant would be preferred. So an example of this would be in patients with antiphospholipid syndrome, which we know Warfarin would be the preferred agent.  They looked at a variety of outcomes, including their primary outcome as well as some imaging and safety outcomes. So their primary outcome was recurrent venous thrombosis,  which included both VTE or CBT during their follow-up duration.  They also assessed recanalization status as their imaging outcome, and they assessed it as either complete, partial, or no recanalization.  There were additionally a variety of safety outcomes and this included major hemorrhage, which was defined as intracranial hemorrhage or major extracranial hemorrhage.  They looked at symptomatic intracranial hemorrhage  as well as any death during follow-up.  A variety of statistical tests were used for the various outcomes, including PTAS, Chi-squared, Fisher's exact test, and Wilcoxon-Rank sum test.  They did carry out propensity score matching to test associations between the group outcomes given differences seen between the treatment groups at baseline.  They also had two additional adjustment models that they applied, which was an adjustment for some pre-specified variables as well as an adjustment for differences observed  in the baseline characteristics of the two populations.  Now moving on to the results of the study. So listed here are the populations seen. So we have, for the efficacy and safety analysis,  a total of 845 patients broken down by what agent they used. So 279 used a DOAC and 438 used Warfarin. And there was a group of 128 patients  that did use both at some point during follow-up.  They also had the recanalization rate analysis, which was a somewhat smaller population because they excluded patients who had endovascular treatment  or who had used both a DOAC or a Warfarin during follow-up. So this population had just under 180 DOAC patients and 346 patients using Warfarin.  Baseline characteristics from the two groups can be seen here in this table. Some things I want to highlight are outlined in red.  So the DOAC group did have higher rates of VTE at baseline. And then the Warfarin group was more often to contain active smokers, although this was not statistically significant,  but just important to note given the known increased risk of thrombosis seen with smoking.  Another thing to note is the higher likelihood of at least one positive antiphospholipid antibody in the Warfarin group. But again, as I mentioned,  patients with antiphospholipid syndrome were excluded.  Finally, just want to point out the median duration of anticoagulation for both groups was beyond the recommended minimum duration stated in those recommendations earlier  of a minimum duration of three months. So that was surpassed for both groups.  This figure shows the breakdown of which DOACs were used in our DOAC group. So Apixaban was by far the most commonly used agent, just under 70% of patients using Apixaban,  and then followed by Dabigatran and Rivaroxaban, and then a very small percentage using either other or multiple DOACs. Unfortunately, the information on specific dosing strategy  of these agents was not provided.  Here we have the different efficacy, imaging, and safety outcomes, as I mentioned before. And we have them for both the unadjusted analysis, one of the weighted models,  as well as the propensity match model.  As you can see for the primary outcome  of recurrent venous thrombosis, there was no difference between groups in any of the analyses or in the unadjusted column. This remains true for death,  as well as partial or complete recanalization rate.  What I do wanna highlight is the hazard ratios for our safety outcome of major hemorrhage. As shown in the table, the rates of major hemorrhage were marginally lower  in the DOAC group versus Warfarin, and this did become statistically significant in the weighted model shown in the middle, where you can see a hazard ratio of 0.34,  with a 95% confidence interval, 0.15 to 0.80.  I also want to provide for you the Kaplan-Meier survival curve. So shown here is the survival curve for survival without recurrent DTE,  with the Warfarin group shown by the blue line and the DOAC group by the red line.  And here is the Kaplan-Meier curve for survival without major hemorrhage. Again, Warfarin group being represented by the blue line and DOAC being represented by the red line.  So you can see kind of the difference here in survival without major hemorrhage with the DOAC having less hemorrhage.  Getting into the conclusions by the authors. So they went on to conclude that DOACs provide a reasonable alternative to Warfarin in patients with CVT. They do note, however,  that there needs to be confirmation by larger prospective studies. Of note, there are two studies currently underway, which are listed here for you.  DOAC CVT is looking at DOACs versus Warfarin and is predicted to end at some point in 2024. And then secret is Rivaroxaban versus Warfarin  and predicted to conclude at the end of 2022. So if interested in more of the studies of that design and enrollment, the QR codes will bring you to their clinicaltrials.gov page.  Next, I want to touch on some of the strengths and limitations from this study. So starting off with the strengths. The data shown in this trial is pretty consistent  with the non-inferior efficacy as well as favorable safety profile that we've seen with DOACs in other patient populations, such as atrial fibrillation or other VTE,  such as PE or DBT.  This trial included multiple centers and was international, so this provides us with some more real-world data that has good external validity.  Additionally, the multiple adjustments that they took with the weighted models and propensity matching produces results with real clinical relevance.  And finally, a good majority of the population did have adequate follow-up with at least 90 days. This is important because we know that the highest risk of re-thrombosis  or propagation of initial thrombosis is in the more acute phase following the initial insult.  Next, for limitations. Although one of the major things highlighted by the authors in this study was the reduction in major hemorrhage seen with DOACs,  this has not been reproduced with other studies in this specific population, so I would just personally interpret this as caution  until there's more data available to confirm this finding. Additionally, the ideal DOAC of choice as well as what dosing strategy to use remains fairly unknown,  but again, could potentially be extrapolated from other VTE populations.  Similarly, there was a lack of INR values for the warfarin group as well as relevant drug-drug interactions that may have been present  and contributing to the patient's anticoagulation status. And finally, we have limited information regarding the use of the INR  and finally, we have limited information regarding adherence or how they monitor adherence for these patients.  So, wrapping up with my take-home points for the article, overall, there's really a limited amount of data to pinpoint the ideal anticoagulation agent for this patient population.  However, data favoring DOACs for other VTE indications such as CVT and PE have been extrapolated to the CVT population and as it's a similar pathophysiologic concept.  Overall, I believe unless there's a strong indication for patients to be receiving warfarin, DOACs do provide a reasonable alternative  with a potentially safer adverse effect profile for patients with CVT.  And I'll be wrapping up the presentation with a few polling questions for the audience.  So, the first question is, in your current practice, which anticoagulation strategy do you recommend for patients with CVT? Do you recommend warfarin, a DOAC,  or you have not encountered a patient with CVT?  All right. So, it looks like a majority of patients do currently recommend DOACs, but a close second would be Warfarin.  So, I think this is pretty similar to what I've seen with kind of half and half almost.  And our second polling question is, now, taking into consideration the results of this study  as well as weighing the risk of benefit versus harm, would you recommend a DOAC for anticoagulation in a patient with CBT? Either yes, no, or you're awaiting some more evidence.  So, it looks like the majority of patients, or sorry, the majority of audience members would currently recommend a DOAC for practice in a patient with CBT, and this is also pretty  consistent with discussions that I've had with some of the preceptors and practicing  So, that concludes my presentation, and I'm happy to take any questions at this time. Thank you, Claire.  If there are any questions from the audience members, you may type those into the question box at this time.  Okay, with no questions, that will conclude our Q and a session for Claire. Thank you very much.  So, in conclusion, we wanted to thank you to our presenters today and also to the audience members for attending.  Please join us on the third Friday of the month from two to three PM Eastern time for the next journal club spotlight on pharmacy. And that will conclude our presentations today. Thank you.  Thank you, Kelly.

Video Summary

The Journal Club Spotlight on Pharmacy webcast discussed three articles related to critical care pharmacy. The first article focused on the hemodynamic effects of ketamine compared to propofol or dexmedetomidine as continuous ICU sedation. The study found that ketamine was associated with less hypotension and bradycardia compared to the other sedative agents. The second article examined the use of subcutaneous insulin protocol for diabetic ketoacidosis (DKA) management. The study found that the use of subcutaneous insulin was safe and associated with better outcomes, including a reduction in ICU admissions and 30-day readmissions. The third article compared the use of direct oral anticoagulants (DOACs) to warfarin in the treatment of cerebral venous thrombosis (CVT) . The study found that DOACs were a reasonable alternative to warfarin in patients with CVT, with comparable efficacy and a potentially lower risk of major hemorrhage. Overall, the findings of these studies provide valuable insights for critical care pharmacists and suggest potential changes in practice for sedation and anticoagulation management.

Asset Subtitle

Pharmacology, Neuroscience, 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.

Tyler Brouse, PharmD
Atchley E, Tesoro E, Meyer R, Bauer A, Pulver M, Benken S. Hemodynamic effects of ketamine compared with propofol or dexmedetomidine as continuous ICU sedation. Ann Pharmacother. 2022 Jul;56(7):764-772.

Michael Young, PharmD, BCPS
Rao P, Jiang SF, Kipnis, et al. Evaluation of outcomes following hospital-wide implementation of a subcutaneous insulin protocol for diabetic ketoacidosis. JAMA Netw Open. 2022 Apr 1;5(4):e226417.

Clare Cycz, PharmD
Yaghi S, Shu L, Bakradze E, et al. Direct oral anticoagulants versus warfarin in the treatment of cerebral venous thrombosis (ACTION-CVT): a multicenter international study. Stroke. 2022 Mar;53(3):728-738.

Follow the conversation at #SCCMCPPJC."

Meta Tag

Content Type Webcast

Knowledge Area Pharmacology

Knowledge Area Neuroscience

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Tag Analgesia and Sedation

Tag Diabetes

Tag Anticoagulation

Year 2022

Keywords

critical care pharmacy

ketamine

propofol

dexmedetomidine

subcutaneous insulin protocol

diabetic ketoacidosis

DOACs

warfarin

cerebral venous thrombosis

	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.