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 Kaitlin Brown. I'm a Neurocritical Care and Emergency Medicine Clinical Pharmacist at Mayo Clinic in Rochester, Minnesota, and I will be moderating today's webcast. A recording of this webcast will be available to registered attendees. You can log in to mysccm.org and navigate to the My Learning tab to access the recording. Thanks 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 questions throughout the presentation, type into the question box located on your control panel. You will also have the opportunity to participate in several interactive polls. When you see a poll, simply click the bubble next to your choice. SCCM provides the following disclaimer. This presentation is for educational purposes only. The material presented is intended to represent an approach, view, statement, or opinion of the presenter, which may be helpful to others. The views and opinions expressed herein are those of the presenters and do not necessarily reflect the opinions or views of SCCM. SCCM does not recommend or endorse any specific test, physician, product, procedure, opinion, or other information that may be mentioned. And now I'd like to introduce your speakers for today. Each will give a 15-minute presentation followed by a Q&A session. Our first presenter today is Katie Dalton, a PGY-2 critical care resident at the University of Illinois Hospital and Health Sciences System in Chicago, Illinois. Our second presenter is Traeger Hintze, a PGY-2 critical care resident at University Health System in San Antonio, Texas. And our third speaker is Erin Knopp, a PGY-2 critical care resident at Yale New Haven Hospital in New Haven, Connecticut. And now I'll turn it over to our first presenter. Good afternoon, everyone, and thank you for the opportunity to speak with you all today. As Kaitlin said, my name is Katie Dalton, and I am the PGY-2 critical care pharmacy resident at UIC. The journal article I will be discussing today is by Gandolfi and colleagues, and it's titled The Effects of Melatonin Supplementation on Sleep Quality and Assessment of the Serum Melatonin in ICU Patients. This was a randomized controlled trial published in Critical Care Medicine Online Ahead of The Circadian Rhythm is a 24-hour internal clock that regulates our sleep-wake cycle by responding to changes in light in our environment. As you all are well aware, the ICU is not an ideal place to get restful sleep due to constant exposure to annoying alarms, irregular lights, lack of procedures, and discomfort from mechanical ventilation. Usually ill patients can also have sleep disturbances due to pain, stress, and anxiety. While we can utilize analgesics and sedatives to make our patients more comfortable and to reduce pain and promote sleep, it's important to keep in mind that sedation has been associated with abnormal EEG patterns that are not absorbed in normal sleep. Additionally, opioids and benzodiazepines have been shown to disrupt rapid eye movement or REM sleep, which is our deep restorative sleep. Melatonin is a natural hormone with hypnotic, antioxidant, and immunomodulating properties. Critically ill patients have reduced peak melatonin levels, and this has been correlated with delirium and poor clinical outcomes. Melatonin dosing in ICU patients has varied from 3 to 30 milligrams, and therefore the optimal dose is currently unknown. Literature evaluating melatonin has been shown to have mixed results. Most of the literature evaluates melatonin's effects on delirium, and more recently, there have been newer studies coming out assessing sleep quality and quantity. The pre-randomized control trials that have been completed are very small, and additionally, there are several issues with the quality of the studies, as the sleep length and quality have been endpoints. However, there are heterogeneous methodologies in terms of the sleep analysis. Oren and colleagues looked at quality of sleep by FIS monitoring in 24 patients that had respiratory failure. They used 10 milligrams of a dose of melatonin in this trial, and patients had a 7% decrease in the fifth AUC, which suggested better sleep in the melatonin group. Ibrahim and colleagues assessed the length of sleep by a bedside nurse, and it averaged about four hours per night, and they utilized a 3 milligram dosing strategy in this trial, and no differences were seen between the lengths of sleep. Shiloh and colleagues conducted a randomized control trial in eight patients with DOPD to assess sleep duration, and the duration was about six hours, and there was no difference between patients. When the 2013 TAD guidelines were updated in 2018, immobility and sleep were highlighted as priorities in intensive care. However, current TAD guidelines make no recommendation on the use of melatonin to improve sleep in the ICU due to a low quality of evidence. Before I discuss this trial, please respond to the poll about current practice at your institution. Do you routinely recommend melatonin or remeltion in the ICU? Please select A, if yes, you use melatonin, B, if yes, you use remeltion, or C, no melatonin or remeltion. All right, so it looks like about half are recommending melatonin. The effects of melatonin supplementation on sleep quality and assessment of serum melatonin in ICU patients was a multi-center, double-blind, randomized placebo-controlled trial that was conducted between March 2016 and June 2019. Patients were admitted to the ICU at two centers in Brazil and were randomized in a one-to-one session to receive melatonin 10 milligrams for seven consecutive days or placebo. If patients were transferred out of the ICU, they were able to continue on melatonin. Adults that were greater than 18 years of age and spent greater than or equal to one night in the ICU were eligible for enrollment. Patients that were excluded from the trial were unable to answer the questionnaire, had a seizure history, neural or psych illnesses, and renal or hepatic impairment. There were no significant differences in baseline characteristics, and overall, the patient population was 57% male, 60 years old, had STAT-3 scores of 43, had a SOFA score of one on admission, which was zero by day two of the study. Patients were 39% surgical patients that were post-op, 33% had cardiovascular disease, 13% had cancer, and only 5% of patients had sepsis. The authors were expecting to see a 20% difference in good sleep quality between groups and aimed to recruit 200 patients. The statistical analysis was by intention to treat with significance set at 0.05. The study primarily aimed to evaluate whether the use of exogenous melatonin in critically ill patients improved sleep quality. The primary outcome was assessed utilizing the Richards-Campbell Sleep Questionnaire, or the RCSQ, which is a validated tool to assess sleep quality. It is composed of five subcategories, including sleep depth, sleep latency, or the time to fall asleep, awakening, returning to sleep, and sleep quality. I would like to point out here that although the RCSQ score in total evaluates sleep quality, it is also an individual component in the scoring tool. Each subcategory is assessed on a 100-point visual analog scale, and the categories are added together and divided by 5 to get a total score from 0 to 100, with 0 indicating the worst possible sleep you've ever had and 100 indicating the best possible sleep you've ever had. The RCSQ tool can be patient-reported or are unassessed at the bedside. The study secondarily aimed to assess whether the use of melatonin decreases the prevalence of delirium, pain, and anxiety, the duration of mechanical ventilation, the length of ICU and hospital length of stay, doses of sedatives and analgesics, and adverse reactions. RCSQ scores are broken down into four categories, as you can see on the bottom left of the screen. Scores of 1 through 25 indicate very poor sleep, while 26 through 50 indicate poor sleep, 51 through 75 indicate good sleep, and 76 through 100 indicate very good sleep. Since patients only had to be in the ICU for greater than or equal to one night and sleep differed significantly in the ICU versus that on a general medicine score, data analysis was performed by separating ICU nights and ward nights. The benefit of melatonin was not observed in the ward as it was in the ICU as none of the components were statistically significant and neither was the total score in ward patients. The total score of RCSQ in ICU patients was 70 versus 61, and this did reach statistical significance, although both of these scores would put the patients into the good sleep category. Additionally, the only RCSQ subcategory that was statistically significant was sleep depth. All of the others, including returning to sleep or sleep quality, latency, and awakenings were all not statistically significant, so this total score of the primary endpoint is driven by sleep depth. The table at the bottom represents which category of sleep each patient fell into. While there were no differences in the poor sleep, good sleep, or very good sleep categories, about 15% of patients in the placebo group reported very poor sleep compared to about 3% of patients in the melatonin group. All secondary outcomes displayed on this slide are in melatonin versus placebo, and none of the secondary outcomes were statistically significant in this trial. Overall sleep in the ICU was assessed by bedside RN and was roughly six and a half hours in each group. Delirium was assessed utilizing the intensive care delirium screening checklist and had low rates overall compared to other studies. Patients spent on average about three nights in the ICU, were ventilated for about three days, spent 10 days in the hospital, and had about a 10% mortality rate. No differences were seen in adverse events, or no differences were seen in pain, anxiety, or doses of sedatives and analgesics. The investigators also obtained melatonin plasma concentrations in the trial on days one through three in nine patients. There were six patients in the melatonin group and three in the placebo group, and at 2 a.m. melatonin concentrations were 150 versus 32.5 picograms per milliliter. For your reference, in a study of about 130 healthy volunteers, melatonin concentrations were about 80 at 3 a.m. and 30 at 8 a.m. The author's conclusion of this trial was that melatonin was associated with better sleep quality, which suggests its possible role in the routine care of critically ill patients in the future. Some strengths of this trial is that it was very well designed. A large number of patients were enrolled compared to previous melatonin randomized control trials, which only evaluated about 60 patients. Patient cohorts were also well matched, as there were no differences in baseline characteristics. Also thought it was reassuring that the investigators obtained melatonin levels as patients can have poor absorption and concentrations can be significantly decreased by first-pass metabolism. There were also several limitations in this trial. First, while melatonin is relatively inexpensive, it is an over-the-counter medication that is not regulated by the FDA. This has led to concerns for quality and consistency in products, which has led to many hospitals being hesitant to add melatonin onto their formulae. Another limitation in this trial is that nurses assessed sleep duration at the bedside, which is extremely subjective. RCSQ scores were never compared to an objective sleep measurement method, such as polysomnography, otherwise known as a sleep study. This trial would have been strengthened by comparing RCSQ values to a sleep study, although this is cumbersome and expensive. Finally, the acuity of patients enrolled in this trial were not our thickest ICU patients, as evident by the SOFA scores of 1 at enrollment and 0 by day 2 of the trial. Patients also had a short ICU length of stay, so it may not have been able to capture the circadian rhythm derangements to see the beneficial effects of melatonin by that point. There also was a very low incidence of delirium in this trial, so any results regarding that would be inconclusive. While the authors concluded that melatonin is associated with better sleep quality, the RCSQ score was driven by sleep depth and not sleep quality, as patients did not report an improvement in sleep quality that reached statistical significance. There was a study done by MICU RNs and patients that compared the RCSQ scores between patient reported versus nursing assessments, and they found that nurses overestimated patients' perceived quality of sleep in the ICU. So, the RN assessing the duration of sleep may not be as accurate and may be a little overestimated. Additionally, this study was underpowered, as the authors were expecting to see a 20% difference in the number of patients that experienced good sleep quality, but they only observed a difference of 2%. Also, the trial had enrolled 203 patients, but 11 patients, 6 in the melatonin group and 5 in the placebo, were unable to answer the questionnaire for reasons that authors did not report. Additionally, 8 patients had dropped out of the trial for having poor sleep. This was 7 patients in the placebo group and 1 in the melatonin group. The RCSQ score of 69 versus 60 is likely not clinically significant, as both are in the category of good sleep. While it did reach statistical significance, I personally don't believe this is clinically significant. There's also the possibility of having low external validity, as patients had low acuity, and although the rates of delirium were a secondary endpoint, it would be difficult to come to any conclusions based on this trial about melatonin's effects on delirium. Critically ill ICU patients experience sleep disturbances, circadian rhythm disorders, which can lead to increased delirium, and poor outcomes. Based on this study, I would not routinely recommend the use of melatonin 10 mg in critically ill patients. Although patients had supra-therapeutic melatonin concentrations at 2 am in this trial, it was reassuring that there were no significant adverse effects from melatonin. I would possibly consider utilizing a lower dose in critically ill ICU patients. Overall, I think the benefit of melatonin may outweigh the risk and the harm, since this drug is well-tolerated and relatively inexpensive, but I don't believe this study would be practice-changing. For future directions of evaluating the use of melatonin in ICU patients, I would like to see patients enrolled in the trials that have a higher severity of illness and the trials to evaluate patient-centered quality outcomes. Additionally, it would be better if trials utilized these objective sleep assessments, however they are very expensive and difficult to interpret. As pharmacists, there are a couple things we can do to help promote sleep in our ICU patients, one of them being reviewing chronic medications that patients were on prior to coming to the ICU to avoid any withdrawal syndromes, and to utilize opioid incentives at their minimal effective dose. And that brings us to our final learning question. What dose of melatonin do you recommend in critically ill ICU patients? Do you recommend 10 milligrams, 5 milligrams, 3 milligrams, a different strength, or you don't routinely use melatonin in the ICU? All right, we've got a wide range here. It looks like some are recommending the 10 milligrams, some are recommending the 5, the 3, and then some don't use it. And that concludes my presentation, and I would be happy to answer any questions. Thank you, Katie. The first question we have is, did the authors report any non-pharmacologic sleep interventions used, for example, reducing the frequency of nursing assessments at night? They did not. That was one thing in this trial that they did not control for, so they were allowed to go in the room at night and do procedures or lab draws or whatever they had to do for the patients. None of the non-pharmacological aspects were controlled in this trial. Thank you. The next question, did they report medications commonly used for agitation that caused sedation like Seroquil? They did not report, or I'm sorry, yes, they did. There were no differences in the amount of catepin or haloperidol administered between the two groups. Thank you. The third question we have here, common issues with sleep-inducing meds, for example, Lopidem, are over sedation and fall risk the following morning. Were the high concentrations of melatonin the following morning associated with a hangover effect? They did. I think that's a great question, and that was one of the things that I was looking for as well in the trial. The authors did report that patients did not experience any drowsiness the following day, and the concentrations did decrease from the 2 a.m., the 150 concentration, down to normal physiologic levels by the morning. And then our final question before we go on, has the RCSQ been validated by both nursing assessment as well as patient-reported assessment? And with that, do you know the percent breakup between RN assessments and patient-reported scores? Sorry, can you repeat that, the percent? Yeah, so it's a two-part question. Let me ask you the first part first. Has the RCSQ been validated by both nursing assessment as well as patient-reported assessment? I guess big picture, has it been validated at all? So it was validated for patient-reported. I'm not entirely sure about the validation in the RN-reported, but there was that study that showed that there's significant variability between nursing assessment and patient assessment, and the nurses kind of overestimated the perception of sleep in patients. So they were reporting that patients were getting better sleep than the patients were reporting themselves. And then with that, in this particular study, do you know what percentage of the patients had RN assessment versus patient-reported scores? So in that particular study, it was 100% because they were looking at the inter-rater reliability of the scoring system. Sorry, what about in the specific study you just presented on? By Gandolfi and colleagues? Sorry, so you were asking which percent was patients versus nursing? Yeah. So the study didn't explicitly say, but I did say that the scoring tool was administered by psychologists. However, I'm inferring from the trial that it was patient-reported because they excluded patients that weren't able to answer the questionnaire. And then they did report in the trial that 11 patients were unable to answer it. But they didn't specifically say that it was patient-reported or nurse-reported. Duration of sleep was assessed by the bedside nurse. And then I believe the score was reported by patients answering the questionnaire and not the nursing. Sounds good. Thank you, Katie. So this concludes our Q&A session. And now I'd like to introduce our second speaker, Traeger Hintze. Good afternoon, everybody. It's great to be with you today. As Caitlin mentioned, my name is Traeger Hintze. I'm a PGY-2 critical care resident at University Hospital in San Antonio. And today I'm going to be reviewing the study tranexamic acid during prehospital transport in patients at risk for hemorrhage after injury. This was published in JAMA Surgery. It's e-published ahead of print. But the title that you'll probably remember a little bit better is the STAMP trial, which stands for the Study of Tranexamic Acid During Air, Medical, and Ground Prehospital Transport. I'm going to refer to tranexamic acid as TXA throughout the presentation to avoid being tongue-tied as much as I can. But the first thing I wanted to start with was just a little bit of background information. The guideline evidence for prehospital administration of TXA has been extrapolated from studies that have shown a benefit in early in-hospital administration of the medication. However, high-level evidence demonstrating prehospital safety and efficacy of TXA is lacking, and thus information from this study could provide some benefit to us for this patient population. Just as a brief refresher, TXA forms a reversible complex with plasminogen via the lysine receptor. This prevents plasmin from binding and stabilizing the fibrin matrix and ultimately inhibits fibrinolysis. The most important adverse effect of TXA to consider is thrombosis, but the incidence when it's used for massive transfusion per previously published studies is pretty low. And we're actually going to look at a few of those studies right now just as kind of some background. The first one I wanted to cover was the CRASH-2 study that was published in 2010. It included over 20,000 patients and enrolled them to receive TXA within eight hours of presentation. The authors from this study found a decrease in 28-day mortality when patients received TXA within the allotted time frame. And in subgroup analysis, they found a greater reduction when the medication was received in less than three hours. The MATTER study that was published in 2012 was a retrospective military registry study at a single center in southern Afghanistan. Patients in this study received TXA within 24 hours, and they also had to receive one unit of blood as an inclusion criteria. They included almost 900 patients in their analysis of these combat-related injuries and found a decrease in 48-hour and in-hospital mortality in this patient population. And lastly, the CRASH-3 study that was published just last year enrolled nearly 13,000 patients to receive TXA for traumatic brain injury and evidence of bleeding on CT scan. Patients in this trial received TXA within three hours of arrival to the hospital, but the investigators found no difference in 28-day head injury-related death between TXA and placebo. So the study objective of the SAMP trial was to assess the effectiveness and the safety of TXA when it's administered pre-hospital, of course, compared to placebo, and this was in patients at risk for hemorrhage. The design of the study was placebo-controlled, a randomized controlled trial, and it included four U.S. Level I trauma centers. We here at University Hospital were actually one of those trauma centers that were included in the trial. And the primary endpoint that they looked at was 30-day mortality. A couple of the secondary endpoints, specifically as they relate to safety that I wanted to point out, they looked at the rate of pulmonary embolism and DVT in this patient population, and they also looked at the incidence of coagulopathy and hyperfibrinolysis per the INR and the TAG in these patients. I wanted to kind of go through this flowchart a little bit to describe the different groups that they included, because there ended up being four different groups or treatment arms in this particular trial. So they went through pre-hospital randomization to receive either one gram of TXA or placebo. And the placebo group received placebo infusions throughout the remainder of the trial period. But the TXA group was then stratified to receive one of three different treatment options. The first, the study investigators called the abbreviated dose group. This was just a one gram bolus of TXA pre-hospital. There was also a standard dose group, which was TXA one gram pre-hospital, and then a one gram continuous infusion over eight hours once the patient was in the hospital. And the final group that they looked at was what they called a repeat bolus dosing regimen, a one gram infusion of TXA bolus, and then one gram, again, a re-bolus when the patient arrived to the hospital. And then finally, a one gram continuous infusion over eight hours. The statistical analysis of the trial, they utilized the two-sided mantle Hansel for primary outcomes. And the study investigators used the mortality estimate of 16% from the CRASH-2 trial to estimate that they would need to enroll 994 patients to provide sufficient power to detect a 90% difference in this patient population. And for those that were missing 30-day mortality outcomes, multiple imputation was performed. As far as the study population goes, the patients that they chose to include were patients at risk for hemorrhage that could be transferred from the scene or an outside emergency department to a participating site within two hours, and they had to experience at least one episode of hypotension, which was defined as systolic blood pressure less than 90, or tachycardia, which was defined as a heart rate greater than 110. The patients that they chose to exclude from this trial were those likely with a very poor prognosis due to the inciting factor of the injury, and so keeping that in mind with the patients that they chose to exclude from the study. A few of the baseline demographics that I wanted to point out. First was a blunt mechanism of injury. About 85% of the patients in both groups had a blunt mechanism of injury. The pre-hospital heart rate, the median heart rate for these patients was in the high-one-teens, and so keeping that in mind as far as the study population that we're looking at and the patients that they chose to include. And then the injury severity score. In the placebo group, the median was 11, and in the TXA group, the median ISS score was 13. Overall, a lower injury severity score than you might suspect to see in many level-one trauma centers. As far as the primary endpoint goes, they were looking, again, at 30-day mortality, and in the placebo group, there was a 9.9% mortality rate, and in the TXA group, there was an 8.1% mortality rate. This did not reach statistical significance, and the trial was terminated early due to low enrollment and lack of funding. 30-day mortality data was available on 894 patients. They excluded five in the TXA group and four in the placebo group due to that lack of 30-day mortality data. There were no differences in the secondary endpoints that they analyzed, but I did want to point out that there was no difference in the rate of PE and DVT between placebo and TXA between the two groups. In post-hoc subgroup analysis, there were some differences that they noted between TXA and placebo. The investigators found a decrease in 30-day mortality in favor of TXA with the repeat bolus dosing regimen, which, again, was the one-gram bolus pre-hospital, one-gram bolus on arrival to the hospital, and then one-gram infusion over eight hours. Additionally, the patients that received TXA within one hour, there was also a 30-day mortality benefit for these patients as well. And lastly, patients with severe shock defined as a systolic blood pressure less than 70. Remember, the inclusion criteria was less than 90 for this trial. There was also a mortality benefit for these patients as well. The author's conclusion is that there was no difference in 30-day mortality between TXA and placebo when it was administered pre-hospital in patients at risk for hemorrhage. There was also no difference in the risk of thrombotic complications between the groups. And so, ultimately, pre-hospital TXA administration is safe, effective, and provides a survival benefit in specific subgroups. But a few of my thoughts that I had on this particular trial is that the outcomes in the pragmatic design allow for meaningful application to clinical practice. Since about 85% of the patients experienced a blunt mechanism of injury, it may limit the applicability of the study to all trauma patients. But overall, this study was underpowered as they did not reach the estimated patient enrollment. Additionally, the differences they did find in favor of the repeat bolus dosing regimen, the receipt of TXA within one hour of injury, and administration to patients with severe shock are all underpowered. And we have to keep that in mind when we're considering the results. Additionally, utilizing the mortality estimate of 16% and an effect size of 7% from the CRASH-2 trial may have been too high for this particular trial when you consider the design. This study was conducted, as we mentioned, at four U.S. Level I trauma centers, whereas the CRASH-2 trial had 274 hospitals in 40 different countries. So, again, the effect size may have been too large for this particular trial design. Lastly, needing to enroll a number of patients, the inclusion criteria they selected with one reading of systolic blood pressure less than 90 or a heart rate greater than 110 may have selected for a patient population with a lower severity of injury. And this is particularly demonstrated when you look at the median injury severity ISS scores of a median of 11 and 13 between the placebo and TXA groups, respectively. So, keeping some of these limitations in mind, my recommendation is ultimately that it's reasonable to consider pre-hospital TXA in patients at risk for hemorrhage if the patient's systolic blood pressure is less than 70 and they're within one hour of injury. I would consider extending the window out to less than three hours at the time of injury based on the results of the CRASH-2 study and their findings of a mortality benefit in this time frame. I would give strongest consideration to patients with a blunt mechanism of injury. This is the majority of the patients that were studied in this particular trial. And lastly, I would actually choose what they called the standard dosing regimen in this study, which is one gram bolus pre-hospital and then that was followed by a one gram infusion over eight hours. When you look at the trial specifically, the mortality rate between the standard dose and the repeat bolus dosing regimens were 7.8% and 7.3%, respectively. But I do feel like this is an area that could be explored further in other studies. And we could further refine our dosing of TXA to find if one or either of the dosing regimens provide a benefit or if one is safer than the other. So I have a couple of polling questions. Based on the results of the STAMP trial, pre-hospital TXA demonstrated a 30-day mortality benefit in all patient populations. I'll give you about 10 seconds to consider your answer. Okay, it looks like about 93% of the respondents agreed with the, they said no, that there was no difference, or the 30-day mortality benefit was not shown in all patient populations. Okay, and the second polling question I had for you was in post hoc analysis of the STAMP trial, which TSA dosing regimen showed a survival benefit over placebo? I'll give you about 10 seconds to consider your answer. Okay, so it looks like most of us said the standard dosing regimen. It was actually the repeat bolus dosing regimen that they looked at specifically that showed a survival benefit over placebo. There was a difference in the mortality benefit or the mortality rate between the standard dosing regimen, which was 7.8% mortality, and in the repeat bolus dosing regimen, it was a 7.3% mortality rate. So keeping that in mind, what the authors analyzed in this study was the repeat bolus dosing regimen. I, however, would recommend the standard dosing regimen if I were recommending TXA pre-hospital for a patient at risk for hemorrhage. Thank you so much for the opportunity to present, and I would love to answer any questions that you have. Yeah, thank you, Traeger. So I think we have time for maybe one or two questions here. The first one, did the authors discuss the ethical considerations for giving trauma patients placebo rather than TXA? They did not specifically discuss the ethical considerations for giving placebo pre-hospital. They did discuss, however, that then they emphasized the point that it is the mortality benefit has been shown in hospital. There are some retrospective analyses that have been small that have looked at giving TXA pre-hospital that have shown a little bit of benefit, but the high-quality data in the form of a randomized controlled trial did not exist until this point. Thank you. Then our second question, did the study report how many patients received massive blood transfusions in each group, and was there a mortality benefit in those patients? Yeah, that's one thing I really would have loved to see as well in this particular trial is reporting how many patients received massive transfusion, and they ultimately did not report it in this trial, but that is information for me that I would have loved to have, too, to provide a little bit more information and guide therapy and that sort of thing. Okay, and then a final question here, did the authors make any adjustments for multiple comparisons? So, just to clarify, that would be like multivariate type of analyses? Yeah. You know what, they did consider a lot of different variables in this particular study. They did do some multivariate analyses. I don't have it right off the top of my head which ones they performed, but they did do some multivariate analyses in this particular trial. Sounds good. And we'll switch it over to our last and final presenter, Aaron Knopf. Thank you, Traeger. Good afternoon, everyone. Thanks for having me. As Kaylin said, my name's Aaron Knopf. I'm a PGY-2 critical care resident at Yale New Haven Hospital. And this afternoon, let's discuss the recently published Chezikovit trial, comparing therapeutic and prophylactic anticoagulation in COVID patients. For the past 10 months or so, all of us as healthcare providers have been facing the severe acute respiratory syndrome coronavirus 2 pandemic with admirable resiliency. COVID-19 primarily spread from human to human contact has presented a unique and extraordinary challenge uncovering over time, a spectrum of severity and myriad of complications secondary to infection. Acute respiratory distress syndrome resulting in respiratory failure can precipitate shortly after the first signs of dyspnea. Cardiovascular effects, including cardiomyopathy and widespread inflammatory response have been associated with critical and fatal illness. Secondary respiratory infections and bacteremias, pulmonary emboli, the advanced thrombosis and acute strokes have all been reported. And ultimately, recovery appears to be correlated with disease severity, age and pre-existing comorbidities with increasing reports of long-term sequelae such as prolonged respiratory impairment. Today, I'd like to focus on thromboembolic complications, a result of a hypercoagulable state and observed in up to one third of ICU patients. Autopsy evaluation has showed that in addition to macrothrombi, extensive microthrombi and increased angiogenesis, particularly in alveolar capillary circulation may be related to severe endothelial injury as a result of COVID-19. An independent study of patients within a New York City hospital demonstrated that certain factors, including age, sex, primary cardiovascular history and higher D-dimer levels at presentation were associated with a thrombotic event, which further correlated to more than 80% higher mortality. Thus, thromboembolic treatment or prophylaxis provides us a possible target to decrease complications and improve mortality outcomes for these patients. Two retrospective studies evaluating the effect of anticoagulation on COVID-related mortality were able to demonstrate a possible benefit. Tang and colleagues evaluated patients receiving heparin and found parameters such as age and D-dimer correlated to 28-day mortality. However, those with elevated D-dimer or sepsis-induced coagulopathy score and receiving heparin had lower 28-day mortality. Furthermore, Paranjpai and colleagues specifically observed mechanically ventilated patients receiving treatment of heparin and found lower mortality that correlated with increased duration of anticoagulation. Overall, these findings are encouraging in support of a role for anticoagulation in COVID-19 patients. In early January, the World Health Organization provided interim guidance and recommendation of pharmacologic prophylaxis to reduce the incidence of venous thromboembolism in COVID patients with a preference for low molecular weight heparin if available and not contraindicated. Later, the Chinese Thoracic Society and Association of Chest Physicians released a similar recommendation, specifically citing use for critically ill or severe patients, but lacking to elucidate a prophylactic or treatment dose modality. Subsequently, the American College of Cardiology and International Society on Thrombosis and Hemostasis released a similar recommendation. Within the document, they recognized the broad variety of prophylactic, intermediate, and therapeutic doses of anticoagulants used in COVID patients, stating that a majority of the authors feel prophylactic dosing is adequate. However, some did feel intermediate or therapeutic dosing was more reasonable. They ultimately cite that the optimal dosing in patients with severe COVID-19 remains unknown and warrants further prospective investigation. Critically ill COVID-19 patients have hypercoagulability commonly characterized by elevated D-dimer, but can also be observed through slight prothrombin and activated partial thromboclastin time prolongation, increased fibrinogen, and occasionally increased platelet counts. Additionally, von Willebrand factor can also be greatly increased consistent with endothelial injury. Together, these findings and society consensus favor a modality of anticoagulant provision to COVID patients. Low weight molecular heparin is preferred when possible, largely related to its less frequent dosing requirements, but also its documented potential to suppress inflammatory bile markers. Ultimately, empiric dosing methods remain unclear in regard to risks and benefits for prophylactic or treatment dosing. Hesicovid seeks clarity on this with a hypothesis of whether therapeutic anastoparin will improve micro-vessel patency in COVID-related pulmonary microvascular thrombi to ultimately improve blood oxygenation. The Hesicovid authors designed a randomized, controlled open-label single-center study, including adult patients with COVID-related ARDS or Berlin criteria, and severe clinical presentation requiring mechanical intubation. Included patients were likewise evaluated with labs potentially suggested of hypercoagulability. Elderly patients older than 85 years were excluded, as well as those with a clinical indication for therapeutic anticoagulation or baseline coagulopathy. Additional exclusion criteria encompassed appropriate baseline comorbidities and remote health-related complication. Overall, the design and patient selection seemed fitting to support a comparable population while reducing potential confounding. 20 patients underwent block randomization between the two treatment arms in a one-to-one ratio. Therapeutic anticoagulation was administered as weight-based anastoparin, dose-adjusted for age and renal function. Prophylactic dosing modalities were based on weight, with agent selection between unfractionated heparin or anoxaparin made per provider preference. Each arm continued the subcutaneous intervention for a minimum of four days, but recommended treatment duration of up to 14 days. This leads us to our first poll question. For patients with COVID-19, with which threshold does your institution-based anticoagulation dose selection? A, a D-dimer greater than 2.5 milligrams per liter. B, a D-dimer greater than five milligrams per liter. C, a D-dimer greater than 10 milligrams per liter. Or D, your institution uses another laboratory parameter or a universal dosing method. Okay, interesting. So it looks like the majority of institutions are not using D-dimer to dose anticoagulation or they're using an alternative laboratory parameter. The primary outcome observed was the change in ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen at 7 and 14 days from baseline at enrollment. Secondary outcomes included time to successful extubation, defined as no need for reintubation within 72 hours, 28-day ventilator and ICU-free days, change in D-dimer at 72 to 96 hours from baseline, and all-cause 28-day mortality. The outcomes of bleeding occurrence were assessed using the thrombolysis and monocardial infarction or TIMI criteria. The authors powered this study to 85% for detection of improvement in PaO2 to FiO2 of at least 50 millimeters of mercury, with an alpha of 5%, again, following 20 patients for 28 days. Continuous and categorical variables were observed over distribution and assessed for differences, including TURQI's multiple comparison test for evaluation of the PaO2 to FiO2 ratio temporal change. Overall, statistical analysis was appropriately selected for evaluation of the data and complete with a survival analysis performed using Kaplan-Meier curve with Cox regression. Each cohort was well-matched with no significant differences in baseline characteristics. Patients were similar in age, primarily male, and only had exposure to prophylactic anticoagulation prior to enrollment. D-dimers were markedly elevated above a commonly accepted normal range of 500 micrograms per liter, and each group's baseline PaO2 to FiO2 ratio was consistent with moderate ARDS. Baseline patient acuity was also similar between groups with mean SIPA scores of 10 and FAPS 3 scores of 56. Digging into the results of the primary outcomes, between groups, no statistically significant differences existed at any time point. However, day 14 oxygenation ratios narrowly missed statistical significance in favor of therapeutic anoxaparin with a p-value of 0.057. Conversely, when observing within-group differences, statistically significant improvement in PaO2 to FiO2 ratio was demonstrated over time at day 7 and 14 in the therapeutic group, as illustrated by the graph to the left. Furthermore, therapeutic anticoagulation patients experienced a statistically significant reduction in D-dimer levels as measured within 3 to 4 days from baseline, while the prophylactic anticoagulation group experienced statistically significant increases in CRM D-dimer at similar time intervals. The therapeutic group also benefited by significantly more ventilator-free days. This correlated to a trend towards fewer ICU days as well, however, missed statistical significance, as did in hospital and all-cause 28-day mortality between groups. Successful extubation, again defined as no need for mechanical ventilator support for more than 72 hours following vent liberation, also supported therapeutic anticoagulation with a benefit of four times the likelihood of successful extubation during 28-day follow-up. This difference became apparent at approximately day 9 to 10 of treatment, as illustrated on this Kaplan-Meier curve. Bleeding was the only actively monitored safety outcome. Minor bleeding was characterized in two of the therapeutic group patients as hematuria and femoral artery catheter bleeding. The authors note that no major bleeding events were recorded in either group, however, two patients in the prophylactic group and four in the therapeutic group experienced bleeding that required medical attention, and these cases were likewise associated with a drop in hemoglobin of at least 5 grams per deciliter, but with no overt sign of hemorrhage. Thrombotic events were reported in two patients from each group, one DVT and one PE in the prophylactic arm, and two DVTs in the therapeutic arm. Overall, no significant differences were observed in any of these events related to bleeding or thrombosis. When interpreting these results, it's important to observe additional interventions within each group to determine how the clinical approach reflects current management practices for COVID patients. In regard to ventilation techniques, all patients had similar lung protective settings with low tidal volume, averaging 6.1 milliliters per kilogram. Prone positioning of at least 16 hours for patients with a PaO2 to FiO2 ratio less than 150 millimeters of mercury was also performed at similar rates in both groups, 80% in the prophylactic and 70% in the therapeutic. The same number of patients in both groups received corticosteroids. However, agent selection, dosing, and duration are not reported. And lastly, it's important to consider that no patients in this study with enrollment through April to July of this year received agents reflective of current treatment approach, such as remdesivir or recently less supported interleukin-6 inhibitors. Ultimately, the author's conclusion states that based on this data, therapeutic enoxaparin was able to improve oxygenation over time and increase the rate of successful excavation. However, these results are ultimately proof of concept and support a need for additional larger trials to evaluate empiric therapeutic anticoagulation in COVID-19 patients. Critically evaluating this study, it demonstrated a strong prospective and randomized design reflective of robust clinical trial. The authors selected appropriate inclusion and exclusion criteria targeting a patient population that we do seek further clarity for while also reducing the risk for confounding of the results. Likewise, both arms were well-matched, and this helped to solidify the observed outcome. Alternatively, though, a risk for bias is introduced by its open-label design. And while selective population criteria help to support internal validity of this small trial's results, external validity may be limited by a younger cohort, a trend towards larger BMIs, and D-dimers that could be considered mildly elevated on the spectrum of COVID infection. While clinically relevant results were evaluated in secondary outcomes, the study was not powered to detect significant impacts in these areas, such as mortality. And lastly, while we were able to gather some perspective from the COVID, HEPA COVID trial, patients were not exposed to the current approach to care, including remdesivir, and there is no delineation of corticosteroid use. So, based on this data, do you feel your institution should implement empiric therapeutic anticoagulation in COVID-19 patients? A, yes for all patients. B, yes in select patients, such as those with a D-dimer greater than a specified threshold. C, no, not for patients without a clinical indication, such as an identified thrombosis. Or D, more clinical trial data is needed to guide management. Interesting. So, I would say the results are split between yes and certain patients with a D-dimer threshold or that more clinical trial data is needed. I can say that based on our practice here at Yale-New Haven Health, we do select anticoagulation dose based on a D-dimer level, but I would agree that further clinical trial data is needed in choice of an empiric therapeutic approach. In summary, it appears that therapeutic anticoagulation is safe and effective at improving oxygenation specifically when observed through the PaO2 to FiO2 ratio in mechanically ventilated COVID-19 patients and may also decrease days on the vent. Therapeutic intensity anticoagulation trended toward a reduction in intensive care unit length of stay, which may benefit health systems and widespread resource use during a pandemic. Ultimately, this study was not powered to detect clinically significant outcomes, such as differences in bleeding, thrombosis, or mortality, and should only offer a proof of concept for further studies to determine the optimal prevention of macro and microthrombi in critically ill COVID-19 patients. With that, I'd like to thank you all for joining us this afternoon and also for this opportunity to present, and I welcome any questions that you may have. Thank you, Aaron. The first question we have is, was data reported for the percent of patients that were paralyzed? Yes, that's a great question. So, they don't do any type of subgroup analysis in regards to that. However, for patients within both groups, all 100 percent, 20 patients, received neuromuscular blocking agents. Thank you. Our next question, we are currently empirically escalating from prophylactic dose to intermediate dose, which is half or full therapeutic dose for critically ill patients. Do you think HESA-COVID justifies at least intermediate dosing over pure prophylaxis dosing at this time for ICU COVID patients? Yeah, I think that's a great question and certainly a possible perspective that this trial offers. So, I do think that it shows a compelling reason to investigate more intensified dosing for anticoagulation, whether that be intermediate or therapeutic dose, because while a PAO2 to FIO2 ratio is not an entirely clinically significant outcome, it is correlated to reduced healthcare challenge and workload. I can say that here at Yale New Haven Health, once we have patients with a D-dimer greater than 5 milligrams per liter, we do escalate to an intermediate dose anticoagulation as well and only pursue therapeutic once we have confirmed imaging and suspicion of thrombosis. So, I do think that this supports some clinical approach for intermediate and therapeutic dose, but I would say that there is more prospective trial data necessary to confirm that benefit. Thank you, Erin, and it looks like we have time for maybe one more question. The efficacy of this intervention seems promising with the noted small sample size. Do you have any concerns about the higher rates of bleeding in the treatment group given the small sample size, for example, not power to detect adverse bleeding events? I do think that that's an intriguing finding for this study. I think that some of the inclusion criteria also limit the external validity of this trial to a broader population, particularly when looking at their age. As we know, older patients have more severe reaction to COVID infection and likewise elevated risk for bleeding. Overall, the authors find that the bleeding events were not different between the groups and they actually found that there were decreases in hemoglobin within both groups, regardless of bleeding status, that they don't think was clinically relevant or required full intervention. So, I think from an extrapolation standpoint, I, again, agree that we could certainly be observing this in more populations, but I think that it should be limited to maybe younger populations as cited within this study because I have the same concern for increased bleeding and possibly older or more extremely severe patients. Sounds good, Aaron. Thank you. This concludes our Q&A session. I'd like to thank our presenters and the audience for attending today, and you can join us the third Friday of the month from 2 to 3 p.m. Eastern time for the next Journal Club Spotlight on Pharmacy. And thank you. This concludes our presentation. Thank you.

critical care medicine

melatonin supplementation

sleep quality

ICU patients

randomized controlled trial

therapeutic anticoagulation

prophylactic anticoagulation

COVID-19 patients

acute respiratory distress syndrome

mechanically ventilated patients