Hello and welcome to today's Journal Club Critical Care Medicine webcast. This webcast, hosted and supported by the Society of Critical Care Medicine, is part of the Journal Club Critical Care Medicine series. In today's webcast, we feature two articles from the September issue of Critical Care Medicine. This webcast will be available to registrants on demand within five business days. Log in to mysccm.org and navigate to the My Learning tab. Registrants of today's webcast will also have access to a bonus Journal Club recording which features articles from the April issue of Critical Care Medicine. My name is Thomas Zagmani and I'm a senior lecturer in intensive care at Cardiff University in the United Kingdom. I will be moderating today's webcast and I've got no disclosures to present. Thank you for joining us. Just a few housekeeping items before we get started. First, during the presentation, you will have the opportunity to participate in several interactive polls. When you see a poll, simply click the bubble next to your choice. Second, there will be a Q&A session at the conclusion of the presentations. To submit questions throughout both presentations, type into the question box located on your control panel. Third, if you have a comment to share during the presentations, you may use the question box for that as well. And finally, everyone joining us for today's webcast will receive a follow-up email that will include an evaluation. Please take five minutes to complete the evaluation because your feedback is greatly appreciated and will help us. 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 would like to introduce today's two presenters. Our first presenter is Dr. Eugene Urydyski, attended undergraduate and medical school at the University of Maryland, and completed his internal medicine residency at the Hospital of the University of Pennsylvania. Thereafter, he completed a cardiology fellowship at Georgetown University, Washington Hospital Center, and moved on to the cardiology faculty at the New York University School of Medicine. Presently, he serves as associate director of the CCU. His clinical and research interests include venous thromboembolism and cardiac arrest. Our second presenter, Dr. Paul Vesper, is the assistant dean for critical care medicine research, the Gary L. Brederson Family Chair in neurointensive care, and professor of neurosurgery and neurology at the David Geffen School of Medicine. He is ABPN board certified in neurology and UNCS certified in neurocritical care. Dr. Vesper is leading the university-wide initiative to enhance the multidisciplinary academic and research operations in critical care medicine at UCLA as the assistant dean for critical care medicine, and is the director of neurocritical care program at UCLA and clinical director of UCLA Brain Injury Research Center. Dr. Vesper is the current co-director of the Neurocritical Care Society Research Central. His clinical and research expertise is in traumatic brain injury, post-traumatic epilepsy, intracerebral hemorrhage, and bioinformatics. He pioneered the role of continuous brain monitoring in coma patients, imaging and microdialysis studies of brain metabolism, ICU informatics and telemedicine, and stereotactic treatment of brain hemorrhage in the ICU. He has published over 275 scientific research articles. Currently, Dr. Vesper is leading a large NIH-funded observational trial in post-traumatic epilepsy, and has been an NIH-funded co-principal investigator of the MISTI-ICS studies on minimally invasive surgery for intracerebral hemorrhage. Thank you very much, Eugene and Paul, for joining us today. Before we begin, could you each tell us if you have any disclosures to note? Eugene? No disclosures. Paul? I have disclosures that will appear on my slides. Okay. I will now turn things over to our first presenter, Dr. Eugene Urydziski. All right. Thanks for having me. So, I'm going to be talking about thromboelastography, which is also known as TEG, and what those profiles look like in critically ill COVID-19 patients with ARDS. The slides advance. All right. So, let's start off with a polling question. So, if you're taking care of a patient with COVID-19 who is admitted to your intensive care unit, what would you do to prophylaxe against venous thromboembolic events, or VTE? So, the options you have is you just use standard prophylactic dose anticoagulation with low molecular weight heparin or infractionated heparin. You start the patient on full-dose therapeutic anticoagulation with low molecular or infractionated heparin. You make your decision in part based on what the D-dimer value shows. You base your approach on some other clinical or laboratory parameters, or you use a non-pharmacological approach. So, I'll give you a few seconds to submit your answers. Okay. So, I'm seeing some responses come in. So, half of you voted that you use prophylactic dose anticoagulation, and the other half voted that you vary your approach based on the D-dimer. And actually, the latter is what we have been using at NYU. So, to dive in a little bit into the introduction. So, when the COVID pandemic hit the U.S., New York was one of the first cities that was hit really, really hard. And I think everyone turned their attention to the fact that a lot of these patients, particularly hospitalized COVID patients in ICU, were at significantly increased risk for venous thromboembolic events. And one of the early studies that made quite a splash showed that almost 50% of ICU COVID patients have a VT. The majority of these events turned out to be pulmonary embolic events or pulmonary embolism, which was thought to be maybe slightly different mechanistically than just clots that form in the legs. We later did a study that followed this with about 800 patients in ICU, and we found that the rates are still pretty substantial for VT, but definitely lower than this kind of 50% mark. For us, the rates were close to about 14%. And what was interesting is that patients with COVID-19 ARDS seemed to be quite different from run-of-the-mill patients with ARDS. In fact, one study showed that if you have COVID-19 and ARDS, your odds ratio for having a VT event is 6.2 compared to non-COVID ARDS patients. So, the question that we were asking at our institution is, why is this going on? What is it about these COVID-19 ICU patients that makes them hypercoagulable? And there are many ways to look at this, and traditional markers of coagulation, such as your PTT, your PT, etc., have certain limitations as they only interrogate small aspects of the coagulation cascade. But one very useful way of looking at coagulation is thromboelastography or TEG. So, unless you work in trauma, I think for a lot of folks who work in medical ICUs, we probably don't order this too much, so I want to just briefly review what a TEG is before jumping into the meat of the presentation. So, what TEG allows us to do is to really recreate the low shear stress of venous blood in an experimental setup. So, a blood sample is taken from a patient, it's placed in a cup, and that cup is placed into a machine equipped with a pin or a wire that is dipped into the cup. As the cup rotates around the pin, clot starts to form. And as the clot forms, this pin or wire starts to deflect. And you can actually graph these deflections as amplitude of the clot over time. And this is sort of what a clot would look like as it forms. And there are various aspects of clot formation and clot lysis that we can actually look at individually in this kind of holistic approach. So, the first thing that we see is something called the reaction time, or the R, which is how long does it take from the very start of this reaction as this cup is spinning around the pin to the point where the clot actually starts to form. And in part, this is related to your various coagulation factors. Next, we look at a factor called kinetics, or K, which is how long does it take from the beginning of clot formations for the clot to form up to 20 millimeters of amplitude. Similarly, we look at something called the alpha angle right here, which is a tangent of amplitude and time graph, which both that and the K somewhat represent your cross-linking or your fibrinogen in the clot. And then, very importantly, we look at a factor called MA subsequently. The MA is the maximum amplitude of the clot, which is largely related to your platelet count, your platelet function, and your G2B3A receptor interactions with fibrinogen and such. And then, subsequently, we can look at clot lysis at 30 minutes. So, these are kind of the main aspects, these five aspects here of the tag that we look at that can give us some sense of what's going on in terms of blood clotting in a patient. And if I skip ahead here, I can give you an example of what a normal tag profile looks like. Again, we have amplitude over time versus a hypercoagulable tag. And you can kind of see almost from the door these two look very different. A hypercoagulable tag has a very short R-time, so the clot starts to form pretty quickly compared to the sort of more prolonged R-time. You can see that the takeoff of cross-linking, this kind of alpha angle, is very, very steep here in a hypercoagulable patient compared to the normal tag. And you can very clearly see that the amplitude of the clot, which is, again, your platelet function, your platelet count, or cross-linking, is much, much broader than it would be in a normal tag. So, we wanted to see what does this look like in a COVID-19 patient. Let me skip over this slide out of interest of time. So, a tag has actually been looked at before in non-COVID critical illness. You can imagine there's been kind of a lot of interest since critically ill patients are at risk for clot formation. And one thing that has come out of this study of 82 patients and consistent with a number of other studies in intensive care units is that critically ill patients tend to have an elevated MA, or maximal amplitude, of the clot. So, for example, in this study of 82 patients with a variety of different medical critical illnesses, about 60% of patients tend to have a maximum clot amplitude, which is above the reference range. Whereas, specifically in this study, other aspects of clot formation, reaction time, kinetics time, alpha angle, all seem to generally fall within the normal range. And it's really this MA that seems to be really kind of elevated in critical illness. And generally, while tag is used to look at bleeding patients to try to determine how factors should be repleted, there have been a few studies like this one that have actually looked to see whether tag can actually predict DVTs or various forms of venous thromboembolic events in patients. This is probably the largest study that looked at about 1,000 trauma patients. And what they found is that while no single parameter on tag can really predict thrombosis, if you take an aggregate of multiple parameters, you can predict the risk of DVT in these critically ill patients with an odds ratio of about 2.4%. So, not amazing, but definitely gives you some predictive value of DVTs. So, for us, before kind of diving into some results, is what we did is we took patients who were being admitted to our ICU. And all of these patients had ARDS. All of these patients were intubated. The threshold was pretty high. And we collected tags on them along with a number of other labs that I'll go through within the first 72 hours of admission. Not to confound the study, we've excluded a certain population of patients that we thought may have deranged tags at baseline. So, we excluded individuals with cirrhosis. We excluded those with thrombocytopenia, malignancies, or any other reason to be on anticoagulation really just to try to isolate patients where COVID ARDS is their main issue. So, before actually diving into Table 1, I wanted to go through this a little bit backwards. But here, I have various parameters of tag and kind of what these profiles look like on our patients. And the total number of tags that we collected was 64. So, 64 individual tags. In addition to the parameters that I discussed that you can see with the tag, there's also one that I didn't discuss, which is called the CI, or your clotting index. And this is kind of a proprietary formula developed by the tag folks who put together the assay, which takes into account the various different coagulation aspects of the tag, puts it together in a formula, and says, if this value is above 3, you're hypercoagulable. If this value is less than negative 3, you're hypocoagulable. And anything between negative 3 and 3, you're in the normal range. And that's been tested in some other studies that I won't go through. But what we found in our 64 COVID ARDS patients is that half of them, by this formula, were hypercoagulable. What we also found is that 44% of patients had abnormal R times and K times. So, remember that the R time, the shorter the R time, the more rapid the reaction time. And then with the K time, it's just quite the opposite. So, the faster your K, the more likely the clot is going to form to that amplitude of 20 millimeters faster. We also know that 70% of our patients had very steep alpha angles. So, that metric that's related to cross-linking of fibrinogen was quite high. And that 60% of our patients had an elevated maximum amplitude. All of these are being shown in reference to kind of the standard. And this is, as I showed you, very similar to what was shown in the non-COVID critical patients, that 60% of them have an elevated MA. So, very consistent. Just a graphical way to represent this. So, I kind of broke down all of these type parameters by clotting index, reaction time, kinetic time, alpha angle, maximum amplitude, and then clot license. And these kind of dash bars represent what the normal limits are. So, you can see the kind of where our clotting index fell. That's at 50% of our patients had an abnormally elevated clotting index. And then, importantly, if this is the upper limit of normal of your alpha angle, quite a number of our patients had elevated alpha angles. If this is your upper limit of normal of your maximum amplitude of 70 millimeters, you can see that the majority of our patients seem to fall above that. And then, since I did this a little bit backwards, I'll jump into our table once you actually get a sense of what our patient population looked like. So, as I mentioned, we wound up having 64 patients that met our inclusion and exclusion criteria. And then, we separated them into those with elevated clotting index based on that formula and clotting index that was normal or hypofogulable. And the kind of key things to point out is that most of our patients didn't have too many pre-existing comorbidities. A fifth of them had underlying cardiovascular disease, but less than 10% had pulmonary disease, CKD, but they were quite sick. So, you can see that 60% of our patients were in shock. About half of them were in renal failure. And as I'll show you later, about over the course of three weeks, 30% of them wound up having a venous thromboembolic event. You know, I showed you all of these tech parameters. You can also see kind of some of the other factors that we looked at. We collected labs probably every 24 to 48 hours and looked at things such as platelet counts, d-dimers, which on average were quite high at 2300 in our patient population, INRs, fibrinogens, TRPs, ferricins. You can see that they're all quite elevated above the reference range. Our mean fibrinogen was about 700, our CRP on average was about 100, and our ferritin levels were about 1400. And if you break these down by clotting indexes I have here, other than the tech parameters, which not unexpectedly are going to differ based on your clotting index since your tech parameters are included in that formula, the only real difference that we found is that your INR was just slightly different, but not specifically significant. Sorry, this box is over a different value. Your d-dimer was different based on whether or not you were hypercoagulable with an elevated clotting index or not hypercoagulable. We also took an interest in whether or not any of these tech parameters correlated with other markers of inflammation, just since there is clearly a link between inflammation and coagulation. And we really found that if you're looking at spearmint correlation coefficients that correlates between various tech parameters and markers of inflammation are moderate to low at best. So the MA did correlate somewhat with fibrinogen and your CRP, but these R values aren't particularly too impressive. The alpha angle, which is your fibrinogen cross-linking, correlated to some extent with platelet count and your fibrinogen. Again, not a particularly robust correlation. And the reaction time did correlate with the d-dimer as expected. So it would give you a negative range to higher d-dimer the lower your reaction time. I showed you a little bit about some of our clinical outcomes. So because we weren't systematically getting ultrasounds or chest CTs on these patients to look for clots, we kind of have some limitations to conclusions that we can draw. But at the end, 63% of our patients did wind up having lower extremity ultrasounds and only 16% did wind up going for chest CTs to look for pulmonary emboli. And what we found is that in our population, we diagnosed 31% with venous thromboembolic events. Almost all of them were DVTs, unlike some of the previous data that I showed you where most of these events were PEs. One patient wound up having clot in transit and one patient had a P as well as a DVT. About 30% of our patients died. And if you look at death or venous thromboembolic events as a composite outcome, 52% of our patients met that. So quite impressive. Unfortunately, we didn't find that various TEG parameters were able to predict death or venous thromboembolic events as a composite endpoint. And the only thing that we found is that those who met this endpoint were more likely to have an elevated CRP of 136 versus 82 and elevated fibrinogen of 713 versus 562. But TEG didn't seem to predict these outcomes. And then lastly, what we decided to do is we broke down the total patient population by D-dimer cutoffs. We used 2000 because for two reasons. One of them, the 2000 D-dimer was close to our median value in our ICU population, but also because when the COVID pandemic or the surge really hit us, we started to use D-dimer of 2000 or rapid delta D-dimer as our marker to start therapeutic anticoagulation on patients. So, if you break patients down by this, we were also interested to see whether or not this at all played a role into our decision. The only thing that we really found is that our time was the only thing that tracked with D-dimer. Not surprisingly, patients with a particularly high D-dimer had a shorter reaction time, which means that their factors were more limited and that their clots started to form earlier. Clinically, not by much. We're talking about 4.8 seconds versus 5.6 seconds. Any other markers that we looked at, including your actual clinical events, which is your venous thromboembolic events, your alpha angle, your maximum amplitude and such did not seem to track with that. So, looking at our key conclusions. So, the majority of our patients in ICU wound up having a hypercoagulable tag. And these tag profiles seem to be somewhat similar to non-COVID-19 critically ill patients, mainly in the sense that the maximum amplitude of the clot seemed to be similar. Other markers in COVID-19 patients seem to be more dramatically different than those in non-COVID-19 critically ill patients. But perhaps what was the most interesting conclusion of our study was that it seems to be that platelet function and fibrinogen seem to be playing an important role in these tag derangements. I mentioned that the maximum amplitude and this alpha angle, which are factors of platelets, G2V3A receptor interactions with fibrinogen, they all seem to be somewhat deranged. And because there are certainly quite some data that look into sepsis, viral illness and demonstrate that there's increased platelet reactivity, this really brings into question whether or not we should be thinking about antiplatelet therapy in COVID-19 critically ill patients. But more to come on that. So, that's it for my presentation. So, I'm going to hand it over to Dr. Paul Vespa. Thank you, Dr. Uritzky. That was very interesting. The title of my presentation is The DECIDE Trial, Does Use of Rapid Response EEG Impact Clinical Decision Making? And I'd like to thank the Society for this opportunity to present in the webinar. These are my conflicts of interest. This study was sponsored by Cerebel, and I am a consultant and got funding for the research in this trial. The following question that we'll start with is as follows. What is the consensus recommendation for getting an EEG on a patient with status epilepticus? Select one of the following. Within one hour, one to two hours, two to six hours, longer than six hours. Okay. And pretty good split there between within one hour and one to two hours. That's very good. We can go back to the presentation. Okay. So, we'll explore that here. So, background and significance of the study. Patients suspected of having non-convulsive seizures in an emergency or acute care situation is a common problem. There's often uncertainty about whether the patient is seizing. The patient is encephalopathic, unconscious, and there's a lot of uncertainty. Current practice is driven by guidelines to obtain an EEG as soon as feasible within one hour, and that's supported from the references as noted there below. The consensus conference recommendation from the Neurocritical Care Society. However, there are long delays in diagnosis and treatment of status epilepticus, and numerous studies have demonstrated that there are delays in treatment, and often there are improper dosing or selection of drugs and or delay in treatment overall, and that delay in treatment is associated with poor outcome. So, question number three here. How long does it take to get an EEG on a critically ill patient with suspected seizures or status epilepticus in your hospital? And the choices are less than one hour, one to two hours, two to six hours, and longer than six hours. Okay, great. And there's a pretty good distribution there and suggesting that it takes longer than an hour to get an EEG at most of the respondents' hospitals. Great. Thank you for answering that. And so this was the study that was done by my collaborators at multiple academic institutions and was published this year in Critical Care Medicine. Our hypotheses were as follows. The use of a rapid EEG system by a bedside consultant, a physician in this case, would reduce the time to obtaining EEG findings in patients with suspected status epilepticus. And hypothesis two is rapid EEG would inform clinical decision making. The methods for this study were a prospective, non-randomized, multi-center observational study. There are really two studies in one. One was a physician survey study and the second was a patient observational study. 181 patients were encountered during this study. 164 patients had clinical and EEG data available. And there were 37 physician respondents in this study. Our outcomes primarily were time to the first EEG, change in physician's diagnostic assessment plus confidence in diagnosis, change in physician's treatment plan plus confidence in the treatment plan, and then ease of use of the technique by a non-EEG trained medical staff and adverse events. This is a synopsis really of the steps in this study. The idea is that the consult starts, patient is evaluated by the consultant, there's a decision that one needs an EEG, and then before doing the rapid EEG, the physician would complete a survey questionnaire that would really outline the level of suspicion for seizure, confidence in the diagnosis, intention to escalate treatment, and confidence in treatment decision. And then the rapid EEG would be performed, as we'll show in a minute. And then after the rapid EEG is performed, and usually for a fairly short interval of time, a post-questionnaire is answered. It's the same questionnaire to see if the physician treating professional has changed their perception. And then a conventional EEG is obtained as soon as possible. This is the rapid EEG system schematic. This is a headband that goes around a person's head. It's prefabricated. It's kind of a one-size-fits-all. It has a number of contacts here that have some gel in them. You twist these contact knobs and the gel comes out and makes contact. It gets this headband, which is disposable, gets connected to a portable device. The portable device plays the EEG in real time and also has a sonification if there is a seizure. It makes like a scratching sound if a seizure occurs. And the physician at the bedside or health professional at the bedside can look at the EEG in real time. This also is uploaded automatically to a HIPAA-secure cloud and can be reviewed via the cloud. And monitoring is performed right there in real time by the clinician. These were the results of the patient and physician characteristics. The top is the physician characteristics. These were patients who were in ICU or clinical practice, years of EEG experience, and then patients enrolled per site. These were the patient characteristics as outlined below. And the next slide shows the clinical diagnosis, suspected clinical diagnosis or actual clinical diagnosis. And what one can see is that the majority of patients had either suspected, possible status epilepticus, intracranial hemorrhage, or some sort of altered mental status of unknown cause. And then there were handfuls of other diagnoses for which the patients were encountered. The primary results here are shown above. So this is the hypothesis one. With the use of rapid EEG, shown on the top line here, the median time to getting the setup of EEG going was about five minutes, with the interquartile range there, four to ten minutes. And it was true basically across all of the five sites. The time to conventional EEG, and it's important to note that conventional EEG is ordered at the time of the setup of the rapid EEG. So this is time for conventional. This is where the EEG tech comes with the machine and hooks them up. And you can see here the median time is 239 minutes, so about four hours. And you can see the distribution across various sites. And some of, you can see the percentage of after hours at EEG, after the normal sort of nine to five hours. And you can see the other information on that slide. So this is a graphic showing the distribution of time to rapid EEG, which is on the bottom. And you can see that the y-axis here, ten minutes, is indicated here. So the median times and then the distribution, versus the time to conventional EEG, which is on the upper slide, really is in the number of hours. And again, it's about, the median was about four hours for conventional EEG. Phase two is, would, can be defined as, does rapid EEG inform clinical decision making? So again, we looked at the confidence in the diagnosis, suspicion of seizures, treatment, and confidence in the overall treatment plan. So the suspicion of seizures went down after rapid EEG across patients. The confidence in the diagnosis went up, high and very high confidence went up. With regards to treatment, escalating treatment, the intent to escalate treatment went down. And the confidence in the treatment plan of high or very high confidence went up, as you can see here. So it appears that the rapid EEG information really did inform clinical decision making. This is the clinical impact of rapid EEG broken down here, in terms of patients, in terms of seizure and non-seizure and how it changed before and after rapid EEG. And same thing with the various categories that we just talked about, it's broken down there in detail. The incidence of seizures, EEG seizures, in the DECI trial is shown here. We had 164 patients that had EEGs that were reviewable, 11% of the patients had electrographic seizures, 17 patients had seizures during the rapid EEG itself, 11 patients had seizures during rapid EEG and follow-up conventional EEG, 6 had seizures on rapid EEG but not during the conventional EEG, 5 patients had seizures later, they didn't have them on the rapid EEG but had it on conventional EEG later, and 12% had what's called highly epileptiform patterns without seizures, and the rest of the patients had slow or normal appearing EEG. And I think this highlights an important point that conventional EEG after the rapid EEG is important because A, patients can either continue to seize or some patients who weren't found to be seizing initially with the rapid EEG were found later to be seizing on conventional EEG. The rapid EEG did inform clinical decision making and so you can see here the change in accuracy, sensitivity, and specificity before EEG, rapid EEG, and post-rapid EEG. And what one can see is that all of these parameters really improve and so the clinical judgment improves it seems by all three of these measures with the use of both clinical and rapid EEG information. Rapid EEG was easy for these physicians to use. We scored the use of the headband which is this device that goes around the head as well as the actual recording instrument and this was sort of a satisfaction score if you will of how easy it was to use and you can see that and you can see the distribution across sites as well, pretty similar across sites. Adverse events, this headband device has these knobs which twist and the twisting sort of abrades the skin a little bit and moves the hair out of the way and then gel goes onto the skin. So one always worries about skin irritation, infection, serious adverse effects, events related to the procedure itself. There were no adverse events or serious adverse events noted in this study. And in conclusion, it seems that the rapid EEG enabled bedside EEG enabled the EEG to be started within five to six minutes after decision to perform an EEG. And this really seems to delay a several hour, seems to avoid rather a several hour delay in conventional EEG, again five to six minutes versus about 239 minutes median time. Rapid EEG informed clinical decision making at the bedside and based on the questionnaire survey, it seemed to enhance confidence in diagnosis, enhance confidence in decision making and made seizure diagnosis more sensitive and specific. And finally, the ease of use scoring demonstrated that the rapid EEG was easy and was safe to use. And with that, I will conclude and move on to the question and answer session. Thank you very much for the presentations. And here is our first question from the audience to Dr. Juricic. The correlation between these traditional clotting markers that you have also measured and the tag is relatively weak. Do you have an explanation or a hypothesis that why is that? Because the patients... Great. Great question. So this has actually been looked at in some other studies, not necessarily in the critically ill population. And the results seem to be consistent. The traditional markers of coagulation don't correlate so well with tags. One of the thoughts is that with these traditional markers, you're really interrogating very specific isolated aspects of the coagulation cascade in plasma, whereas here you have whole blood and multiple other factors that go into it. Similarly, you know, if you have a patient with cirrhosis and an INR of four, it would seem that that person is predisposed to bleeding, but by tag, they might actually be hypocoagulable or hypercoagulable. So it doesn't seem to correlate probably just because of the number of different factors you're interrogating with tag as compared to isolated factors such as PT and PTT. Thank you very much. Dr. Vesper, it seems that this rapid EEG was really easy to use. Can I ask that what training was provided for the clinicians at the different centers and how can this be upscaled? Yeah, that's a very good question. For the purposes of this trial, there was a training session for each site of the potential users that potentially could use it, and there was a demonstration of the application of the headband and connection using a human volunteer, and this training session really was about 45 minutes to 60 minutes in duration. And I think that the scaling of this really is something that needs to be done across institutions, and potentially it would be something that could be part of a future skills workshop like many other skills workshops where devices are introduced. But it seems to, you know, I think in this study this was really mostly neurology-based physicians, so I think that there probably is some degree of familiarity with EEG in general, and to scale it up to teach people that don't have that sort of perspective might require additional types of training, but it's a good question, I think, for future study. Well, thank you very much. Eugene, why did you choose the cutoff of the D-dimer at the level you set it? Do you have any prior evidence, or is it just based on your results from the groups? Yeah, it was based on two things. One, it was based on that the 2,000 was very close to our median, off by a couple hundred, and the fact that at our shop we chose 2,000 very early on during the surge as the marker to say we should probably start therapeutic anticoagulation to prevent VTE. That was partly chosen arbitrarily. I mean, early data showed that at a D-dimer cutoff of 1,000, patients' mortality starts to increase. So kind of as a large group of intensivists, people got together and decided if your D-dimer is skyrocketing or if it's above 2,000, we as a group think as just expert opinion that we should start. But yes, it was not chosen based on much more complicated data than that. Thank you very much. Dr. Vespa, what are the cost implications of this rapid EEG device? Again, it is going towards the upscalability, and how can it be delivered probably not just in a neuro center? Yeah, I think that the cost of doing rapid EEG need to be compared with the cost of staffing conventional EEG on a 24-hour basis. It's very difficult to staff people on a 24-hour basis, technologists, and it's costly to do so. And I think there are comparative cost estimates about having this device and having it be applied by physicians or other healthcare professionals at the bedside without the additional cost of having on-call EEG technologists. And so I think there's potentially a cost benefit, although that would need to be demonstrated really in prospective studies. But I think that the unknowns really are how this information, the rapid information, affects other sort of cost variables, too, and treatment and whether or not the person is going to be admitted to an intensive care unit directly, how much treatment they're going to be getting, et cetera. And I think so there's really a need to study the full spectrum of costs associated with patient care, and that's to be done in the future. It's still not defined yet. Thank you very much. A follow-on question on that is, looking at the results, the confidence in the diagnosis went up and the treatment for epilepsy was, or the epileptiform activity went down. Is it, again, a possible benefit that if the rapid EEG is interpreted appropriately, then less is more and drugs can be stopped, which are not necessary? Yes, we were pleasantly surprised to see that. I think it's common experience that people worry about whether or not seizures are still occurring and medication is provided out of due diligence and an abundance of caution. However, we all know the literature of sedation in intensive care units and how over-sedation really is an undesirable goal at this point. And so this kind of speaks to that concept of perhaps withholding additional anti-seizure medications, which are usually sedating and may have other side effects. And if you have a better sense of the EEG, then you can avoid that medication. It's sort of similar to an EKG example of monitoring the EKG and not prophylactically giving an antiarrhythmic, which, just because you don't know what the rhythm is, with EKG you have direct feedback of the rhythm and you know exactly when or if you should be giving an antiarrhythmic medication. So it's sort of akin to that, where now with the EEG, you can pause and say they're not seizing, I don't have to give another medication, and that's what the survey results showed in this study. Thank you very much. One of the questions which are coming is that the VTE rate at your population was around 30%. Is that comparable to others or is it higher or is it lower? So there have been quite a number of studies published, including a meta-analysis of VTE rates. It probably falls somewhere in between, as I've shown that when we actually looked at all of our critically ill patients, about 800 plus or so, the VTE rate seemed to be lower at about 13 to 14%. Probably that 50% mark by the clock paper early on, published in Thrombosis Research, was on the higher side. So I would say probably in between looking at the number of studies that have been published prior to ours and post. Okay. And having said that, we weren't systematically looking for them, so it's very possible that our patients had higher rates. Thank you. Following on that, the anticoagulation, what was your anticoagulation strategy when you for therapeutic anticoagulation? Unfractionated heparin or low molecular weight heparin or a combination? Right. So I failed to mention that, but what we did in this patient population, these 64 patients, 86% of them were on therapeutic anticoagulation without having VTE events diagnosed prior, just because of the D-dimer cutoffs and the perceived risk. Generally speaking, we favored unfractionated heparin to low molecular weight heparin, simply because of the rates of renal failure. So in this cohort, it was close to 50%. But if that wasn't the case, or we watched patients for a number of days without significant renal injury, sometimes at that point, we would put them on full dose low molecular weight heparin. Thank you very much. Dr. Vespa, it is a sort of personal question. I was really fascinated by the talk. In Wales, we struggle with electrophysiology resources. So I was fascinated to see that there is a rapid way of analyzing the EEG. What are your thoughts in that? Is there a possibility to pool these resources if you would have these devices available more widely? Is there a role for telemedicine, perhaps? Yeah, that's a great question. I definitely think there's a role for telemedicine. These types of devices, even conventional EEG is nowadays valuable via web interfaces. The rapid EEG in this device has a web interface that's HIPAA compliant and real-time, and it has maybe a couple minutes away where you're seeing EEG near real-time. I think that is quite good, and I could envision clinicians teaming up to be able to review and cross-cover other hospitals. Telemedicine is growing. It's been demonstrated to be quite feasible for critical care and for other specialties, and I think this is one of those examples. I think the problem that you face locally is not unique. I think it's a problem that most places have where it's a very labor-intensive, person-intensive procedure, typically. It's costly to have people on call. It's difficult to find individuals who want to be on call all the time, and yet the patient needs are emergent, and brain conditions in general are being seen and viewed appropriately as being very emergent, and we can see that with the stroke treatments that are out there, and so we're coming upon an era of needing this type of technology, including telemedicine, to enhance care. Telemedicine has done a wonderful job at enhancing stroke care, and some of the most important advancements in stroke care has been the use of telemedicine, so I think this is a logical extension of that general concept. Thank you very much, and this concludes our Q&A session. Thank you to our presenters and the audience for attending. Again, everyone who joined us for today's webcast will receive a follow-up email that will include an evaluation. Please take five minutes to complete the evaluation. Your feedback is greatly appreciated. And on a final note, please join us for our next Journal Club on October 22nd for a discussion on articles from the Discovery Network. This concludes our presentation today. Thank you so much. Thank you.

TEG

COVID-19

ARDS

coagulation

platelet function

fibrinogen cross-linking

EEG

status epilepticus

diagnostic confidence