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 Critical Care Medicine. My name is Thomas Zagmani and I'm a Professor of Intensive Care at Cardiff University in the United Kingdom. I will be moderating today's webcast. Thank you for joining us. Just a few housekeeping items before we get started. First there will be a Q&A session at the conclusion of both presentations. To submit questions throughout the presentation, type into the question box located on your control panel. Second, if you have 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 it will include an evaluation. Please take five minutes to complete the evaluation. Your feedback is greatly appreciated. Please note, this presentation is for educational purposes only. The material presented is intended to represent the 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. Dr. Elliot Bennett-Guerrero is Professor and Vice Chair for Clinical Research in the Department of Anesthesiology and Medical Director for Perioperative Quality and Patient Safety at Stony Brook Medicine. He graduated from Harvard Medical School in 1991 and completed training in cardiac anesthesiology and critical care medicine at Duke University. His research interests include the safety and effectiveness of blood transfusion, surgical site infection, post-operative morbidity and cancer recurrence. He has received significant funding from numerous sources, including the NIHR as well as industry and has led diverse clinical research projects during, including large multicenter randomized trials, pharmacokinetic studies, healthy volunteer studies and large database analysis. Dr. Pablo Sanchez is an Argentina board certified thoracic surgeon focused on lung transplantation and lung failure. Dr. Sanchez did his thoracic residency in Brazil, along with receiving his doctorate of physiology in pulmonary medicine from Brazil. He was a postdoctoral research fellow under Dr. Berthe Griffith at the University of Maryland in Baltimore. Dr. Sanchez joined the University of Pittsburgh Medical Center in 2017. In 2018, he was promoted to vice chair, Benign Lung Diseases, Chief Division of Lung Transplant and Lung Failure, ACMO, Director of Lung Transplants Research and EVRP Program, along with being an associate program director for the Integrated CT Surgery Residence Program. Thank you both for joining us today. Now I turn things over to our first presenter, Dr. Elliot Bennett-Guerrero. Good morning, or good afternoon, everybody. It's a pleasure to be able to speak about our research that was published in Critical Care Medicine. In terms of the background for this project, we know that in some settings, convalescent plasma should work for the prevention and or treatment of infections. I'm sorry. And historical data, mostly from non-randomized studies, suggest that convalescent plasma may be a useful tool in the treatment of some viral illness. However, it's important to recognize there are very little high-quality data. For example, an NIH-funded, multi-center, double-blind, randomized trial of convalescent plasma for the treatment of influenza was conducted. And in this study, they showed that two units of convalescent plasma were not effective compared with two units of standard plasma, this being one of the few studies done in the past with another type of virus that did not show any efficacy. So despite the large-scale use of convalescent plasma for COVID-19, very few randomized trials have been published. We know that it's theorized that use in early treatment, that is within the first few days after infection, may be beneficial. But benefit later on in the disease is less clear. So at our hospital in New York, during the spring surge of 2020, our 610-bed hospital had approximately 500 COVID-19 patients, of which 135 were intubated. So we were quite overwhelmed and overrun with patients at our hospital. And we felt compelled to investigate convalescent plasma as a potential treatment, given the lack of any proven benefits at that time. So we focused on hospitalized patients where there was a large unmet need, but we recognized that it might be too late in these patients' disease course for them to benefit. It was too challenging at our center to administer study plasma in an outpatient setting earlier in the course of treatment. So based on feasibility, we really needed to focus on hospitalized patients. So the objectives of our trial were to determine whether the administration of convalescent plasma to patients with COVID-19 infection would increase antibodies to SARS-CoV-2. And more importantly, we wanted to see if the administration of convalescent plasma would improve outcome in this setting. So at this time at our center, almost all other research had come to a grinding halt, just because our whole institution was overrun with COVID. Everything was closed. So we were fortunate. And with the dean and other people's support, we were able to assemble a very large team, about 75 individuals, to conduct this randomized trial. And we were able to redeploy many people who had been involved in other types of hospital work or research projects to work on this project. So we had eight multidisciplinary physician investigators. We had three individuals helping us with obtaining an IND and regulatory support. We had a team of 13 individuals who were calling potential convalescent plasma donors who had completed an online screening tool. Then we had a team of 16 individuals who were doing in-person consenting and screening of convalescent plasma donors for the trial. And then we had five individuals at our blood bank involved in the collection of convalescent plasma. Then we had another team of 22 individuals involved in the consent and administration of recipients. Then we had a smaller group of five individuals who were unblinded randomizers and antibody testers, since these individuals needed to be unblinded. And then we had an independent data safety and monitoring board for the trial. So we needed to move very, very quickly, given the large number of patients at our institution. And so within approximately two weeks, we were able to write a protocol and submit it to the IRB, which was approved on April 1st. And then on April 2nd, our IND was approved by the FDA, and we began online screening. On April 8th, we began in-person consent of potential convalescent plasma donors, which involved a written questionnaire, for example, a standard donor health questionnaire that all blood donors fill out. Then we did point-of-care antibody testing to SARS-CoV-2 nucleocapsid protein. I just wanna point out that the FDA's guidelines for antibody testing changed several times during the course of our trial, but initially, no antibody testing was required for people who had recovered from COVID-19 infection, but we felt that it was an essential part of our trial, and so we included this, even though it was not required by the FDA at that time. And for the assay that we used, we used an assay threshold of 145 optical density units, corresponding to the minimum neutralizing antibody titer of one to 80 suggested by the FDA. PCR testing was done if people had been less than 28 days from symptom resolution, and then, as with all blood donors, routinely, venipuncture was done to send transmissible disease testing labs, CBC, other routine tests. And I was very happy with how successful we were collecting plasma at a single center. We collected 865 units from 262 convalescent donors over a very short period of time. And in a randomly selected subset of units, the convalescent plasma had high levels of neutralizing antibodies to SARS-CoV-2. The antibody tests that we had used to screen people was not a neutralizing antibody test, since these are much more labor-intensive, but using two different types of assays, both the gold standard plaque neutralization assay, as well as a pseudotype assay, we showed that the convalescent plasma that we collected and administered had high neutralizing levels on the order of, as you can see, one to 334 or one to 526, in terms of the titer. In terms of the recipients that we enrolled, we restricted it to adult patients who were hospitalized with a confirmed diagnosis of COVID-19 infection. So we had pretty broad criteria. But although we wanted to include hospitalized patients, we didn't think it would be worthwhile to enroll people who had, say, been in the hospital already for six weeks. We thought most of the experts agreed that it would be too late for there to be benefits. So we restricted enrollment in the trial to people who were randomized within 14 days of hospital admission. And we also excluded patients where the treating physician did not believe they'd be able to tolerate two units of plasma, so approximately 500 mLs, as well as prior reaction to transfused blood products, and female patients with a positive pregnancy test. But these other exclusion criteria, as you can imagine, were quite rare. So we randomized patients four to one using interactive web response randomization tool in RedCap. So four to one means that for every five people enrolled in the trial, four would receive convalescent plasma and one would receive standard plasma. And the reason we did this is because given the dire need at our institution to try to help patients, we wanted to have as large a group as possible receiving convalescent plasma in case there was a benefit. But we wanted to design the trial such that there would be a large enough control group to be able to assess the safety and efficacy of this unproven therapy at the time. And we stratified by non-intubated versus intubated patients. The intervention was simply two units of convalescent versus standard plasma that was collected prior to 2020, each over one to four hours. And everybody in the trial was blinded except for the blood bank members and a small dedicated study team as I mentioned before. And the plasma had an identical label stating convalescent plasma or standard plasma to preserve blinding. So our primary endpoint was the total number of ventilator-free days from randomization to day 28. This is a commonly used endpoint in many critical care trials. But we also wanted to look at all-cause mortality through 90 days post-randomization. And then similar to most other clinical trials that were being done, we also included the World Health Organization Ordinal Scale, which looks at how much oxygen support people are on. And we wanted to detect at least a two-point improvement in that. Then we also looked at changes in antibody levels from baseline to several days after administration of the investigational product to see if the antibody levels were increased. And then we had certain pre-specified safety endpoints, a safety monitor not otherwise involved in the trial reviewed these, and then an independent DSMD also was in place to review adverse events. So all data were housed in a REDCap database and we had a dedicated data manager and statistician for the trial. And our primary analysis population was all patients randomized. So an intent-to-treat, which is the most conservative analysis method for clinical trials such as this. But we also planned a per-protocol analysis for patients who were randomized and received both units of plasma to which they were assigned. And the original sample size would have provided enough participants to detect a two-and-a-half-day difference for the convalescent plasma group versus standard in ventilator-free days with at least 90% power and two-sided significance level of 0.05. I just wanna point out that when we were enrolling people, it was interesting that there was very strong interest in the trial by families of patients who were extremely sick and had multi-system organ failure. So patients who had been ventilator-dependent for several weeks had very strong interest by their families in wanting to be in the trial. And it's interesting that in contrast, many of the patients with mild COVID-19 infection who had just been admitted were reluctant to try a blood product transfusion experimental therapy. So it was interesting to see this dichotomy in terms of reluctance by people who were doing pretty well and weren't that sick versus kind of a desperate desire to be in the trial by people who were beyond the window where they could be benefited. So we stopped enrollment of the trial early after the FDA granted the emergency use authorization for convalescent plasma. It was unlikely that patients would wish to participate in a randomized trial for an approved therapy at that point. So we felt like we had no other option but to terminate enrollment. And at that point, 82 patients had been enrolled with 74 were randomized. And consistent with the plan four to one target, we had 59 convalescent plasma patients enrolled by then and 15 standard plasma patients enrolled. And there were only two patients who were randomized who did not receive plasma, one in each study arm. One patient in the convalescent plasma arm experienced a serious transfusion reaction after a small volume was administered. This person recovered after receiving treatment. And obviously we did not complete administration of the two units of plasma in this patient. No patients were lost to follow up through the 90 day study period. So even though the enrollment was terminated early, so we had a relatively much smaller group of patients than we had wanted to have when we initiated the trial, the study groups were pretty well balanced in regard to most baseline characteristics. For example, age, sex, BMI. For example, patients intubated at baseline, 19 versus 20% was similar. The median days from symptom onset to randomization, which is a very important part of the trial, was nine days in both of the arms. And I just wanna point out that this, like many other trials in patients who are hospitalized, shows we were not administering this to people early in the course of their disease. So this was clearly not early treatment. We also showed a similar percentage of patients who had any FDA defined sign or symptom of severe COVID-19 infection. And then we also calculated an APACHE-2 score in all the people, which is a good composite index of comorbidities and physiologic status. And this was well balanced between the two arms. There were some imbalances at baseline. There were more patients receiving convalescent plasma, receiving remdesivir and on supplemental oxygen. But in general, the groups were pretty well balanced. So we were able to show that administration of convalescent plasma did increase antibodies to SARS-CoV-2. You can see in this top left panel, here we have the control plasma group and then the convalescent plasma group to the right. And this is looking at the antibody levels. And this is at baseline. This is before they received any standard or control plasma. And you can see that there was quite a distribution where many of the individuals who we enrolled already had antibodies to SARS-CoV-2, which was not surprising and is consistent with some other studies. And this was similar for both IgM antibodies on the left in panel A and IgG antibodies to the right in panel B. And with respect to the percent change in antibodies from baseline, so to day one, day seven, day 14, 21 and 28 from baseline, we saw that the administration of convalescent plasma shown in red increased antibodies to SARS-CoV-2 as shown in panel C. And this was also consistent in terms of the absolute change in antibody levels versus we saw no significant increase or just in some cases, a decrease in antibody levels in patients randomized to standard plasma. So although we were able to increase antibody levels to SARS-CoV-2, unfortunately, we didn't observe any significant difference in the study groups for the primary endpoint of ventilator-free days. There wasn't even a trend towards any benefit. The improvement of at least two points in the World Health Organization ordinal scale was achieved by 20% of subjects in both study arms. So again, no trend towards any benefit with convalescent plasma. Mortality at 90 days was numerically lower with convalescent plasma, 27 versus 33%, but given the small sample size, did not achieve statistical significance. On this slide, I'm just showing the time to all-cause death, the survival curves, Kaplan-Meier analysis. And you can see in panel A in the upper left-hand corner, a survival through 28 days with what looks like perhaps a trend towards more people in blue receiving convalescent plasma surviving, but it was not statistically significant and the sample size was small. And then again, survival through 90 days is shown in panel B. You can see the decrease in survival over the 90-day period. No difference between study groups. We did have a pre-specified analysis to look at people who were intubated. And in these smaller group of patients, there was what looked like maybe some signal for a benefit to convalescent plasma, but again, the numbers are very small, so really cannot conclude anything definitive from this. So in conclusion, we didn't see any significant difference between study groups for the primary endpoint. No, I'm sorry. Yeah. So in terms of results and lesson learned, I think we learned several things doing this clinical study. First, I was very proud at our institution. We were able to quickly mobilize a very large team to design and initiate a randomized trial within several weeks and collect a large number of convalescent plasma units from 262 donors and at a single center. And I think this is important because who knows, down the road in the future, there may be some pathogen, some viral illness where people are dying even worse than with COVID and where it may be important to try convalescent plasma again, and I think that our studies showed that a single center like ours can mobilize to do this. I think we also learned that the administration of this plasma increased antibodies, but it wasn't associated with improved outcome. And this is now consistent with many clinical trials that have subsequently been published showing that the administration of convalescent plasma to hospitalized patients relatively late in the course of disease does not appear to improve outcome. And consistent with other studies, as I had mentioned before, most of the patients in our trial, over 80% already had antibodies to SARS-CoV-2. So I think consistent with perhaps a trial by Lifster published in the New England Journal of Medicine looking at early treatment in the first 72 hours, there may be a benefit if you can administer convalescent plasma very, very early, but certainly by the time most patients are arriving at our hospitals, they already have mounted a humoral immune response. And as I had mentioned before, there was a lot of interest by families of patients who are desperately ill, but it was interesting that many people who were not that ill were reluctant to try a blood product transfusion. And I think the EUA by the FDA, while I'm sure it was well-intentioned, I think it hampered enrollment in our clinical trial as well as probably others. And so we concluded it would be too challenging to consent patients for this experimental treatment if there was an FDA-approved treatment once the EUA had been issued. And then finally, there was no simple mechanism at our institution to expand or move convalescent plasma to other geographic areas. It's a bit of a shame we collected all this plasma, but then we weren't able to redeploy it easily to the South and the Midwest areas where once the surge in New York died down, we then saw the surges in other areas of the country. But from a regulatory standpoint, it wasn't simple to think about how we could redeploy our plasma to those sites if there was interest. Well, thank you very much for your attention. There'll be question and answer at the very end of both talks. So at this point, I will hand it over to Dr. Sanchez. Thank you, Dr. Benaguerrero. That was really interesting. We're gonna make a little bit of a change of gears here, and we're gonna be talking about a little bit of our work that just got published in Critical Care Medicine now in July of 2021. These manuscripts, it's focused mostly on our anticoagulation strategies here at the University of Pittsburgh when it comes to patients on VV ECMO. I do don't have anything to declare on this conflict of interest. For those of you that are not completely aware of ECMO use around the world, as you can see here in this graph, it has been a significant increase on the number of ECMO centers. Of course, the biggest surge was after the 2009 pandemic, and I'm sure with the 2020 pandemic, we'll continue to see this happening. In regards of patients with pulmonary failure that might be required to escalate the treatment of their hypoxia, as you can see, the number of patients that we currently trace, it's pretty extense, and the survival has continued to improve both to decannulation and to discharge from the hospital. So that leaves us with a series of aspects of research when it comes to the technology and the method that still need to be approached. And recently, I think this was in May of 2021, the SCOPE study investigators had a modified DALPHI study through a sequence of questions through ECMO experts and profusionists and so on across the world, in which they asked them what they thought were the areas of focus in regards of ECMO research. And as you can see, when it comes to the domain of adverse effect, a major bleeding is still an area of significant interest in our practice. With this in mind, we didn't know about this until recently, but we were working on this publication for some time, so we decided to get our data published, and I'll walk you a little bit through the manuscript that you probably have access to. The design, of course, this is a retrospective cohort study, and it's set in here at the University of Maryland. We just got our Platinum Award as ECMO Center of Excellence, so we're a large-volume ECMO high-referral center. We included patients between 2013 and 2020 that needed to be supported with veno-veno ECMO due to respiratory failure, and to be able to have a little more clarity in our data and don't go the path of previous studies when it comes to the, we decided to exclude the veno-arteriole ECMO patients because we thought that it was just going to make the data a little more difficult to understand. We also excluded patients that, for diverse reasons, were managed without anticoagulation, and we also excluded patients, even though this was a really small number of patients, that had to be transitioned between anticoagulants, and these transitions mostly was on patients that were started on heparin and eventually were changed to Vival because of a high suspicion of probably heat. So, I wanted to share briefly what is it that we were doing in terms of our anticoagulation protocols, and as you can see, they're pretty standard, starting here with our unfractioned heparin protocol. We normally start at 8 units per kilo per hour, and at a rate that will not exceed 1600 units, and then we follow an anti-tan protocol in which we try to keep a level between 0.25 and 0.35, and this is mostly for our nurses to be able to titrate heparin based on this protocol. Once we have reached a level, we normally have two consecutive testing of the anti-tan, and then we just repeat it daily for the next five days and once we reach stability. In regards of our Vival protocol, we follow APTTs, and we start with a relatively low dose, 0.025 milligrams per kilogram per hour. This, of course, is in patients that are non-CRT, and then we titrate our Vival protocol to achieve an APTT between 61 to 75 seconds, and again, we follow daily APTTs in this population. So, overall, when you look at the patients that we included in this retrospective analysis, we included 162 patients in the heparin group and 133 patients in our Vival group, and again, these were only VV ECMO patients, and not significant differences in terms of age or gender, but the biggest thing that you start to see here is the change that we made in 2016. This is when our program decided to start thinking about changing from heparin to Vival, and it was mostly implemented in the following years, in 2017 and 2018. You can see that slowly we migrate to the point in which, at this point, we only use Vival as an anticoagulation strategy in our patients on ECMO. These patients that were on respiratory failure, many of them related to ARDS, we have a pre-VC lung transplant program, so we also included our post-lung transplant patients. We also have a pre-VC thoracic surgery practice, so we included post-thoracic surgery resections that require VV ECMO due to respiratory failure post-operative cleavage, and we do bridge some patients to transplant on ECMO, and so this compromised the study group. When you look at the way these shake up in terms of this was a medical or a surgical admission, of course, the only ones that were included in the surgical admission was the post-lung transplant patients and the thoracic surgery patients. Everything else was within our medical group, so we have a significant number of patients included in both arms. The primary inputs of our study were mostly related to what would be the need of anticoagulation in a patient on VV ECMO, so we thought that one of them was an in-circuit thrombosis. We defined that, but one circuit exchange, and then the counterpart of that would be a major bleeding event, and we utilize at that point the ELSO recommendations, which for those of you that are not aware, is a 2 grams of hemoglobin drop, 20 mLs per kilogram of bleeding, or a transfusion need of more than 10 mLs per kilogram. When we look at both groups, we start picking up on some significant differences. When you look at the in-circuit thrombosis, the percentage of patients that experienced anticoagulation with heparin was around 32 percent compared to only 17 percent of our bioavailability group. We also decided to look at this as the number of days on ECMO. A more prolonged ECMO round will lead probably to a potentially higher probability of a thrombotic event on the patient, so when we look at the in-circuit thrombosis per exocoperm membrane oxygenation day, we also see a significant difference favoring still the use of VIABLE. In looking at the major bleeding events, we see here significant differences in the amount of not only red cell blood transfusion, but also the number of fresh FFP units and platelet pools that all of them favor the VIABLE group. When in regards of the major bleeding event, we saw that our heparin group, it happened in 40 percent of our patients, which compared to only 11 percent of our patients that were on VIABLE as an anticoagulant. This is a little graph of what we saw in terms of the number of days free from a thrombotic event, or what would be a circuit exchange. Again, this happened in 32 percent of the heparin patients and in 17 percent of the VIABLE patients here. Of the patients that require a circuit event, most of that circuit exchange happened as an average of eight to nine days after the patient was placed on ECMO. Because of the retrospective nature of our study and the multiple variables that can be associated favoring one over the other strategy, we did conduct a multivariate cost regression study regarding mostly the odds of developing in-circuit thrombosis. This was based on some general values as age and gender and the admission type. It was surgical or medical, the patients that were on transplant, or if the patient was started on an anticoagulant within 24 hours of being placed on ECMO. Again, you see that the patients that receive heparin as an anticoagulation strategy in their ECMO run had almost a two-fold increase on developing in-circuit thrombosis. We did look also at some secondary outcomes like the length of stay of these patients, how long they needed to be on ECMO for, how long did they survive to the cannulation or the survival post one year. I'm sure that many of these have not only or eventually just related to the anticoagulation strategy, though, you know, massive transfusion and polytransfusion can definitely have an effect on the survival of these patients. And when we look at all these outcomes, of course, many of them favor or all of them favor the use of IVAL as an anticoagulation strategy. So the conclusions of our study are pretty simple and pretty straightforward. You know, the change that we made, it did favor the VIVAL use in this population. We did see a decrease in the number of in-circuit thrombosis and the need for changes on the ordinary circuit. We do see a decrease in the number of our patients and decrease in the number of blood product that this patient needed to be transfused. And one of the things that is included in our manuscript but is in a text form and not really in a graph is that one of the things that we observed in the nursing care of these patients, the VIVAL protocol requires significant less than a nursing intervention because reaching a therapeutic level was much easier and much faster in these patients. As an average, a patient that was on heparin required 48 hours to reach an anticoagulation, anti-10 adequate level, which a VIVAL patient, it only required around 30 hours. So it did simplify our nursing management of anticoagulation in the ICU. And I believe that is it. So I truly appreciate the opportunity to share in all this and I thank you very much for it. And I think Tomas is going to be running the Q&A. So thank you. Thank you very much both for the excellent presentations. I learned a lot as always. And there are some questions coming from the audience. So the first one is going to Dr. Bennett Guerrero. And it is probably the obvious question that do we really need more studies on convalescent plasma after the recovery and the REMAP-CAP trials, sort of refuting effectiveness? I don't think we need any more studies using convalescent plasma late in the course of disease. I think most of this, the data seems to pretty clear that there's no benefit. I think it's still a bit of an open question. I know the initial Lipster trials that was published in the New England Journal of Medicine showed a benefit when it was administered within 72 hours of the beginning of symptoms. Whereas in the recovery trial, and even when they looked at the subset of patients who were very early in the course of the disease, they didn't see any benefit. So I guess that seems contradictory and unclear. So I guess for this disease as well as future diseases, I think really focusing on patients early on in the course of disease is where the money is. Thank you very much. There is another question for Dr. Sanchez, which you have started to allude to in your presentation. Did you have any major issues implementing the protocol? Because you said that the nurses were pretty much on board with it. How did you do this? Well, in all honesty, Tomas, actually, it was a pretty simple transition for us. I think that the biggest concern at the beginning was pricing, mostly regarding the concern of the increased price of Vival. But as we've been working with our pharmacists and the dilution of Vival in smaller volume bags, it comes down to probably around $175 per day to run this protocol. So once the leadership and the rest of the personnel saw that facilitated the pathway of making these our only anticoagulation strategy. And to me, it's so very simple these days to have these patients on Vival. It's just, you know, it's just become very, very straightforward for nursing management, like I mentioned, and to achieve our levels very fast and very consistently. So very little resistance from anybody, if so. Thank you very much. Going back to Elliot, have you looked at, to get any mechanistic insight, how either standard plasma or convalescent plasma might work, or your study was just a clinical effectiveness study? I mean, ours was focused on clinical effectiveness. We did look at neutralizing antibodies with two, you know, the gold standard, as well as a pseudovirus type assay. I mean, I think that convalescent plasma should work in certain diseases. I mean, it makes a lot of sense. Obviously, not everything that makes sense ends up being safe and effective, you know, when studied rigorously. But, you know, I think human bodies are amazing. We are, you know, if you look at somebody who's recovered from an illness, you know, they have mounted an immune response. And, you know, if you give convalescent plasma, you don't need to guess which is the right antibody. You know, the convalescent plasma will have many thousands, many, many different types of antibodies. So I think it's much simpler than trying to develop a therapeutic that's much more narrow, where you really have to focus on a specific, you know, specific, say, antibody, monoclonal antibody, or another approach. So I think it is attractive, but I do think that we need to be cautious, because I think oftentimes it's really hyped. And if you look at, you know, the NIH-funded study for influenza virus that showed no benefit, as well as the current studies, I think we need to be just very cautious before assuming that this is going to work with a new illness that may arise in the future. Thank you. My question is going back to Dr. Sanchez. Obviously, you have shown a very nice difference and even efficacy difference between the heparin and your new protocol. Do you think that this should be trialled in an RCT before it gets adopted more widely? And if yes, do you have any idea about sample size, which would be needed? Thank you. Well, you know, when it comes to clinical trials, I always think that, you know, choosing your endpoint and how easy it is to measure and how big of a sample size you might require. So if you ask me from my own practice, I don't absolutely believe that there is a need for a randomized clinical trial in anticoagulation. I know that the field is pushing in some cases for no anticoagulation at all in patients on VV ECMO. I believe there is something registering in clinicaltrials.gov from the Toronto group that they haven't recruited any patients yet. So, you know, it'd be nice to see more randomized clinical trials on ECMO. That's for sure, not just the EOLIA. But even then, you know, it's difficult to prove a lot of things and including almost 140, 150 patients on each arm. So short answer is for my practice, I don't think I believe in the superiority of VIVO to heparin. But as a scientific community, why not? Okay, can I push you on that a little bit? Because, again, this mortality improvement between the two groups, could it be explained by that, you know, the newer protocol was implemented after you had more experience with ECMO, more runs, more patients, you had maybe a better patient selection, and maybe a different case mix. And that's why you see that improvement. I'm just playing devil's advocate. Absolutely. I mean, we all get better, the more we do it. You know, the only thing that I would say is that this program is probably one of the oldest ECMO programs in the country. ECMO has been part of our practice since the mid-1990s. So we are not new to ECMO, though we know we're always learning. But when it comes to survival to deconvolution and survival to home, absolutely. You know, it's all patient selection most of the time. That's why I get a little worried. We had to include it in our manuscript, but I get a little worried how anticoagulation can improve survival. That I'm not 100% sure. The other question that we have gotten a few times is, well, maybe your transfusion amount of units is related to the change in practice on when it comes to transfusion. You know, one of the also recommendations in patients that are hypoxic is to drive the hematocrit up. So in many of the ECMO patients, they're still receiving several units of blood just to, not because they're actively bleeding, just because you're trying to increase oxygen delivery. I have to say that very early on, around 2013, our program ran a transfusion threshold of less than seven grams of hemoglobin per deciliter. So I don't believe that our transfusion numbers are based on running higher hematocrit levels. I do believe that are related to the major bleeding plans related to anticoagulation. Thank you very much. Switching back to Dr. Bennett-Guerrero, you said that maybe there is a question if we use convalescent plasma early, then we can show some benefit. Now, how could we use that in a much bigger population who are early in the disease? Do you have any opinion that how could we maybe get the population right with biomarkers or with other population enrichment techniques? I think it would only work in a environment that has very good public health measures in place. So maybe it would be more successful in more single payer, coordinated care places versus the United States where we're just so siloed in how we deliver care. I mean, I think you would need early testing to identify people who have the infection. And then you'd have to have a mechanism for bringing people as outpatients into a clinic and screening them and then giving them the convalescent plasma. And I'm not sure how effective that would be. And I think it would certainly be very challenging to do that in the United States where understandably we have a lot of regulations related to giving blood products to patients. So I think it's possible, but I think it's much more challenging than doing what most people have done, which is kind of looking at sick hospitalized patients. Thank you very much. And one of the last questions to Dr. Sanchez is that what was, if there was any crossover from the bivarudin to heparin in your series? So we don't. So once a patient is on vivo, we rarely cross them to heparin. I would say I would say never. We do run them off anticoagulation. And it mostly are post-surgical cases in which, you know, it was a high-risk bleed due to a difficult dig out or something that it might favor surgical bleeding from having a patient anticoagulated. We did have some crossovers from initially in the data set from heparin to vivo when vivo was not a standard of practice. But I'll say what looking back, I was just looking back into that, I think it was less than, it was around 16 patients. So a very, very, very small percentage. Thank you very much. And I would like to thank both of 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. And we would really appreciate your five minutes to complete it because your feedback is very helpful to us. And then a final note, please join us for our next journal club, Critical Care Medicine, on Thursday, August the 24th. And this concludes our presentation today. Thank you and goodbye.

convalescent plasma

COVID-19

randomized trial

SARS-CoV-2

ventilator-free days

mortality

anticoagulation strategies

veno-veno ECMO

respiratory failure