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 November 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. My name is Tomas Zagmani and I'm a Professor of Intensive Care at Cardiff University in the United Kingdom. I will be moderating today's webcast. I've got no financial disclosures. 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 both presentations. To submit questions throughout the presentation, 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. Your feedback is greatly appreciated. Please note 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. Dr. Mauro Panigata was trained at the School of Anesthesia and Critical Care Medicine in Milan, Italy, and completed his research training as a visiting fellow at the National Institutes of Health, Bethesda, Maryland. He has been working as a staff physician at the intensive care unit of Polyclinical University Hospital in Milan since 2001. Dr. Panigata is experienced in ARDS management and veno-venous extracorporeal membrane oxygenation support for respiratory failure. His research interests include ventilator-associated pneumonia and coagulation management in critically ill patients. Our second presenter, Dr. Nick Ingram, received his doctorate of medicine from Creighton University School of Medicine in Omaha, Nebraska. He's currently a pulmonary critical care and sleep medicine fellow at the University of Minnesota in Minneapolis. And he serves as a staff physician at the Minneapolis Veteran Affairs Medical Center and as a house physician in the Fairview Health System. Dr. Ingram's research interests include provider variation and its effect on ICU outcomes and trends in critical care admission diagnosis and related mortality. Thank you, Mauro and Nick, for joining us today. Before we begin, could you each tell us if you have any disclosures to note? Mauro? Hello, thank you for this presentation. I do not have anything to disclose. Thank you. And Nick? I have nothing to disclose as well. Thank you very much. Okay. I will now turn things over to our first presenter, Dr. Mauro Panigata. Good morning. My name is Mauro Panigata. I am a critical care physician at the Fondazione Cagranda Ospedale Maggiore Policlinico of Milan, Italy. On behalf of my co-authors, I would like to thank the Society of Critical Care Medicine for giving me the opportunity to talk about the results of this study. The study was partially supported with the Grifo's Antithrombin Research Award. This is the first polling question. Do you use heparin to anticoagulate patients on extracorporeal membrane oxygenation? Yes. No. I don't know. Please select one option. Okay. We see that all respondents selected yes. So everybody uses heparin to anticoagulate patients during ECMO. Next polling question is, do you routinely supplement heparin therapy with antithrombin during ECMO? Yes. No. I don't know. Please. Okay. Okay, now we have 83% of the respondents answered no and 17% of the respondents answered yes. During extracorporeal membrane oxygenation, anticoagulation is required to prevent thrombotic complications due to the contact of blood with the large surfaces of the extracorporeal circuit and the unfractionated heparin is the anticoagulant of choice because of its short half-life, because it has an antidote, rotamine, and because it can be used in case of renal failure. But we should remember that heparin is an indirect anticoagulant. That means that it has no intrinsic anticoagulant activity but requires antithrombin to anticoagulation. In fact, antithrombin without heparin neutralizes coagulation enzyme in a slow and progressive manner but when it complexes with heparin, it increases its anticoagulation activity by 1,000 times and it irreversibly inhibits mostly factor Xa and factor IIa, that is thrombin. When antithrombin is consumed because large amounts of heparin are used or it is leaked in the extracellular space due to sepsis or less antithrombin is produced because of a spalanchnic hyperperfusion, for example, as it frequently occurs in critically ill patients, a phenomenon called heparin resistance can occur. That is basically the reduced ability of heparin to inhibit thrombin and fibrin formation. And this is a specific concern of patients on ECMO since antithrombin activity is commonly decreased during ECMO because it is consumed by heparin. Although the guidelines suggest supplementing antithrombin during ECMO only when its deficiency coexists with heparin resistance, antithrombin supplementation is a current practice. We recently conducted an electronic survey among ECMO centers around the world and we obtained responses from more than 50 countries and we found that the majority of the centers use the heparin as the anticoagulation of choice, anticoagulant of choice, and the majority uses the APTP. But interestingly, antithrombin was routinely monitored in 48.7% of the centers and routinely supplemented in 38% of the centers and those percentages rose to 80% in case of pediatric patients. In this frequency distribution, you can see the antithrombin activity triggering supplementation that is below 60 and 100%. Interestingly, we also found that there was uncertainty regarding the practice of antithrombin supplementation because we observed conflicting results when we asked the ECMO experts and the centers coordinators to justify the routine supplementation of antithrombin and the reason that was most commonly reported to justify prescription of antithrombin was that it helps achieve and maintain anticoagulation target, but at the same time, the most commonly reported reason for not supplementing was that anticoagulation is not usually an issue. We should remember that antithrombin supplementation is approved only for the treatment and prevention of thromboembolism in patients with congenital antithrombin deficiency, which is far less common than acquired deficiency that occurs, for example, during disseminated intravascular coagulation, trauma, birth, complicated pregnancy, ECMO cardiovulneral bypass, or thrombosis complicating orthotropic liver transplantation. A review of the U.S. Pediatric Health Information System database covering 4,200 admissions found that 97% of the children who were treated with antithrombin received it off-label with neonates comprising the largest population and, interestingly, the extracorporeal membrane oxygenation was the most common procedure associated with antithrombin use. So, we have seen that antithrombin is supplemented during ECMO, but is there any evidence? Well, most studies that are historical and mostly pediatric report that antithrombin supplementation increases antitenae levels and, of course, antithrombin levels, but results on the effect of antithrombin supplementation on heparin requirements, bleeding thrombosis, and ECMO circuit lifespan are discordant. So, we decided to run a study to answer a question. Does antithrombin supplementation to achieve and maintain normal level of antithrombin compared to no supplementation of antithrombin at all decrease heparin requirements during veno-venous ECMO for respiratory failure? And this study was a prospective, randomised, single-blind, multi-centred, controlled, two-arm trial that was performed in two Italian ECMO reference centres. The coordinating centres is the Fondazione Cagranda in Milan and the participating centre, the ISMET Centre in Palermo. It received the ethical approval of the respective ethical committees and the approval of the Italian FDA in 2017. All consecutive patients admitted to participating centres requiring veno-venous ECMO as a support for respiratory failure were included, and the exclusion criteria was pediatric population and control indication to heparin use or a patient connected to a veno-arterial ECMO. Patients were randomised in a one-to-one fashion in the treatment group that received antithrombin supplementation to maintain a functional antithrombin level of 80 to 120% or a control group that did not receive any supplementation at all with a safety rule that was that if the patient required more than 35 units per kilogram per hour of heparin and its corresponding level of antithrombin was below 50% this patient could receive antithrombin supplementation but just enough to reach antithrombin level higher than 50%. This was the supplementation diagram of the study in the treatment group. At the moment of the cannulation of ECMO the patient received a loading dose of antithrombin according to his body weight and then a maintenance dose was administered daily based on daily antithrombin measurement. We had an anticoagulation protocol with heparin and at the moment of ECMO cannulation the patient received according to his baseline APTT a bolus of unfractionated heparin and then a continuous infusion of 18 units per kilogram per hour. We also had a heparin adjustment algorithm to titrate heparin administration during the study and we followed the ACT during the first 12 hours of ECMO support basically because it was easier during patient transportation because the two centers are ECMO referral centers and using a portable device during transportation was easier and from the 12th hour on we relied on the centralized laboratory APTT to maintain an APTT ratio between 1.5 and 2. We also had the rules for blood product transfusion during the study. We kept hemoglobin higher than 10 grams per deciliter, platelets higher than 50,000 and fibrinogen higher than 150 milligrams per deciliter. Since we know that ECMO circuit coagulopathy affects heparin management and anticoagulation management we decided to standardize the ECMO circuit replacement based on a circuit changeout score based on coagulation parameters platelets fibrinogen and the dimer and also the circuit ECMO circuit was changed when the patient had hemolysis or the circuit itself had the dysfunction. The study ended with a patient who was disconnected from ECMO or study was discontinued prematurely when one of the following condition occurred death course onset of HIT heparin induced thrombocytopenia or unexpected circumstances that prevented proper heparin or antithrombin administration. Antithrombin supplementation was stopped at ECMO discontinuation and the follow-up of patient ended at the ICO discharge or death whichever came first. The study endpoints were primary the total amount of heparin infused to each patient during the treatment with ECMO calculated as the total amount of heparin infused divided by the patient actual body weight and the individual duration of the study and secondary endpoints were anti-factor 10a levels the incidence of bleeding episodes the cumulative incidence of any bleeding the number of products blood products transfused and the incidence of patients of circuit thrombosis checked with the compression ultrasound after 24 hours of ECMO removal and the rate of extracorporeal circuit thrombosis defined as the time to first replacement of ECMO circuit. The sample size was calculated on preliminary data from patients undergoing ECMO treated with the same heparin titration protocol and assuming an estimated difference of five units per kilogram per hour between the two groups a sample size of 40 patients 20 in each treatment group was calculated. This is the flow diagram of the study we assess for eligibility 72 patients randomized 49 one received the wrong was erroneously enrolled because of pre-existing exclusion criteria and 24 patients were analyzed in the control and 24 in the treatment group. These are the baseline characteristics of the patients enrolled in the study in the two groups that were well balanced between the two groups except for a trend towards higher antithrombin in the control group compared to the treatment group same dick scores and severity of these patients. In this slide you can see the mean antithrombin heparin and apt-t during study days in the upper panel you can see the mean antithrombin in the red one is the mean antithrombin in the treatment group and in the blue one is the control group and you can see that they are well separated along the study days instead the middle panel the mean heparin and the lower panel the apt-t they are similar in the two groups. In fact primary and secondary endpoints were not different in the study the primary endpoint that was the heparin was not different in the control and in the treatment group with a median heparin of 15 units per kilogram per hour in the control and 13.5 in the treatment group with a mean difference of minus 1.2. Secondary endpoints were not different the antithrombin were not different bleeding and cumulative incidence categories thrombosis and blood product transfused and the circuit changes were not different in the two groups. Also you can see in this slide that laboratory tag and hemodynamic and ECMO parameters throughout the study were not different in the control and in the treatment group is except for antithrombin of course that was supplemented in the treatment group and reach a median of 109 percent and 84 percent in the control group and except for activated clotting time during the first 12 hours that was higher in the treatment group. There was no difference in the causes of premature discontinuation of the study that was the most prevalent was intracranial hemorrhage. We also conducted a competing risk regression analysis with the competing events premature discontinuation of the studies and of the study and death death before ECMO discontinuation and we found that receiving antithrombin actually reduce the probability of changing the first ECMO circuit. The duration of ECMO and other clinical outcome was not different in the two groups and the duration was 16 days in the control and 15 in the treatment groups and also the ICU length of stay was not different and the survival at the 28 days was not different in the two groups. We also conducted an exploratory subgroup analysis to investigate the effect of antithrombin supplementation on heparin when antithrombin was indeed low and we found that when antithrombin was below 80 percent the estimated reduction of heparin was 2.79 units per kilogram per hour and when antithrombin level was below 60 percent the estimated reduction was 7.36 units per kilogram per hour. So concluding, we know physiologically that the anticoagulant effect of heparin can be achieved only in the presence of antithrombin but probably and probably borrow from the cardiopulmonary bypass experience and many ECMO patients receive antithrombin probably based on the preconcept that managing anticoagulation is easier when antithrombin level is normal. However, this practice is not supported by much evidence and due to the results of the study we can say that until future studies will explore more the efficacy of antithrombin administration especially when the plasmatic level is low the practice of routine antithrombin prescription during veno-venous ECMO should be questioned. This was my last slide and I want to thank you all and my collaborators also that made the completion of the study possible and I turn the screen over to Dr. Nick Ingraham for his presentation. Thank you. Thank you so much. That was wonderful. So good morning. My name is Nick Ingram. I'm a pulmonary critical care fellow at the University of Minnesota and on behalf of my co-authors I would like to thank the Society of Critical Care Medicine for giving me the opportunity to speak to you guys today about trends of functional status deterioration in the critically ill. We have no conflicts of interest to declare and funding was provided by University of Minnesota's Critical Care Research and Programmatic Development Program. So determining quality critical care is a topic of controversy nowadays. Intuitively mortality has been used and appropriately so for a long time. Fortunately through research we've been able to make great strides in overall mortality over the past few decades throughout the world. Different studies have seen mortality decrease in the ICU by over one-third in the U.S. and the U.K. Absolute reductions of mortality range anywhere from four to eleven percent during these time periods which reinforces the importance of what we do with research. The success of our mortality improvements also brings new challenges in determining what is quality critical care. The lower our mortality rate goes the less useful this measure may be in evaluating interventions or ranking quality critical care delivery and furthermore we are keeping more patients alive. We need to put more effort into ensuring that the lives saved also have quality. So for a long time outcomes have been binary in medicine especially in critical care sick not sick life or death and prior improvements have made new challenges with an ever higher bar of success and so one of the blogs I like to follow by Josh Farkas and others has a really nice figure seen here which he was nice enough to allow us to use today and it describes the difficulty of determining an intervention and its effectiveness. So on the left you have outcomes that are easier to measure and more ability to prove causality. This starts to dissipate as you move to the right but on the bright side moving to the right becomes more important to the patient with this in mind we were wondering is there an outcome that would be patient centered and prevalent enough to warrant further investigation. So over the last few decades the United States specifically has seen an increase in ICU beds disproportionate to hospital beds and like we had discussed over the same time period inpatient mortality has been decreasing. So with an increasing patient base and more survivors our ICU population is more fluid than ever and this inherently brings new challenges which include a larger illness spectrum more survivors and new comorbidities and new outcome goals. One of those outcome goals that has increasingly gained attention is functional status and its importance spans the entire spectrum of a patient's health care interaction. For example let's take baseline functional status and this has been found to be associated with short and long-term mortality quality of life and discharge location. In fact decisions to withhold or withdraw care by physicians has been found to be associated with baseline functional status. Other studies have actually found that baseline functional status is actually a better predictor of mortality than the timing of antibiotics and age. Now once the patient is hospitalized regardless of their baseline functional status there is still an association with functional status at discharge and mortality. In fact one of the studies found that if a patient had improving functional status prior to discharge it was associated with a decreased 90 day mortality. Now despite these associations there are still many unknowns as to why some patients are able to bounce back others are not but what we do know is that poor functional status at discharge is associated with readmission and poor quality of life. Increasingly over the past decade we have more survivors of sepsis who were previously productive members of society. We also know that nursing homes and rehab units they result in permanent health care costs specifically when we compare those to patients that were able to go home. So some argue that to deliver cost-effective and patient-centered care that we should consider outcomes other than survival. These include quality of life and neuropsychological function functional status and so we chose functional status in this study and another important aspect to keep in mind is that despite the fact that our ability to predict survival is pretty abysmal our ability to predict functional outcomes is even worse. So with this in mind we sought to further characterize functional status outcomes in the critically ill patient population in the United States. So this brings us to our first polling question of this talk. So does your institution have a dedicated protocol to identify and routinely deliver early mobility and or exercise to ICU patients? The different answers are yes, b, I think so, c, I'm not sure, d, I don't think so, and e, no. That's pretty good. And it actually tracks with some of the data that we're seeing is I think as especially the ABCDEF bundle is becoming more prominent in ICUs as we gain more research. It's good to see and it's reassuring to myself that people are starting to have dedicated and implemented protocols which is wonderful. So it's good to see that. So getting back to our study. So to accomplish this we use the Cerner-Apache database. It's a national database with over 200 hospitals. Data is manually abstracted from trained individuals and this occurred from 2008 to 2016. We included all adults with a documented functional status on admission and discharge and this included about 200,000 patients across 90 hospitals. We did exclude patients with a primary surgical admission, hospitals with less than 100 patient encounters, as well as patients that died. Our final analysis included patients just under 130,000 patients. So spanning across the 90 ICUs we split the admission diagnoses into 11 disease categories which I'll show in a second. We used quasi-poisson regression to identify temporal trends and all of these models were adjusted for age and Apache score. Our primary outcome was assessing functional status deterioration by disease. Data for functional status again was manually abstracted from the EHR and categorized patients into either independent, partially dependent, or fully dependent. And like I said this was done on admission and on discharge. Our secondary outcome included looking at temporal trends along with looking at the severity of the deterioration. This slide here is just kind of showing you where admission diagnoses were categorized into the 11 disease groups. So before we get to our second polling question for this talk I wanted to show kind of a visualization of the binary outcome that we used for functional status deterioration. Again we have three designations for functional status that were recorded on admission and discharge. And for those that remain in the same functional status group or if they improved they were put in the no change group. I should note briefly that less than 1% of patients had an improvement in their functional status from admission to discharge and thus we did not consider this to be drastically different from a no change group and so we included them together. On the right we see the different ways that a patient could have a decline. So this brings us to our next polling question. So what percentage of ICU patients have functional status decline from admission to discharge? A, 5% B, 20 C, 30 D, 40 or E, 65%. So it's actually about 30%. I think it's interesting that a lot of people thought a little bit more so and I think that's important and that'll be something we could probably talk about in the Q&A because the granularity of the data probably missed some patients that had a clinically important functional status decline but the way we set up our variables was very specific to ensuring that we knew that the functional status did decline during the admission and we focused more on kind of the course comparison. But in fact in our data we did show that 30% of patients have declined in their functional status on hospital discharge when comparing to their admission functional status and to further distinguish between the severity of functional status decline we developed an ordinal variable. We parsed out those with functional status decline into three different groups which you'll be able to see here. So now within the patients that experienced a decline in their functional status the majority is actually coming from those that are independent on admission and then became partially dependent. But strikingly of those with a functional status decline 20% of patients had a drastic change in the way they were previously independent and now they've actually become fully dependent. So a broad overview of the patient population that we have in our data set. It was a median age of 63 years old mostly white and male patients. Median ICU length of stay was two days and the average Apache score was 47. Over 50% of the population had Medicare followed by private and Medicaid for insurance. 80% of the patients were independent on admission however 55% were discharged fully dependent. Here are the 11 admission categories and their distribution and what we can see here is that cardiac pulmonary neuro and sepsis admissions make up about two-thirds of the admissions. Next we'll take a brief look at the hospital demographics and on average hospitals in the study had over 3,800 ICU admissions each year 570 beds. Main contributors were from the Midwest and the South and around 45% of the hospitals were large teaching hospitals. Just to give you an idea so in univariant analysis patients with functional status decline comparison to those without a decline these patients were older more likely to be female and white. On admission they were more likely to be independent but more dependent on discharge. They were more likely to be DNR and were sicker and patients with a decline in their functional status also had a longer length of stay and lastly they were more likely to be admitted with a neurological trauma septic related disease. So for the first analysis we looked at functional status change over time by disease category. So for this model we used a quasi-poisson regression model and you can see the incidence rate ratio on the x-axis and categories on the y-axis. Anything above one is indicating an increase over time as this is the incidence rate ratio for the year variable in the model. So what we can see here is that every disease category had an incidence rate ratio above one along with the overall model which showed an 15% per year increase. The highest ratios for the disease models included GI, toxicology, and cardiac. Next we performed a similar modeling but this time using the ordinal regression which we discussed earlier. Again using year to evaluate temporal trends of the severity of functional status decline over time. So as we what we saw with the binary outcome the overall severity of functional status decline is increasing and this is seen specifically with septic and hemonic admissions. The only exception to this was non-surgical trauma which was found to be decreasing in the severity of functional status decline. Next we wanted to evaluate if there was a difference between functional status severity among the admission diseases themselves. And so there's 71 admissions here which I know that nobody can read but we'll talk more specifically about the outliers in the top and the bottom groups. And so for this ordinal regression was completed again looking at the severity of functional status change in these groups and for the analysis we assume that the variable is continuous but also worse as it changes. And the odds ratios here were based on the average risk of admission. So when we look at the top five diagnoses we find a recurrent theme with many of the same bad actors. We see trauma and neuro and not only are we seeing increasing rates of functional status decline but many of them are also associated with increases in severity and severe functional status decline. Now interestingly when we compare the top five and bottom five diagnoses and kind of take the mean of their odds ratios we notice that there's a nine-fold difference in functional status decline. And then if we look at the actual top and bottom there's a 17-fold difference in the odds of functional status decline. And I think this just speaks to the breadth of the heterogeneity of severity of illness that we see in the ICU and kind of speaks to the population. So finally we put the diagnosis based on severity into different quartiles. The figure displays the percent of the diseases from each body system based on the quartiles into which they fell. So again trauma has the highest proportion of diseases that fell into the highest quartile which is associated with the highest severity of declining functional status as would be expected. So I think there's some important limitations that we need to consider with the study. First of all it's a retrospective study. The data may not be representative of all ICUs. Importantly the database is from a benchmarking program so this could lead to selection bias given the fact that hospitals had to participate in this program. Performance status was manually abstracted however it was pulled from the documentation and it was not recorded with a validated functional status tool which would have been ideal. We did not consider death in the ordinal outcome and thus these findings must be considered in the context of patients that are going to survive their ICU admission. Furthermore post-operative patients were excluded and these data should not be extrapolated to this patient population. And finally we could not determine temporal trends for the specific admission diseases due to model conversion issues and given the fact that we wanted to take into account hospital differences. So in summary we noted that functional status deterioration is rising overall and in all of the subgroups to upwards of 200% over the time period in some categories. Further analysis found that there's up to a 17 fold difference in the risk of admission resulting from functional status decline on discharge. Neuro pulmonary trauma and GI disease categories were the most common to be in the top quartile for both risk and aside from trauma these were all increasing over time for severity. And finally we assess the severity of the deterioration after ICU admission and as with the increases in the worst functional status in our binary outcome the severity as determined by the ordinal outcome is also increasing overall and within each subcategory except for trauma. So with the increasing decline and market variation and functional status deterioration these findings support the need to adjust how we determine well what is quality critical care. Mortality will continue to be used as an important target but the more we continue to have lower mortality it's going to be harder to use that and we need to focus on other outcomes. By targeting cognition functional status we can improve the quality of life of our survivors while also starting conversations to change how we measure quality care delivery when evaluating institutions as well. So next steps are already underway and they're trying to focus on survivorship with post ICU clinics and mobile rehab units and these are crucial to not only treatment but also gathering more information on the population of patients that are now surviving at their ICU admissions. So I thank you guys for your time and I appreciate any questions concerns and critiques and now I'll switch it over to for the Q&A and I'll turn it back over to Tomas who will moderate. Thank you guys. Thank you for both of the presenters for the excellent talks and there are a couple of questions from the audience so I will start with the first question for Mauro. You have alluded to this both in the presentation and in the paper that what do you think what is the main reason for the lack of efficacy? Is that because the centers already practiced a little bit of less is more so they reduce the heparin exposure for the patients? Thank you for this very interesting question. Actually I think that the main reason for the lack of efficacy is that even in the control group the antithrombin level was almost normal. Well it was a median of 80 84 percent so it was normal. So this probably explains the results of the study and one explanation is that during veno-venous ECMO we are not administering so much heparin. It's not like in cardiopulmonary bypass where there is a physiologic rationale for supplementing antithrombin because of heparin resistance. Here instead the supplementation is not actually really physiologically, with a physiological rationale supplemented because if you don't have like a low level of antithrombin it does not make sense actually. I think this is the main reason why the lack of efficacy but interestingly the exploratory analysis that looked for the effects in patients who had indeed low antithrombin level we saw a signal towards lower heparin administration in patients with low antithrombin level. This is the reason why in our center for example we switched our practice to supplementing antithrombin only when antithrombin level is below 60%. Thank you very much. Nick, there is a question about the diagnosis which were associated with functional decline. How do you explain that the pulmonary and renal sepsis as well as the bacterial pneumonia had lower odds for the functional decline whilst there is a ton of literature which suggests that these are the groups who are at highest risk of developing complex post-ICU syndrome, complex PTSD and all sorts of other problems. So do you have an explanation for that because the odds ratios are really high for trauma and especially for head injury as you would expect but not so for sepsis. That's a great question and I think one of the important things to consider specifically with this data set is that one of the benefits of having everything manually abstracted is you don't have to clean up the database as much because there's somebody actually going through and doing that. So from a logistics level and from the health services researcher type perspective that's nice. But one of the things that then causes a limitation and we've actually seen this with some sub analysis that we've done that you have to consider when somebody has an admission diagnosis of bacterial pneumonia or renal sepsis or pulmonary sepsis that probably means that that was a clear diagnosis within the first 24 hours or what have you. You'll notice that there are some other disease categories to which patients were put into which may be pulmonary NOS and so that there's these more kind of vague disease categories which if you look are more higher up on the odds ratio list and so I think it actually speaks more so to did they identify what was truly going on with the patient and were able to really get source control or specifically identify that within the first 24 hours versus the other person who may have remained kind of this vague very ill patient that they couldn't quite put their finger on it and be confident to say this was their admission diagnosis. And that's something that we've actually seen come out of data when we've actually looked specifically at the sepsis subset in and of itself and this isn't published data yet but we actually saw that wherever it was an NOS or versus something that was specifically an organ that was identified as the etiology of the sepsis when the etiology was identified and categorized as pulmonary sepsis they had a lower mortality rate and so that's where I believe it comes from but again without kind of the ICD-9 or ICD-10 codes to confirm that I can't say for sure but you know it's a great question. Thank you very much. Back to Mauro and we are sort of diverging from the trial itself but you know you cannot go and look at a study nowadays without trying to make sense how it would affect our treatment plans in COVID-19. We know that in COVID-19 micro thrombi are implicated in the pathophysiology. Do you think that in COVID-19 ARDS and especially ECMO supported COVID-19 ARDS anti-thrombin supplementation could provide benefit? Do you have any data or any experience? Thank you Tomas for this very interesting question. Well actually I don't personally have so much data but the literature around is emerging and showing that in this patient, COVID patient, anti-thrombin is not really low. So it's almost normal except for a rare very rare case of disseminated intravascular coagulation. So I don't think it's a very different setting speaking about anti-thrombin from the ECMO settings, non-COVID ECMO settings, unless we are using a very high dose of heparin because since these patients, the COVID patients, are very inflamed and they have very high levels of fibrinogen and factor VIII. So they require a lot of anti-thrombin to reach the target, the anti-coagulation target. So in this case they can consume anti-thrombin. So this could be a situation in which anti-thrombin should be supplemented. However actually we didn't see in our experience in our centers with the COVID ECMO patients a big difference in anti-thrombin levels with non-COVID ECMO. Well however the anti-coagulation issue in COVID is of course extremely interesting. This is the reason why there are I think almost more than 20 trials on anti-coagulation in COVID and we will soon have an answer I hope on which is the best practice. So thank you very much for this interesting question. Thank you. I'm switching back to Nick again. Did you have any data if the functional status decline is present on ICU discharge or is it developing post ICU or is there a proportion of patients who are moving from one category to another while they are discharged from the ICU but still in the hospital? Well the short answer is I wish. No we don't have that data. They specifically have the hospital discharge functional status but this brings up you know an amazing point that really needs to be further investigated because I think you know especially with the implementation of the ABCDE and the F bundles we we know that there's you know improvements to certain patient outcomes from that standpoint but one of the important things especially when we look at long-term outcomes after they leave the hospital it's unclear what gets diluted out as far as what goes on when they get to the floor at least in my limited experience I don't know that anything that we do ABCDEF wise is really transitioned to the floor so much you know whether physical and occupational therapists are really talking and communicating and maybe there is maybe there's some protocols on the floor that are synergistic to what we're doing in the ICU and and that really needs to be investigated in further projects so that's a really really good thought. Thank you. Mauro if if we think that you know heparin is the gold standard and antitrombin is not really influencing the clinical outcomes what do you think about other anti-coagulation methods one of the newer DOACs and NOACs for instance or Gottroban do you have any experience would they be better especially in the in the heparin resistant cases? Thank you this is another very great question well actually yes we have some experience on the use especially of Gottroban and but in Italy is licensed or we can say the own label use is just in case of HIT so we don't have so much cases on that but when we use our Gottroban the stability of anti-coagulation is much higher so I think the use of indirect thrombin inhibitors because Gottroban like by valirudin does not require antithrombin to inhibit thrombin so this makes much more stable the management of anti-coagulation because it does not rely on other enzymes and also well it does not it is a reversible inhibition not irreversible like the heparin and antithrombin complex with other coagulation factors so I think this is very interesting however well there should be well at least we should run some trials on that to see the real efficacy because there are some cavities because well they have a long half-life therefore for example our Gottroban well should be used with caution in patient with hepatic failure and also by valirudin in only well we should take care in patient with renal failure so they are actually interesting drug but I don't know if they are very safe we currently have a lot of experience of heparin and this is the reason why it makes so reliable and attractive in the management of anti-coagulation during ECMO. Thank you. Thanks Mauro and I think a question leading from from that is that you showed some data on on the TAG in in the trial do you routinely monitor TAG parameters on ECMO or was it just for the for the trial? Well no we routinely monitor TAG and this is one of the means that we have to check the anti-coagulation level well this in this trial we relied only on the APDT before because it was well to conduct a trial we should uniform and standardize between the centers the management of anti-coagulation but we currently use the TAG for managing anti-coagulation during ECMO especially we conducted the trial that was published a couple of years ago on the management of TAG anti-coagulation with the TAG and it was a safety trial and we found that using TAG allows a lower amount of heparin to be administered during ECMO and this lower amount of heparin was not associated with the higher risk of thrombosis neither a higher risk of bleeding so this were interesting results that well they need or they merit some further studies and I think we will do something in this trial we found actually that the TAG the R the reaction time of TAG is well time to the initial fiber information was longer in patients with antithrombin but it did not reach the statistical significance but there was a trend towards a longer reaction times in patient with antithrombin and the TAG is more sensible than the APDT so probably patients receiving antithrombin they had a slightly higher level of well level anti-coagulation at least the efficacy of heparin was higher so this is a very interesting field thank you for this question. Thank you and I would like to thank both our presenters and our audience for attending everyone who joined us for today's webcast will receive a follow-up email that will include an evaluation so please do take that five minutes to complete it your feedback will be much appreciated and on a final note please join us for our next journal club critical care medicine on Thursday January the 28th at 1 p.m. Central Time this concludes our presentation today. Thank you very much. Thank you.

Journal Club

Critical Care Medicine

antithrombin supplementation

extracorporeal membrane oxygenation

functional status deterioration

CERNER-Apache database

neuro diseases

pulmonary diseases

patient outcomes