Hello, good morning, and welcome to Year in Review. My name is Lyle Geraghty. I'm an anesthesiologist and intensivist at the University of Vermont, and I'll be moderating the Year in Review session for anesthesiology, as well as the Continue the Conversation session to follow. A bit more about that, the Continue the Conversation session following this one is meant to take the place of a question and answer session or discussion that would occur at a live event. So the hope is to give participants a chance to interact in this virtual setting like we would at a live conference. It'll be an informal discussion conducted over a video conference to give folks the opportunity to discuss some of the topics raised here in the Year in Review. So I'll lead that discussion with these faculty, which I'll introduce shortly, and attendees can ask questions and add to discussion or submit questions via the interactive chat, and I'll make sure that they get asked to the appropriate faculty. Dr. Philip Sommer is a clinical assistant professor of anesthesiology and critical care at NYU Langone Medical Center in New York City. I suspect he took care of a lot more COVID patients than I did and than many of us did this year based on his location in New York, and I look forward to his talk on mechanical ventilation and COVID-19. Hello, my name is Philip Sommer, and I'm an anesthesiologist and intensivist at NYU Langone Hospital. I'm here today to talk to you about the respiratory management of patients with COVID-19 ARDS. I have no financial disclosures to make. Some of the lecture objectives are when should mechanical ventilation be initiated, learn if noninvasive ventilation should be used, is COVID-19 ARDS different than classical ARDS, how to personalize mechanical ventilation in different phenotypes, and adjunct therapies in mechanical ventilation. During the past year, the world has become familiar with acute respiratory distress syndrome and its role in COVID-19. Critical care physicians have encountered ARDS since its first description in 1967, and even though we have decades of experience seeing it, there are still many unknowns. In the past six months, any critical care physician across the world has seen what most would have seen in a lifetime, and I believe that it is this volume of cases that has led to the question of, is COVID-19 ARDS the same as classical ARDS? This question is, of course, only the beginning of the discussion, as the answer to this guides therapies, morbidity and mortality estimations, and resource utilization. On my first shifts in the COVID ICU, I would walk through the floor, seeing patients on phones, sitting in chairs, watching television, and assume that things were fine. Then a quick look at the respiratory monitor would show saturations of 70%, but little to no patient distress. This scenario was seen around the world, and the term happy or silent hypoxia arose from this. As the reasons we'll discuss later, many of these patients were intubated, but that may have been proven to be incorrect for some of the following reasons. Hypoxemia alone rarely is an indication for intubation, but rather increased work of breathing, which is usually based on the physician's judgment, analysis of vital signs in labs, and the physical appearance of the patient. Isolated hypoxemia is usually able to be treated without the need for intubation. These two figures by the COVID ICU group for the RIVA network show that with improved understanding of physiologic principles, the rates of intubation have decreased, and with other therapeutic options available, there have been a simultaneous decline in mortality. Understanding some of the principles of hypoxia may prevent unnecessary intubation and improve the overall survival for patients diagnosed with COVID-19. How do our respiratory centers respond to different stimuli? These two figures show the response of the respiratory system over a variety of PaO2 and PaCO2 levels. As you can see on the figure on the left, small increases in PaCO2 levels cause rapid increase in minute ventilation. These increases are not well tolerated in terms of patient comfort and dyspnea. In comparison, the PaO2 response curve on the right does not begin to rise until the PaO2 drops below 60. This means that at relatively low levels and a wide range of values, patients will remain comfortable and free of dyspnea unless there is a concurrent elevation of PaCO2. Blunted hypoxia truly exists when there is a low PaO2 in combination with elevation in PaCO2, which should elicit a large response in minute ventilation, but instead this response is blunted. And for these patients, they are at high risk for respiratory complications. You may be asking, what is the mechanism behind this blunting of respiratory systems in the hypoxia? Some known reasons include that older patients have blunted respiratory centers as well as patients with diabetes. As you have seen in multiple case descriptions, the most critically ill COVID patients and the ones that have the highest mortality are older patients and those with medical conditions like diabetes. This altered respiratory response may play a role in their worsening outcomes. Another possible mechanism of this is that the SARS-CoV-2 virus utilizes ACE2 receptors throughout the body to infiltrate cells and have its widespread effects. Pertinent to this discussion is that ACE2 receptors located on the carotid body can allow viral entry and potentially affect the chemoreceptor's response to oxygen levels and potentially blunt them as well. Fever causes a rightward shift of the O2 dissociation curve. And that means that for any given PaO2, the SaO2 will be decreased, creating desaturation on the monitor, but not dyspnea, because chemoreceptors in the body are only going to respond to PaO2 and not the saturation. Certain conditions such as critical illness, skin pigmentation, nail coloring, and just low oxygen levels can falsely lower SpO2 readings. Understanding the body's response to hypoxia is critical in determining whether or not intubation is necessary. As I stated before, hypoxia alone should not be the sole factor for intubation, but rather an assessment of the patient's work of breathing will be more accurate. While we start to gain understanding of the mechanisms behind this severe hypoxia, the question of how to best treat this is next. If one were to walk through a COVID care report now versus in March of 2020, there would be some significant differences. The evolution has followed that in March, scenario one of the intubated sedated patient dominated the census numbers, while only a few patients were utilizing noninvasive positive pressure ventilation as in scenario two. Today, there are entire floors of patients on BiPAP, CPAP, and high flow nasal cannula. This change to embracing the use of these modalities comes with the questions of, is this treatment safe for healthcare workers, and is it efficacious for our patients, or are we just delaying the inevitable? Given the concern over the aerosol spread of COVID-19, initial efforts were made to reduce any procedures or techniques that were thought to be aerosolizing. This is where the idea of early intubation came through. The NIH, SCCM, ATS, and WHO created guidelines stating the use of noninvasive positive pressure ventilation in COVID is an acceptable modality for treatment. Still other societies favored early intubation mainly for the concern of the risk of the healthcare worker infection. In 2012, a meta-analysis by TRAN was performed showing not only did high flow nasal cannula not carry an increased risk of transmission, but that it may have actually been protective and that intubation itself had a higher rate of transmission of respiratory infections to healthcare workers. This means that if high flow nasal cannula could prevent intubations, fewer healthcare workers would be exposed to infectious particles. Ferrioli conducted a study using laser smoke visualization to track the distance of dispersion of aerosolized particles. As you can see, they found that regular nasal cannula dispersed particles the furthest, while CPAP had a negligible dispersion and high flow nasal cannula at 60 liters was similar to breathing oxygen on a non-rebreather. This study helped show how safe noninvasive positive pressure modalities can be for healthcare workers. A study performed in France by Guy et al., specific to COVID-19 patients on floors that exclusively use high flow nasal cannula found that of the 44 healthcare workers who worked Only one nurse tested positive and it was from a presumed domestic infection source. The use of noninvasive positive pressure ventilation has been demonstrated to be safe for healthcare workers, given a few key points. Appropriate PPA is available, BiPAP and CPAP masks are tight fitting, and that patients wear surgical masks in the room when healthcare workers enter to prevent droplet dispersion. Now that we have deemed the use of noninvasive positive pressure to be safe, we must investigate its efficacy. While it is too early to say for certain whether or not noninvasive positive pressure is effective in COVID-19 patients due to the lack of clinical trials, case series and meta-analysis are beginning to show that the use of high flow nasal cannula helped prevent intubation in COVID-19, including the study by Guy et al., who looked at the 34 consecutive patients treated with noninvasive positive pressure ventilation and successfully weaned 70% of them without intubation. While we are lacking these large studies, we can use other information from previous experiences like MERS and STARS and influenza to help guide our treatment options. A 2013 study looking at noninvasive positive pressure ventilation in the H1N1 outbreak showed that 41% of 177 patients treated with high flow were able to remain extubated, much lower than the intubation rate that was published. Notably as important is that of the patients who were intubated after failed high flow nasal cannula trials, their mortality remained the same, indicating that they did not miss a critical window or inflict harm by not performing an early intubation, although that result may be biased due to study design. When looking at effectiveness, obviously we would like to see mortality benefits, but the benefits of noninvasive positive pressure may come in indirect ways. By keeping patients off ventilators, we create more resources for other patients. These patients who are not intubated require less sedation, less muscle weakness, and possibly shorter length of stays, all which are important as resources become scarce in the pandemic. From initial reports across the world in the United States, intubation rates of patients going to the ICU were sometimes upwards of 80% at the start of the pandemic. The global scale has led to thousands of patients being placed on ventilators and observations made that had not been seen before in ARDS. Reports from Italy began to emerge of an ARDS phenotype that met Berlin criteria but had significant dissociation between respiratory mechanics and the degree of hypoxemia. With this potential for a new presentation, the foundation treatments of ARDS came into question. While this initial observation was heard around the world and speculation that this could be a new ARDS rose, larger studies began to come out comparing the new ARDS to the classical ARDS. Looking at various databases, such as the LungData database, conducting observational studies and prospective matching studies, large studies were looking to see whether or not similarities existed between the two ARDS modalities. Haudenburg et al. did prospective cohort matching group and found no difference in plateau pressures, driving pressures, respiratory system compliance, and respiratory system resistance. What they did find was something I will talk about later, is a higher recruitability to inflation ratio in COVID-19 ARDS. One of the largest studies done by Ferrando et al., 742 patients from 36 hospitals in Spain were enrolled. They also had similar lung compliance, driving pressures, plateau pressures, and mortality outcomes to previously studied COVID-19 studies. Zara et al. then took a look at 66 patients with COVID-19 ARDS. The respiratory mechanics were once again similar to other large published studies in ARDS, and importantly enough, the response to proning was also similar with a significant improvement in the PA to FiO2 ratio. It is important to note that in all these studies, the foundations of ARDS treatment, like low tidal volume ventilation, higher PEEP, proning, and muscle paralysis were employed. What these studies are finding is that it's something that we have known for decades and that ARDS is a syndrome of heterogeneity, and within that syndrome of heterogeneity, there is no come one, come all treatment, and we must tailor our therapies to these patients and their specific needs. We will now discuss some of the aspects of mechanical ventilation that could be used in COVID-19. Once the patient is on mechanical ventilation, we have not solved anything. This is our time to achieve the goals of allowing the body to heal, initiating disease-specific treatments, and preventing further damage by the ventilator. The concept that is being proven again is that the ventilator management, regardless of inciting factor, must be tailored to patient-specific variables if we are to see an improvement in mortality. One recurring trait is that we see COVID and COVID ARDS is the elevated respiratory drive and patient effort being produced. While spontaneous respiratory effort can be beneficial for a number of reasons in ARDS with improved oxygenation, but when respiratory drive is elevated and ventilatory asynchronies occur, patient self-inflicted lung injury begins through excessive lung stress, increased blood flow resulting in pulmonary edema, and worsening lung function. Airway occlusion pressure, or P01, is an easy bedside test that estimates respiratory drive during mechanical ventilation. This value measures the first 100 milliseconds of inspiration independent of muscle weakness and respiratory compliance. What it does not provide is information about lung stress during the respiratory cycle. This information is provided by an end-expiratory airway occlusion pressure, or POC, which after a conversion factor can be used to measure respiratory muscle pressure. Together, these two tests can identify patients who are making excessive respiratory efforts that could lead to patient-induced lung injury. Dr. Snell, et al., looked at these two variables in 28 mechanically ventilated, spontaneously breathing COVID ARDS patients as they were trying to wean from the ventilator. He was hoping to try to predict if patients were using these two values, if patients would have a worsening, predicted a worsening respiratory relapse if these values were elevated. The prediction became very reliable with the P01 greater than 4. POC also performed well at predicting failure, but not as well as the occluding airway pressure. While his paper was looking at ways to wean the ventilator, it has been suggested that these measurements be made serially throughout the course of mechanical ventilation to help guide therapy. For example, if there is an increasing respiratory drive and effort being shown, this may trigger a need for increasing sedation or initiating paralysis in the hopes of preventing further lung damage. Since ARDSNet trial, PEEP has been the cornerstone of treating ARDS. While high PEEP can be beneficial in some patients, in patients who have low recruitability, increasing PEEP can lead to lung overdistension, cardiac dysfunction, and decreasing oxygen delivery to the tissues. In looking at those studies assessing the clinical characteristics of COVID-19 ARDS, there continued to be a wide range of PEEP values, but in general, moderate levels of 10 to 14 were found to be used in clinical practice, much within the normal range of previous ARDS studies. As we try to tailor this parameter of PEEP, one can do that by using the recruitment to inflation ratio to determine whether or not the patient is a recruiter or non-recruiter. The recruitment to inflation ratio is a way to practically assess whether patients have recruitable lung or if the addition of PEEP will have negative side effects. There are labels as recruiters or non-recruiters that the R to I ratio is greater than 0.5 or less than 0.5, respectively. In practice, the application of this equation is that the R to I ratio is representing the proportional volume that is distributed to the lung recruited versus the baby lung when the PEEP is adjusted. Therefore, the lower the ratio, the more volume will be distributed into the already baby lung and risk hyperinflation and further lung damage. This patient would be labeled a non-recruiter. In contrast, a high R to I means that the increased PEEP has mostly gone to new recruited lung segments, meaning a potential for increased gas exchange. The study by Chen shows many concepts which in practice appear to be true and accurate as seen in the figures below. As the R to I ratio decreases, so does the PAA to FiO2 ratio at low PEEPs, and when the R to I ratio increases, the normalized change in dead space will decrease, which makes sense for improved gas oxygenation. This concept was also used in Haudeburg's study when he was comparing respiratory mechanics and found that the patients with COVID ARDS had a higher R to I ratio than patients that were in regular ARDS. Other methods of recruitability do exist, including esophageal pressure monitoring as well as electrical impedance tomography, but both are rather cumbersome and their utility in clinical use is limited. This makes the R to I ratio appealing method performed easily at the bedside to determine optimum PEEP levels in patients and reduce the risk of atelectasis and overdistension. One of the foundations of ARDS management that arose from the publication of the PROCEVA trial and several other meta-analyses was proning in moderate to severe ARDS. While having positive reports in the literature, the rate of proning patients, even at large academic medical centers in the United States, is still relatively low, stemming from discomfort with the actual positioning, not enough equipment or staff, or fear that the patient would not tolerate it from a hemodynamic standpoint. In this global pandemic, we have shifted the paradigm to now we are proning patients who are not even intubated and do not qualify as moderate to severe ARDS in hopes of staving off intubation. Whether this strategy is truly beneficial is still up for debate due to the lack of randomized trials, but several observational trials have shown improvements in PA to FiO2 ratios in awake patients. Because the optimal timing of intubation in ARDS is not known, close monitoring must be done on these patients to make sure their respiratory rate and their tidal volumes do not increase too much to induce lung injury. The only absolute contraindication to prone positioning is an unstable spinal fracture, while there are a number of relative contraindications that must be taken into consideration, such as increased intracranial pressure and open abdominal wounds, this must be taken in terms of the patient. A positive takeaway from COVID is that the comfort level and experiences of patients and providers and proning patients has improved dramatically, and this will be helpful in treatment of ARDS for years to come. Given this information, it is recommended that patients with moderate to severe ARDS be prone, barring any contraindications. There are several mechanisms why proning works. The quantity of tissue made available for gastric change in the dorsal area is much greater than the area of derecruitment that is seen in the ventral area when flipped over. This improves gastric change and ventilation mismatch. The lung tissue on the posterior side is also more homogenous, and this creates also an improvement in the ventilation perfusion mismatches as well. With the compliance of the anterior chest wall going down in the prone position, more blood flow and more oxygenation is now vented towards the posterior portions. Since the ACURASYS trial, neuromuscular blockade has been used widely in the treatment of ARDS. The recent ROSE study called these findings into question, though. While there are significant differences between the two trials, there are plausible physiologic benefits to muscle paralysis. It can inhibit large changes in transpulmonary pressures related to strong inspiratory efforts as well as active expiration, leading to alveolar collapse and worsening ventilator-induced lung injury. It can also decrease oxygen consumption by reducing the work of breathing and allow for redistribution of blood flow to other areas of the body. Some negative aspects include the need for deeper sedation and the ICU-required weakness. While paralysis has been used in cases of severe COVID ARDS throughout the literature and with the proper ventilator management should be continued. The potential rescue therapy that was being used at NYU was the introduction of inhaled nitric oxide in patients with refractory hypoxia. INL acts as a smooth muscle vasodilator, and once it crosses the alveolar membrane, it increases blood flow to the oxygenated areas, improving ventilation matching. In previous studies of ARDS and INL use, transient improvements in PA to FiO2 ratio were seen, but not associated with any clinical significant outcomes. The surviving sepsis campaign guidelines for COVID-19 recommends against the routine use of INL, but does recommend it in cases of refractory hypoxia. Previous studies looking specifically at the use of INL therapy in COVID ARDS has shown minimal improvements in PA to FiO2 ratios, although none of these were randomized trials. One trial did show an improvement, but that was mostly driven by an improvement of RV dysfunction in a small subgroup of patients. At this time, we cannot recommend that INL use be used routinely in patients with COVID ARDS. COVID-19 has presented us with an opportunity to prevent and improve our ventilator strategy and knowledge for all patients with ARDS. We should strive to tailor our treatment efforts to specific patient characteristics if we hope to see improvement in outcomes. In treatment of COVID-19 ARDS, we should maintain our core fundamentals of ARDS treatment and lung protective ventilation. Thank you. Dr. Justin Towle is a cardiac anesthesiologist and intensivist at the Medical College of Wisconsin in Milwaukee. He's an assistant professor and also the assistant program director of the Critical Care Medicine Fellowship there in Milwaukee. His practice includes a good deal of critical care related to CT surgery and mechanical circulatory support. And he'll be speaking on a topic of great debate and interest throughout the year, coagulation and COVID-19. Hello. My name is Dr. Towle. I am an assistant professor of cardiac and critical care anesthesia at the Medical College of Wisconsin. And I'll be discussing hematologic testing and challenges in COVID-19. I have nothing to disclose, no conflicts of interest. Dr. Lisa Baumann and I have shared some of the most of these slide decks. So I appreciate her help on that. I just want to mention her. My primary care of these patients is in a CT ICU in an ECMO unit. We'll review COVID-19 inflammation and bleeding and clotting as a part of critical illness, as well as societal guidelines and anticoagulation options and monitoring before we review at the end. So coronavirus coagulopathy has been well demonstrated in prior infections, both with SARS and with MERS, an inflammation that causes something that looks like DIC was well reported. We see high levels of acute phase reactants, fibrinogen and factor VIII associated with these infections and what looks like a typical DIC, low fibrinogen and platelets with prolonged PT and PTT testing as the disease progresses. Even at the very beginning of COVID-19 have seen these coagulopathies in the very ill. So it's been reported that up to 50% of non-survivors end up with some kind of degree of coagulopathy and frank DIC in as many as 71% of non-survivors in early studies. We call this an atypical DIC picture because often in this particular illness, the fibrinogen levels are quite high. We also see inflammatory effects that are the cause of our coagulation problem rather than it just being exclusively a coagulation issue. So there's a positive feedback loop where inflammation causes clotting, clotting causes inflammation and that process builds on itself. So the typical findings in COVID-19 are really a small reduction in a platelet count and a mildly prolonged prothrombin time. We see very high levels of fibrinogen, especially early in the disease. And then very late in the disease, we may see low levels of fibrinogen once we've turned up all that we have. The D-dimer has been a prominent part, the elevation that's quite dramatic in the D-dimer has been something that's sort of the hallmark of this disease as it really progresses to critical illness. So I want to take a little bit of time to remind you guys of where that comes from. The D-dimer is often thought of as the breakdown product of fibrin, so it's fibrin cross-links. And then it comes back being broken down by tissue plasminogens, plasmids. That's how we typically think of it, but there's a bit more going on in this particular disease. So we know that in ARDS and specifically in COVID, we're seeing a ton of fibrin deposits along with the inflammation in the lungs. So not just the alveoli, but also in the septum between those alveoli, there's a large amount of fibrin deposit, which is then chewed up by macrophages and spit back into circulation that we see here as high levels of D-dimer that may or may not be related to intravascular coagulation. Some of the first reports we saw were in the U.S. were autopsy studies. They demonstrated high levels of pulmonary hemorrhage. You can see in this gross sample, there's some dark spots both at 9 o'clock and over around 3 o'clock that represent pulmonary hemorrhage on the gross specimens in the lungs. That's despite all the concerns that we had initially about clotting. The histological slide here is that fibrin deposition we discussed in the prior slide. The blue here is fibrin, and you can see these areas around alveoli and in the septum that are just densely packed with fibrin. The natural question early on was, is this pulmonary hemorrhage the result of prolonged EIC and exsanguination following the consumption of olfactors? It doesn't seem to be that way. There seems to be simultaneous coagulation as well as bleeding in most of these patients, both in the lungs and in the mucosa. We want to put that into some context within critical illness. When we look at studies of ICU patients and their rates of DVTs, we see that a typical ICU patient is somewhere between 6% and 16% at a rate for a DVT, which is why we prophylaxed most of these patients. However, when you really look at sepsis, we see those rates are much higher. The inflammatory process associated with sepsis in the Kaplan study showed rates as high as 37%, with 3% to 4% of patients suffering from PE. We need to remember that sepsis, separate from COVID, does have these strong inflammatory processes that encourage patients to have clot. When we look specifically at COVID-19, we see a huge range of venous thromboembolic disease, anywhere from as low as 7% all the way up to 47%. Suffering rates are not well described empirically in this way in prospective studies, but they're somewhere between 2% and 10%. Seems to be much lower than the risk of clot. Why is there such a large discrepancy? Well, we can think about this in a couple of different angles. Some of it's the patients. Some patients in these studies, especially towards the bottom, are sicker. They have more intense ICU criteria like ventilation requirements or pressor needs, frank sepsis. Others, at the top of the screen, they weren't looking very hard. So when we look at Dr. Klock's study, 37% of what he found, he was really looking very hard. They were using ultrasound on patients that were asymptomatic and really discovering a lot of disease that we wouldn't normally find unless we're really looking. So the harder we look with clots, the more we find. That's a little bit different from bleeding, which is usually a bit more obvious, drops in hemoglobin, blood pressure. We see frank bleeding. But with clots, the harder you look, the more you find. And so it shouldn't be surprising that when we really surveil prospectively, we're gonna pick up a lot more than we would normally see. But if you put that in the perspective of what we saw in the Kaplan study, it's right about at the same level. So it's important to recognize that we do see this typically with all critical illness. And while it may be more on the tip of our noses, it may not actually be all that much more prevalent than what we see in severe sepsis. So early on, society recommendations were lacking. Nobody wanted to speak on COVID-19 specifically, but over the last year, we've seen almost every hematologic society has come out with recommendations for prophylactic dosing rather than anything more aggressive like this intermediate dosing. We talk about intermediate dosing of Lovenox. We're talking about that half per kilo twice a day as opposed to the once a day dosing. We see some places are making recommendations for testing. Others are making recommendations for this intermediate dosing when patients are quite sick. So if you are really critically ill, or if you're showing evidence of clot despite prophylactic heparin, those patients may be considered for this intermediate dosing heparin strategy. Otherwise, prophylactic normal dosing daily of Lovenox or low molecular weight heparin would be appropriate. There are additional recommendations including SCCM itself that come out with a preference for low molecular weight heparin, as well as chest and the OB societies as well. OB are a little different. They don't wanna recommend Lovenox given the anorexial techniques and the bleeding associated with delivery. So there is one interesting randomized prospective trial that looked at really assigning patients to either a prophylactic heparin strategy or a therapeutic heparin strategy when they're an inpatient with COVID. That was halted because of harm to the therapeutic group. So we do wanna avoid therapeutic anti-coagulation unless patients have a specific indication like a PE or clot. There are sort of three main medications that we use for this. The first being unfractionated heparin, which we all know very well. It's a dirty drug. It acts at a lot of places, requires antithrombin three, but it can be completely reversed with protamine. It is renal cleared and temperature dependent for its offloading. When we look at low molecular weight heparin, there are comparable rates of bleeding. It seems to protect patients a little better because of the rapid onset and the longer duration, but it is only partially reversed. So if you do have a risk of bleeding, that is something you need to be careful with. I wanna remind everybody to dose adjust their low molecular weight heparin for both decreased GFR or low renal function or obesity greater than BMI greater than 40. The direct oral anti-coagulants did receive a fair amount of early notice, especially as hospitals were becoming overwhelmed. The lack of subcutaneous dosing, as well as the lack of monitoring need were seen as benefits early on. However, there's growing data to suggest that some of these anti-retrovirals were really pushing up levels. And so the CYP inhibition was really causing dramatic elevations in your anti-TENA levels and people were having problems. So these are generally not recommended for the critically ill at current. They are great for outpatients or people getting ready for discharge. So we'll talk a little bit about testing. The ACT is the most commonly used test for heparin effect that we know well in the operating room as a point of care test. It's not very good outside of that environment. There's a bunch of different machines and they all give you different numbers. And so there's really a lot of variance between these devices. And so we also see that the confounders are quite dramatically deranged in the setting of COVID. So low fibrinogen platelets and high elements of D-dimer can really make your ACT even worse. I wanna draw your attention to this sentence. The ACT is poorly associated with heparin dosing, bleeding, and clotting. Well, what do we care about? We care about bleeding, clotting, and we like to think that our ACT is dose dependent, but actually most of the studies show that it is inversely associated with heparin dosing. So both PTT and anti-TENA testing are preferred to the ACT. When we move on to the PTT and the anti-TENA, this is a bit of a messy slide, but it's important that you see these things next to each other. I've got the short and dirty up top. The non-heparin PTT confounds are strong and they're variable. We see PTT shifting shorter or longer depending on the state of the patient rather than the state of the heparin. Anti-TENA levels are pure heparin test. So the PTT is the most widely associated So the PTT is the most widely available and used test for heparin monitoring. It looks at all the elements of the intrinsic pathway. It has really no standard lab techniques. Your lab and my lab, completely different in the way that they put additives in and whatever normal they believe it is. So they really don't translate from lab to lab. It also is weakly correlated with heparin dosing. The graphs that you see in the top right corner the left graph is the PTT versus the heparin dosing. The heparin dosing is seen at the bottom. The PTT is along the Y axis. We see a weak correlation here. That's different from what we see in the ACT. The ACT, we see a strong correlation around 0.4. So what we really see is our heparin dosing increases, our anti-TENA levels increase. Whereas with the PTT, increases in your heparin level are not strongly associated with your PTT. The reason for that is that there's all these confounds. When we look at the levels of factor VIII and fibrinogen which are quite elevated in COVID-19, we see that shortens your PTT. Whereas other things like the D-dimer, the CRP, the platelets and the other factor levels may actually depress the PTT or prolong the PTT. So what we see a lot with the PTT is what's going on with our patient. We're seeing patient factors and different levels of coagulation factors playing a key role here rather than necessarily what your heparin is doing. On the other hand, the anti-TENA levels are really a pure test of heparin. They are the gold standard. It is a bit more expensive of a test and it's not widely available everywhere. It is a standardized test and it is an optical test. So the confounds for this test are really things that make color changes or cloudy plasma. So triglyceride poisoning or hemolysis can cause the color of your plasma to either be cloudy or turn red. And that really obstructs the light going through it which may cause the level to be reported lower than it really is. These situations would cause you to overdose your patient. In addition, the direct oral anticoagulants cause very high levels of anti-TENA. And so as we transition from an oral agent to a subcutaneous or IV agent, you need to be very careful and avoid anti-TENA monitoring for at least three to four days. Having said that, the platelets and the fibrinogen are almost irrelevant to the anti-TENA levels. So if your patient has very low levels of fibrinogen or platelets and you're dosing them very high on the anti-TENA scale, you may be setting yourself up for a bleed. So it's important to look at these two numbers in the context of the patient. Either one of them tells the complete story in the setting of COVID. So our take-home points, COVID-19 is an inflammatory disease with atypical DIC. Thrombosis is very common, around 30 to 35%. And then bleeding is about 10%. So that really only happens in the severe critical illness patients. We see survivors and non-ICU patients not have any of these problems at very high rates. Standard prophylaxis is recommended by pretty much every guideline that is out there now. And we need to remember to dose our low molecular weight for renal impairment and obesity. And if you're not documenting a patient's clots, if you're not documenting PE, if you're not seeing clots in the arms or legs, you really want to stick with a prophylactic dosing. You can consider that intermediate dosing in your sickest patients, where you're seeing hints of failure of therapy, or where they're really just very critically ill, high levels of coagulation factors in studies. And that's the time to think about it. But in general, we want to stick with that standard prophylaxis. It's really important that you understand your lab and your patient as you interpret your studies. We want to remember that the PTT and the anti-TEN-A provide complementary information, and that the PTT can be altered in either direction and may not monitor heparin very well. Whereas the anti-TEN-A is going to tell you everything you want to know about your heparin, but won't tell you much about your patient. So you want to think about both of these things together and analyze your patient for their particular content. Thank you very much. Dr. Jing Tao is also a critical care anesthesiologist, also located in New York City, at Memorial Sloan Kettering Cancer Center. She specializes in the care of critically ill cancer patients in both the MICU and SICU. And she'll be speaking on some of the most important articles of the year, and in this framework, provide a year in review. Good morning. My name is Jing Tao, and I'm an assistant professor at the Department of Anesthesiology and Critical Care at Memorial Sloan Kettering Cancer Center in New York City. Today, my talk will be on the year in review top 10 articles of 2020. This year has been very much embroiled by COVID. However, I did not want to focus my entire presentation on COVID articles. So I've divided them up into six COVID articles and four non-COVID articles. So I do not have any disclosures. The first study I wanted to mention is the dexamethasone in hospitalized patients with the COVID-19 preliminary report. The dexamethasone study was based on the idea that majority of organ failure from viral pneumonia came not from the virus itself, but from the host response. Steroids, which have been extensively debated in other inflammatory pathologies, was trialed again here to see if it can attenuate organ injury and prevent further spread. The study randomized patients to receive either usual care or usual care plus dexamethasone for about to 10 days. And their primary outcome was 28-day mortality. The study was able to recruit about 6,500 patients. The steroid group was given dexamethasone about one week after symptom onset. And as you can see from the graph here, steroids were shown to significantly reduce mortality with a rate ratio of 0.83, which is almost a 20% drop. This mortality benefit was most pronounced in mechanically ventilated patients seen in figure B on the left, with a proportional reduction of about one-third. Mortality also dropped in patients who required oxygen therapy in figure C. These included anyone from needing nasal cannula to BiPAP. However, no mortality benefit was seen in those who did not receive oxygen therapy. In the case of those who did not receive oxygen, in figure D, there was actually a trend towards an increase in mortality in the steroid group, although this did not reach statistical significance. The evidence was strong enough that in September of this year, the WHO updated their guidelines to include a recommendation for cortical steroid use in severe and COVID-19, and recommended against this use in non-severe cases. The next important new COVID study was on trauma and the rate of thrombosis in the New York City health system. The study was done due to increased incidence of thrombosis noted on autopsies of COVID-19 patients. It set out to assess the risk factors as well as the incidence of this at four large NYC hospitals. It was an observational trial with thrombosis screening that was not routine or standardized, but done by routine clinical care. And the results were quite striking. They found an overall 16% overall thrombotic rate in all hospitalized patients, a staggering 29.4% in ICU patients, and about 11.5% in non-ICU patients. When you break down thrombotic rates by demographics, higher rates were seen with age over 55, as well as male sex, and Hispanic race, and a history of MI and CAD. Thrombosis was not surprisingly associated with D-dimer levels, and the rate significantly increased with increasing D-dimer of greater than 500. Unfortunately, no standardized way to deal with these findings came out of the study. It solidified the idea that COVID patients were high risk for thrombosis and DVT prophylaxis should be given. However, there are many things that are still in debate, including if and how systemic anticoagulation will benefit, if and how anticoagulation should be diatriated based on D-dimer levels, as well as the need for anticoagulation post-hospital discharge. Because there's been no official stance, what hospitals have generally been doing is using low molecular weight heparin instead of heparin for DVT prophylaxis, and giving it once a day for non-ICU patients, and then twice a day for ICU patients. The next article is on the use of remdesivir in the treatment of COVID-19. The idea of remdesivir came from its antiviral effects against SARS and MERS, as well as some in-vitro data showing good effects against COVID. The study was a double-blind randomized placebo trial using remdesivir versus placebo in hospitalized patients with COVID. And primary outcome was time to recovery, which was defined as hospital discharge, as well as infection control. This is a bit of a busy slide, but the results found that overall remdesivir was able to improve hospital outcome or hospital discharge from about 15 days to 10 days. This was most pronounced in patients who were hospitalized and received supplemental oxygen, but was not seen in those who either did not receive oxygen or needing higher levels of oxygen, such as BiPAP, mechanical ventilation, ECMO. The study also found that clinical recovery time was better if you gave remdesivir within 10 days of symptom onset, but this loss of significance when it was greater than 10 days. However, about a month later, the WHO came out with their solidarity trial and found that out of the 2,700 patients who they gave remdesivir to, there was actually no change in mortality, ventilation, or hospitalization. It did recognize that the drug was found to improve recovery time in less sick patients in other studies. However, it didn't feel that the evidence was strong enough for remdesivir's use and issued a conditional recommendation against its use in COVID patients, regardless of disease severity. The next study is the anticipated look at convalescent plasma on recovery time in COVID-19. Convalescent plasma is not new in viral infections. It was tried in multiple other virus outbreaks, but definitive results have been unclear about whether it works in anything. The trial was done after a small case series finding success in using antibody-rich plasma against COVID patients. The study was an open-labeled, multi-centered, randomized controlled trial comparing patients getting either usual care or usual care with convalescent plasma, and the primary outcome was time to clinical improvement within 28 days. Unfortunately, results were quite disappointing as convalescent plasma was not shown to increase overall time to clinical recovery, as you can see in the first red box. The only improvement they were able to find was a faster way to negative viral PCR detected from a nasal swab, which I'm not sure how much that will help you if your patients are not getting better sooner. As a result, the NIH states that at this point, there is insufficient data for the recommendation of either against or for the use of convalescent plasma for COVID-19 patients. The next study is the use of a wake proning to prevent intubation in high-risk patients. At the beginning of the outbreak, it was recommended for patients to be intubated early to prevent patient self-inflicted lung injury. However, when the number of cases really skyrocketed, there was concern that intubating everyone could deplete resources. And this was when the idea of a wake proning began. And the idea was that if we can improve VQ mismatch in patients who are at high risk of intubation, but not yet intubated, are we able to definitively reduce this need? The study was a prospective multi-centered observational trial of 36 hospitals across Spain, and it compared the need for intubation in patients on high-flow nasal cannula versus those on high-flow nasal cannula doing a wake proning. Results, again, were quite disappointing. As you can see here, a wake proning in patients on high-flow nasal cannula did not decrease the rate of intubation or 28-day mortality. The last COVID study I want to talk about is probably the most anticipated one of the year, and that's on the safety and efficacy of the mRNA COVID-19 vaccine. The reason for this study is obviously self-explanatory. The study was a multinational placebo-controlled observer-blinded efficacy trial. Patients were randomized to receive either the trial vaccine at 30 mics IM, 21 days apart, or placebo. Only exclusion criterias were children under 16, active COVID-19 infections, and an immunosuppressive condition or therapy. Primary outcomes were twofolds. One was safety, which was either local or systemic adverse effects up to about six months after the second dose, and the second was efficacy, which was confirmed COVID-19 infections defined as symptoms plus a confirmed respiratory specimen at least seven days after the second dose. In terms of safety, results found that the drug were quite well-tolerated with only mild to moderate local and systemic effects seen in younger patients or younger participants compared to older ones. Pain in the injection site was the most common complaint, followed by fatigue, headache, and fever. Although mild, more than half of the patients did experience some symptom, mostly fatigue and headache, and all the adverse effects occurred within the first one to two days. In terms of efficacy, this was surprisingly positive. In patients who never had a positive COVID test, only about eight patients in the treatment group got COVID compared to 162 in the placebo group. In participants with and without COVID positive tests, only nine got COVID after vaccination compared to 165 in the placebo group. And this translated into a 95% efficacy rate. Although this study was not powered to look at subgroup efficacy, results it did have show that it was equally effective across age groups, sex, race, or ethnicity, as well as the country of origin. When you dive more into the results, this figure shows efficacy just after the one first dose with a zoomed in graph focusing on the number infected within the first 21 days between the first and second dose. And looking at this graph, only 39 patients in the vaccine group ended up with COVID versus 82 in the placebo groups. The others concluded that the efficacy after dose one was about 50%. We can also see from the zoomed in graph up top is that the number of infections really started diverging around day 12, with the placebo group in blue continuing to climb in number of infected and the vaccine group in red plateauing out. This suggests that there is early protection by the vaccine starting as soon as 12 days after the first dose. In terms of protection against severe COVID, the study found that vaccine had an overall 89% efficacy with only one out of the 10 patients who ended up with severe COVID in the vaccine group. And breaking it down to days after dose one and dose two, the efficacy dropped to about 75% when you go beyond seven days after the second dose. So that was it for the COVID studies. Now I want to move on to some articles not related to COVID. The first one is the AXE trial where the ascorbic acid, corticosteroids, and thymine on organ injury in septic shock. The idea for this study came from data suggesting there is synergistic effects of steroids and ascorbic acid in improving outcomes in sepsis. The study was a randomized blinded multi-center trial of about 100 patients on vasopressors. It compared those who got placebo versus triple therapy and primary outcome was a change in SOFA scores between enrollment and 72 hours after treatment. The study group had initially predicted a two point SOFA score decrease in the treatment group, but the actual study results did not find treatment to be very effective. Looking at bigger two on the left, there was no SOFA scored change at 72 hours and looking at bigger three on the right, mortality at 30 days was also similar. This next study focuses mainly on trauma patients. I thought I'd touch on a very popular topic, which is the utility of fiscal elastic acid in trauma resuscitation. As many of us already know, trauma coagulability benefits from an individualized approach. And the classic method of using conventional coag testing really suffers from long turnaround time and it really only gives you this quantitative measurement. Fiscal elastic testing like ROTEM and TEG, which have become quite popular in the last decade or so, have become so popular because results can be seen in real time and it gives you a qualitative look at things. The study was a multi-centered randomized programmatic trial of seven major trauma centers across Europe and looked to assess if empiric massive transfusion protocol was better guided by conventional test or by fiscal elastic assay. Primary outcome was proportion of patients alive and free of massive transfusion at 24 hours after injury. The study included about 400 patients, which is the largest one of its kind and have fairly high baseline injury severity scores at around 26. Unfortunately, results from the study did not find fiscal elastic assay to be superior to conventional tests when it comes to survival at 24 hours or preventing the need for massive transfusion. When you break it down by subgroups, which is here defined as PT greater than 1.2 at baseline, severe TBI and prior oral anticoagulation use, fiscal elastic assay was also no better than conventional test. Secondary outcomes such as mortality, median time to hemostasis and rate of multiple organ dysfunction were also similar. The next article is the one on endoscopy for timing of endoscopy for acute upper GI bleed. It was not specifically performed in critically ill patients, but because of how frequently we encounter GI bleeds in the ICU, I thought this article would be important to mention. The study was done due to unclear international consensus group recommendation on when high-risk GI bleeds should be scoped. There's also been large cohorts showing conflicting evidence on the timing of endoscopy and mortality. This study randomized high-risk GI bleeds, excluding those in shock, to randomly receive endoscopy either at six hours or 24 hours after GI consult. Primary outcome was mortality of 30 days. And just of note, all patients received high-dose PBI infusion before they were scoped. The authors had initially predicted that early intervention would result in better mortality benefit. However, the study did not actually find this to be true. As you can see in figure one, or in figure A, 30-day all-cause mortality was similar between the two groups. Looking at figure B on the right, the study also found that despite the six-hour group receiving more interventions like clipping and banding compared to the 24-hour group, probability of further bleeding ended up being the same. The authors attributed this to the fact that the 24-hour group was able to spend more time getting high-dose PBI than the six-hour group. And again, this study did not include those who were persistently hypotensive despite resuscitation, but it did provide evidence that most GI bleeds can weigh on intervention, especially if high-dose acid suppression is given. The last article looks at targeted temperature management in cardiac arrest with non-shockable rhythms. The need for this study came from the International Liaison Committee on Resuscitation recommending a wide range of hypothermia between 32 and 36 degrees for cardiac arrest. And it makes this recommendation without really distinguishing if that cardiac arrest is from shockable or non-shockable rhythm. And prior to the study, there's been evidence in non-shockable rhythms that are conflicting in terms of benefit of hypothermia. This study was an open-labeled, multi-centered pragmatic study done at 25 ICUs across France. Patients were randomized to receive either hypothermia at 33 degrees for 24 hours, and then re-warmed for another 24 hours versus normothermia at 37 degrees for 48 hours. Primary outcome was survival with favorable 90-day neurological outcome. And the results were quite favorable and showed that hypothermia at 33 degrees benefited both neurological outcomes and survival in non-shockable cardiac arrest. CPC, which is the Cerebral Performance Category, is a scale used to measure neurological outcome. And the lower the number, the better. And as you can see in the first red box, many more patients in the hypothermia group had lower CPC numbers than the normothermia groups. There was also a slight improvement in death in ICU, survival to ICU discharge, and survival to hospital discharge. Death in 90 days was not affected. So just to review, the top 10 articles of 2020 when it comes to COVID include the dexamethasone study on survival in COVID-19, the very disappointing rendesivir, convalescent plasma, and awake proning, and then finally, the efficacy and safety of the mRNA vaccine. And for non-COVID, important studies include ascorbic acid, corticosteroids, and thymine in septic shock, viscoelastic assay in trauma resuscitation, thymine mendosity for GI bleed, and finally, targeted temperature management for non-shockable cardiac arrest. Thank you.

anesthesiology

mechanical ventilation

COVID-19 ARDS

intubation

non-invasive positive pressure ventilation

coagulation

corticosteroids

remdesivir

mRNA COVID-19 vaccine