A really major area of focus, at least in the last few weeks, has been the cardiac injury that we see in hospitalized patients with COVID-19 [coronavirus disease 2019]. Alex, you’ve been very busy, not just taking care of patients but also collecting data on the different types of complications that patients experience when they have a COVID-19 infection. Can you tell us a little about what you know regarding cardiac injury? I can speak to that a bit myself. There’s a lot of thought about how blood clotting begins to predispose to some of these cardiac injuries. Maybe you could tell us about how COVID-19 affects clotting.

 Much like Vic, I really want to give a shout-out to my younger faculty members who’ve put in the huge amount of work on the frontlines and have done the yeoman’s work of taking care of these patients. We’ve had a huge influx in our system at Northwell Health. The numbers are what I call wartime numbers. We have seen 100,000 patients and a few thousand patients—up to 3000&mdash;at once in our health system, and about a third of those are intubated. There are tremendous numbers with this. We’re actually gaining a good sense with respect to the cardiovascular and thromboembolic complications in this COVID-19 population.

Much like you all, we’re starting a registry to better assess how and when these complications appear. The first thing that I can tell you, looking at the literature and our own population, is the risk of thrombosis is about 3- to 6-fold higher than what we would expect in the hospitalized population, specifically in the critically ill population that Vic was talking about. This appears to be a highly thrombogenic disease. What’s interesting is that we really don’t know specifics of the coagulopathy. Indeed, there are aspects of DIC [disseminated intravascular coagulation], but it’s not quite behaving as a consumptive coagulopathy. The D-dimer levels are very elevated, but the platelet counts tend to be normal or near normal. The coagulation parameters, the PT [prothrombin time], tends to be normal as well. Fibrinogen levels also tend to be normal or slightly elevated. It’s not a classic consumptive coagulopathic picture.  

However, as Vic pointed out, we’ve seen some of the early autopsy data, and patients are dying from massive thrombotic events—such as MI [myocardial infarction], saddle pulmonary emboli in the lungs, and things like stroke&mdash;but also a quite consistent pattern of microemboli, specifically in the lungs. One of the purported mechanisms may be this hyperinflammatory reaction, probably cytokine related. An immense release of these inflammatory factors creates a cytokine storm that folks have described, and patients become hypoxic, much out of proportion to what is seen in the regular exams. This hypoxia is leading to a microvascular hypoxemic process and maybe some right heart strain and other aspects.

There’s another school of thought that talks about the oxygen stress, especially at the level of the red blood cell, that may be adding to this. This could be multifactorial. Coagulopathy, yes. Fibrinogen levels, AT [antithrombin] III levels, and factor VIII levels are all affected, but there is evidence of a severe hyperinflammatory response.

In terms of the incidence of disease, you can probably address the cardiac issues better than I. We’ve seen classic acute coronary syndromes, but also things like myocarditis in these patients. On the venous thromboembolic side, there are some studies that suggest a prevalence of anywhere from 20% without screening to as high as 40% with routine lower-extremity screening, which is about 3 to 6 times higher than what we would expect. There is a very high prevalence of venous thromboembolic events in this population. What’s unusual is an apparently high prevalence of in situ pulmonary embolic events in the absence of lower-extremity DVT [deep vein thrombosis]. There are some very interesting things about this disease.

From the cardiovascular system standpoint, as you mentioned, it’s not just the patients presenting with heart attacks. There are patients typically presenting with non—ST-elevation myocardial infarction, but we are seeing some ST-elevation myocardial infarction. Myocarditis seems to be relatively common in these patients, either through a direct effect on the cardiac myocytes or because of some other factor. Heart failure has been seen. There have been reports of new atrial fibrillation in patients who otherwise haven’t had it. There are other types of arrhythmias and sudden cardiac death. All 3 of us are in the camp where we think a lot of these cases are probably massive PEs [pulmonary embolisms], but there have also been arrhythmic deaths. From the coagulation point of view, there is a lot of catheter thrombosis. I don’t know if you have been seeing that, but that’s been a big complaint at my center: clotting of dialysis catheters or central lines.

 Absolutely. At Northwell Health, I’m getting a lot of feedback from my renal disease colleagues. Renal replacement devices, dialysis catheter-based therapies, ECMO [extracorporeal membrane oxygenation] devices—these are all clotting through standard anticoagulant regimens. In our particular institution, we’re advocating for the use of alternative parenteral anticoagulant modalities, such as intravenous argatroban, for example. We’re using that more and more. We’re potentially moving away from the PTT [partial thromboplastin time] method to an anti—Xa method of monitoring on fractionated heparin.

 We’re seeing the same thing, Greg. We’re seeing more venous thromboembolism than arterial embolism. We’ve seen arterial embolism. Arterial clots have been line-related. We’ve had 1 leg amputation in a patient who received ECMO. We don’t see them very often. This probably wouldn’t happen with only a sheath or a catheter in place, but with this hypercoagulable scenario, we’re seeing more problems related to CRRT [continuous renal replacement therapy], central lines, etc.

Cardiac Injury and Mortality

® Peers & Perspectives® presentation titled, “Antithrombosis in Patients With COVID-19.” I’m Dr Gregory Piazza, from Brigham and Women’s Hospital in Boston, Massachusetts. I’m joined by some close friends and excellent colleagues, Dr Victor Tapson from Cedars-Sinai medical center in Los Angeles, California, and Dr Alex Spyropoulos from Northwell Health in New York, New York. We’re going to discuss the prophylaxis and management of thrombosis in patients dealing with a COVID-19 [coronavirus disease 2019] infection. Welcome, and let’s get at it.

Why don’t we dive right in? We have a number of fantastic discussion points and questions, and I know my colleagues are itching to get at this. Here’s the first question. I think it’s great for Dr Spyropoulos. Alex, what risk factors make a person more vulnerable to complications if they become infected with coronavirus?

 Greg, that’s the million-dollar question, isn’t it? That’s been asked during the few months of the pandemic. The first thing to keep in mind, which is very relevant for our discussions, is that patients with underlying cardiovascular comorbidities are at increased risk of experiencing bad complications from COVID-19, especially when they’re hospitalized. Patients with underlying cardiac disease or patients who’ve had a history of stroke, diabetes, hypertension, or thrombosis are patients who tend to get into trouble, especially when they’re hospitalized.

The other thing we’re realizing more and more is that advanced age seems to be a fairly accurate predictor of a bad outcome. Our institution, Northwell Health, just published a very large database in 

 suggesting that—also seen in China and Italy and other places&mdash;patients with advanced age are also at great risk of getting complications from COVID-19.

There are other clinical markers, and Vic and the critical-care community probably know this better than anyone. Things like DIC [disseminated intravascular coagulation] scores, SOFA [Sequential Organ Failure Assessment] scores, and SIC [sepsis-induced coagulopathy] scores, especially a score of 4 or more, also seem to portend a very poor prognosis for this hospitalized COVID-19 population. Lastly, there are key biomarkers that consistently, time and again, show that patients with these elevated biomarkers are going to run into complications. One of the most consistent biomarkers is an elevated D-dimer test score, especially when you’re using cutoffs of 4 or 6 times the upper limit of normal.

These patients tend to have poor outcomes, whether these outcomes are due to thrombotic events or events that include a coagulopathic state and highly inflammatory state. The cytokine storm that’s been described in COVID-19 is not quite known at this point. Patients with very elevated D-dimer are at risk, maybe less so with things like CRP [C-reactive protein] or troponins: these are also biomarkers that suggest poor outcomes.

Exactly. Vic, tell us a little more about some of the scores you use on the frontline in the ICU [intensive care unit] to determine patients who are more vulnerable, such as a SOFA score. What goes into that?

 We use scoring to some degree. We use the parameters that Alex talked about. Sequential Organ Failure Assessment is what SOFA stands for. It’s basically a scoring system that includes a number of physiological parameters and key parameters—renal function, platelet count, things like this&mdash;that we use to score. We use it for all kinds of patients in the ICU, Greg.

When patients get to the ICU, it depends on how you admit patients at your center. At some centers, you’re overloaded with patients, and some of the sicker people can’t even get in the ICU right away. Some people come in on 6 L of oxygen, some people come in on high-flow oxygen, and some people come in and are intubated on the floor. But once they get to the ICU, there’s generally some sort of significant respiratory stress involved. Almost everyone has bilateral pulmonary infiltrates. Some of them are minimal. Some of them are more significant. It’s been striking to me. Currently, I’m the site PI [principal investigator] for the NIH [National Institutes of Health] remdesivir trial at Cedars-Sinai, so I see a lot of these folks come into the ICU who have pulmonary infiltrates.

One of the criteria to get into the study is hypoxemia and pulmonary infiltrates. It’s interesting to me how often patients are very hypoxemic with pulmonary infiltrates. Why are they so sick? Why are they so hypoxemic when their x-ray doesn’t look that bad? On the other hand, we’ve seen some really severe ARDS [acute respiratory distress syndrome] pictures, as well. As Alex pointed out, everybody’s ferritin seems to be in the 1000 μg/L range when they’re sick. CRP levels are very, very high. The IL-6 level is very, very high. The D-dimer levels come in higher than we often see in patients with acute pulmonary embolism. We see these inflammatory markers and D-dimer levels together. Then you see hypoxemia and sometimes shock.

Many of us are thinking, Greg, that this is a DIC microvascular thrombosis picture. We need more data, I think, to support exactly what’s going on. But patients come in sick, hypoxemic with pulmonary infiltrates—sometimes out of proportion to the hypoxemia&mdash;and sometimes hemodynamic problems requiring pressures.

Alex did a really nice job of talking about the comorbidities and factors that go into managing these complicated patients. If you were to describe the type of patient you see coming into your ICU, what does that person look like? Is it predominantly an older patient with lots of comorbidities? Give me the spectrum.

 The spectrum is pretty wide. I’ve been on the PERT [pulmonary embolism response team] for the last month evaluating patients with and without COVID-19, so I’m not on the frontline in the ICU right now. I really respect my younger colleagues and the nurses and the respiratory therapists in there treating this. Greg, regarding what patients look like when they come in, Alex is absolutely right. Age is clearly a risk factor. We have 80- and 90-year-old patients in the ICU, and we’ve gotten some of them out and off the ventilators. But that’s optimistic. It’s a bad disease.

We have a 20-year-old in the ICU right now and a 40- and 41-year-old, 3 men. Men seem to be getting hit with this a bit more. Young people can get this. I have not seen a teenager with it yet. I’ve not seen anyone under age 10. Age really does seem to be protective to some degree, but it is not universal. I don’t know if it’s because of the viral load, but we’ve seen some very sick younger patients. 

Yes, certainly we’re learning that none of us should feel too comfortable about our odds if we were to get COVID-19. There is no immortality bias here.

Complication Risk in Patients With COVID-19 & Thrombosis 

There’s been a lot of buzz in the media, at least early on, about nonsteroidal anti-inflammatory drugs. Then they started to talk about anticoagulants and direct oral anticoagulants [DOACs]. There was a concern that being on a direct oral anticoagulant could increase the risk of having a complicated disease course of COVID-19 [coronavirus disease 2019] and that being on an anticoagulant alone made you a high-risk patient. What do you think about that, Vic?

 That’s a great point, Greg. I think this is 1 area where we need more data. There are a lot of areas where we’re starting to speculate and make decisions, and we just don’t have the data. We’ve learned there’s 1 area with DOAC where we need to be careful, and it’s not cancer patients. We have great data in patients with cancer in the SELECT-D study. We have the ADAM VTE study. We have the CARAVAGGIO study. For patients with antiphospholipid antibody syndrome, we need to be careful. When a patient is triple positive, as the TRAPS study and others have shown, we probably need to use warfarin. In every other situation, DOACs seem to be as good or better. I think we need more data. We need to be studying more aggressive anticoagulation.

Some are doing that already. Saskia Middeldorp tweeted that they’re performing ultrasounds on people every 4 or 5 days in the ICU. What’s the dilemma? You expose people more but get the data. We’re getting better at doing tests at Cedars-Sinai, exposing people minimally and getting appropriate tests done, but it’s still hard.

The bottom line to me, Greg—back to your question&mdash;is that we don’t have enough data. To me, we should get people on anticoagulation and learn, through some randomized trials that are done quickly, how aggressive we need to be with anticoagulation. Then we can decide&mdash;and Alex may want to talk about this&mdash;do we send these folks home on a DOAC? I’m comfortable thinking along those lines right now.

I field this question probably 10 times a day from my patients who either have my email address or find me on the patient gateway. Am I at higher risk because I’m on a blood thinner? One of the issues I talk about with my patients is the fact that it really depends on the reason they’re on anticoagulation. If, for example—and this is a patient population you see a lot of, Vic&mdash;they have chronic thromboembolic pulmonary hypertension and they’re on anticoagulation, they’re at very high risk if they get COVID-19. We don’t want them to get that. But someone who’s had a calf DVT [deep vein thrombosis] may not be in that high-risk category. As you said, we need more data. I’m hearing things from both sides, actually. DOACs could be protective. DOACs could be harmful. As you said, we need data on that. Alex, what do you think? If you were to venture a guess, are DOACs helpful or harmful in this situation?

 My personal bias is that we’re mixing up cause and effect in some ways. As Vic suggested, the reason a patient is on an anticoagulant, especially chronically, is because they have underlying cardiovascular and thromboembolic disease. That in and of itself puts you at higher risk for complications, especially if you’re hospitalized from COVID-19—related processes. I think we’re mixing that up.

One thing that I wanted to at least mention to our audience is that if your patient is on antithrombotic therapy, they should remain on antithrombotic therapy. That is a very key message from a public health perspective, because very early on there was some talk about whether we should remove these patients from their underlying anticoagulant regimens. I agree with both you. We just need more data. My bias is that potentially, patients who are on treatment doses of anticoagulants may be protected from underlying hospital-acquired thromboembolic complications, but we just don’t know. We need those data.

 I would say that if I had a patient with a DVT or PE [pulmonary embolism], perhaps even provoked—and we know patients who have provoked DVT are more likely to have a recurrence when their anticoagulation is stopped than someone who never had a clot&mdash;or at low risk, I might be inclined to consider extending their anticoagulation even longer in the COVID-19 setting. If I got COVID-19 now, even though I’ve never had a thromboembolic problem, I might wish I was on a DOAC during the phases I was getting worse. Maybe I wouldn’t get the microvascular thrombosis. Of course, I’m speculating. But as Alex said, we need data. Don’t stop your blood thinner.

Exactly. There was the same concern about statin medications for lipids. I’ve been getting those questions, though not as frequently as those regarding DOACs. “I read something that said statin medications could be bad or good for COVID-19. What should I do?” I think that the messaging from all the societies right now is to stay on what you’re supposed to be on, and stay the course. The last thing you want to do is stop something that’s preventing you from having a massive PE because you’re worried what it might do regarding COVID-19.

Anticoagulants and COVID-19 Complications 

What are some of the strategies that inpatient anticoagulation services should consider to manage these patients with COVID-19 [coronavirus disease 2019] more carefully and optimize their outcomes? You mentioned using anti-Xa levels instead of PTT [partial thromboplastin time] and potentially using direct thrombin inhibitors instead of some of our heparin methods. Talk to us a little about that. 

 The way I would conceptualize this, Greg, is thinking about 2 buckets: the patients who are hospitalized in the medical ward and the patients who are critically ill in the intensive care units and coronary care units. These represent 2 different patient groups. I think we have better data in the critically ill population, both retrospective and prospective data. There are fewer robust data in the hospitalized population, but in general, it does appear that the overall hospitalized COVID-19 population is at greater risk for VT [venous thrombosis] events than we’re used to. That’s the first message.

The second message is—similar to all your hospitalized, medically ill patients&mdash;your COVID-19 population should have a coinciding formal VT risk-stratification strategy. Whether you use risk-assessment models, such as the Padua Prediction Score or IMPROVE bleeding risk score, or you use gestalt methods, such as individual risk factors, both models fine. However, in the COVID-19 era, if a patient does not have any absolute contraindications to prophylactic anticoagulation, they should definitely be on chemoprophylaxis. That’s the first lesson to be learned.

The second approach that we’ve used—at Northwell Health, and a lot of our other colleagues across the country&mdash;is thinking about more aggressive, even primary, prophylaxis with intermediate doses of heparin that we’re not otherwise used to. There are some protocols out there advocating for intermediate doses of low-molecular-weight heparin, as an example. We also advocate for that, especially in patients whose BMI [body mass index] is over 30. We go to a more aggressive regimen, such as enoxaparin 40 mg twice a day, as 1 example.

The other aspect, of course, is thinking about multimodal strategies, such as adding mechanical prophylaxis and chemoprophylaxis in certain high-risk profiles, even though we all know that the data on mechanical prophylaxis in the medically ill population is quite poor. But we’re in a new era here regarding COVID-19. I think those are the most important points.

The use of therapeutic anticoagulation may be a step too far at this point. We’re initiating a randomized trial, as we speak, at Northwell Health. I know other institutions are also initiating a very similar trial design. We’re assessing whether therapeutic doses of anticoagulants for VTE [venous thromboembolism] prevention offer advantages in terms of efficacy over the standard prophylactic-dose anticoagulants that we’re used to and are mandated by the guidelines. That’s probably 1 of the most important questions in the hospitalized COVID-19 population. Would they benefit from a much more aggressive VT prevention strategy using treatment-dose anticoagulants?

With respect to the use of parenteral anticoagulants, as a system we’re moving toward direct thrombin inhibitors, such as argatroban and bivalirudin in certain situations, especially for device-associated antithrombotic anticoagulant issues. I tend to think that the data behind using D-dimer levels to manage a prophylactic strategy are quite weak, so I wouldn’t advocate for that right now. I would advocate for a strong institution-wide guideline that advocates for, at the very least, routine chemoprophylaxis for all these patients hospitalized with COVID-19, unless they have absolute contraindications to anticoagulation.

Inpatient Services: COVID-19 and DOACs Management   

Of these patients who are going into the ICU [intensive care unit] with COVID-19 [coronavirus disease 2019], a lot of them have underlying respiratory comorbidities, such as COPD [chronic obstructive pulmonary disease] or asthma. How much of that is a risk for having a complicated disease course with COVID-19? How do you manage that from the prevention standpoint up to when you have to ventilate these patients?

One aside I’m going to mention is that I’ve talked to a couple of colleagues now, and none of us with busy pulmonary hypertension centers have seen a case of COVID-19 in a patient with pulmonary hypertension yet. I don’t know why that is. It might just be coincidence, and we’ll see. But I wonder. There’s maybe some genetics influence here.

Could it be that they’re on anticoagulation, Vic?

It could be. We haven’t even seen the COVID-19 infection yet. I’m even wondering if somehow the 

 gene mutation plays a role here. It’s an interesting point, Greg. I’m certainly not going to stick my neck out too far, but I would love to look at that a little further. To get at your point, if someone comes in with ILD [interstitial lung disease], COPD, or emphysema, I think they’re more susceptible. This disproportionate hypoxemia that Alex mentioned is a big point. It’s similar to a big PE [pulmonary embolism]. It’s dead space ventilation. You’re seeing infiltrates of your hypoxemia, and you’ve got underlying emphysema or ILD. You’re already on 6 liters of oxygen at home. I think your risk is higher.

Do…patients benefit by using prone ventilation in those situations or not? Is it better to delay intubation or not to delay intubation? 15 years ago, we didn’t use high-flow oxygen. Now we use high-flow oxygen. People are on 40 liters per minute of oxygen, and that’s a ton of oxygen. For many people, you can stave off an intubation longer if you’re careful and monitor oxygenation carefully. But you don’t want to wait too long and have a crash intubation. We’re lucky. I’m not in the room intubating patients now. We have very good anesthesiologists who do that and do it very quickly.

Yes, I think these patients with underlying lung disease clearly have a higher risk of doing poorly. I want to keep sticking an optimistic point in there. I’ll tell you, at some of these ICUs, as we get more experience, we have done a really great job. We’ve gotten elderly patients with bad lung disease off ventilators, but not all of them. While it’s a problem and, clearly, does increase the complication rate, we’re getting better at dealing with it.

 You’re a penultimate optimist, but I’ll say something pessimistic. What’s interesting to me, Vic, is that when you look at the literature—albeit very low-quality data, but also our own experience at Northwell Health&mdash;these patients, whether they’re in the ICU or not, seem to have breakthrough clots, despite being given what were previously adequate anticoagulant regimens. Despite being on chemoprophylaxis, we’re seeing a lot of patients experience breakthrough clots and be “resistant” to these types of regimens. We have one nonpublished experience at Northwell Health and have found a prevalence, screening with lower-extremity 2-point ultrasounds in the ICU of DVT [deep vein thrombosis], as high as 40%. Indeed, some of the most recent data out of Europe are suggesting a prevalence that high, 30% to 40%, using screening methods. To me, it’s fascinating. There’s something about this disease that’s very thrombogenic and very resistant to the usual modalities that we’ve used for the last two decades or so. 

I think it’s a real dilemma, Alex, because I still have equipoise randomizing patients who are on enoxaparin 40 mg once a day versus a more aggressive regimen who come in with COVID-19 and no proven DVT yet, maybe with criteria including high-inflammatory markers, D-dimer, etc. But it worries me more and more, as the week goes on. I think we’re going to get data here. Maybe it’ll be your study, Alex. We’ll get some data here pretty soon, and I hope we can make better decisions, because I’m starting to worry more and more that prophylaxis alone may not be enough.

Underlying Respiratory Comorbidities and COVID-19

Alex, this is a good question for you. How do we define the high-risk patients, especially in this COVID-19 [coronavirus disease 2019] setting? Are they all high-risk, and would they all fall into that medically ill, high-risk category? How much do their underlying comorbidities come into it? Or, is it really simple? Is it just that you have COVID-19, you are in the hospital, and you are high risk?

 I think, Greg, you hit the nail on the head. From what I’ve seen at this point, if you’re hospitalized in a US hospital with COVID-19, you’re at high VTE [venous thromboembolism] risk. You probably have different shades of high risk: high to very high risk, especially if you’re critically ill. But in my view, I don’t think we need to slice and dice. I think what we should assume is that these patients have a high risk of VTE, period. In my view, they should all get chemoprophylaxis, unless they have absolute contraindications.

It makes sense to me. As you mentioned, the burden of VTE in these patients is high: higher if you’re in the ICU, but still high if you’re hospitalized. It’s happening, even in patients who are receiving our usual best effort, which is prophylactic anticoagulation. Do we have any sense of why patients may be resistant to our usual anticoagulant interventions?

I’ll take a stab at it, but I would love to hear what Alex has to say. I would just say that we’ve seen other viruses, such as H1N1 [

]. Other viruses seem to be associated with an increased risk of VTE or hypercoagulable state. This seems to be unprecedented to me. We’re all noticing this. When someone comes into the ICU, we don’t routinely see elevated ferritin in everybody. Now we’re seeing ferritin levels of 4000 ng/mL, CRP [C-reactive protein] levels of 200 mg/L, and D-dimer levels that are 1000-fold elevated.

We all think of VTE as an inflammatory state, but not when you have a set rate of 150. We always say this. It’s an inflammatory state, but it’s not a grossly inflammatory state with fevers and high set rates. There is inflammation. But this case seems to be different. There are fevers in these patients, and we see these dramatic inflammatory markers. My thought is that there’s a lot of inflammation that ties in with thrombosis, and we’re seeing what I believe is a microvascular thrombosis picture in these patients. That’s why we’re seeing dead space ventilation, high oxygen requirements, and sometimes, not so much on the x-ray.

There are other issues, too, and I don’t mean to digress. Some of the patients we’ve had have had tons of secretions, and they even need bronchoscopic suction. I’m glad I’m not the one in there performing that, either. I think that they’re sick people with inflammation, and there may be something even more peculiar about the coagulation pathway and the DIC [disseminated intravascular coagulation] picture that Alex described. It’s not a classic DIC picture.

Stratifying Patients as High-Risk With COVID-19

Now that we’ve had some experience under our belts, what are the current recommendations or guidelines for prevention of DVT [deep vein thrombosis] and PE [pulmonary embolism] in patients with COVID-19 [coronavirus disease 2019]? Different societies have issued guidance on their websites. Obviously, there’s no way you can put together a big guideline document this quickly, but what do you use? What do you recommend those joining us for this conference use as a resource?

 Greg, first of all, I have to take my hat off to the various cardiology and thrombosis societies that have gone through herculean efforts to produce in 2 weeks what usually would take 6 or 8 months—guidance, or even a guideline document. You and I both worked on a very recent guidance document that was multi-institutional, international, and multidisciplinary. This represents a state-of-the-art summary of what we know at this point, with respect to thrombotic disease management.

I’m actually leading an effort by ISTH [International Society on Thrombosis and Haemostasis] as part of the Scientific Standardization Committee in creating a VTE [venous thromboembolism] document, both for prevention and management. What we see, as Vic was discussing earlier, is a unification and harmonization of all the guidance guideline bodies that, at this point, using the best available evidence, suggest we should recommend routine chemoprophylaxis at the standard doses. Right now, the 1 thing we should do is provide standard-of-care chemoprophylaxis using the standard dosing regimens of either unfractionated heparin—especially if you’ve had elements of renal insufficiency&mdash;or low molecular weight heparin, which seems to be the preferred agent. Of course, we should have a low threshold for increasing the dose, especially in obese patients. They seem to get into more trouble in terms of outcomes. I think all the guidance guideline documents advocate for the need for high-quality trials, which is what Vic was talking about.

At the very least, have a method of identifying these patients and prophylaxing them. All the listeners out there, get your patients into these key randomized trials that we’ve been talking about.

One very quick point I’ll add is that these guidelines and documents are great, but they’re going to need to be updated in a couple of weeks or a month if we do the right studies. We enrolled 1000 patients around the world on the remdesivir trial, 1000 patients at 70 centers in a matter of weeks. We enrolled 25 patients at Cedars Sinai in 10 days. My point is, these guidelines are great. They’re going to have to be malleable and updated, because things are changing so fast.

 I couldn’t agree with you more. In my view, these are living documents. In the COVID-19 era, these have to be updated quite frequently, because this thing is changing by the hour and by the day and by the week. Absolutely.

DVT/VTE Prophylaxis Guidelines 

If we do our very best and we can get these patients out of the ICU [intensive care unit], to the step-down unit, and eventually out of the hospital, our work doesn’t seem to be done. There seems to be a reasonable rate of patients suffering complications once they go home. Venous thromboembolism [VTE] seems to be front and center for that. Why is the risk of venous thromboembolism still elevated after patients are discharged from the hospital? Aren’t they better? Shouldn’t they be out of the woods?

There are a couple reasons for that. One is this markedly reduced mobility we see. We see quarantined DVT [deep vein thromboembolism]. People come into the hospital without COVID-19 [coronavirus disease 2019] because they’re staying home and getting DVT just from stasis. If you put that together with COVID-19 and then you put someone through hospitalization—when you put someone in ICU, they’re losing days and days and days in terms of getting back on their feet.

What this points to is that we really should study and be aggressive with postdischarge prophylaxis. Look at EINSTEIN CHOICE and AMPLIFY-EXT. Look at MARINER and MAGELLAN. We’ve learned that low-dose DOAC [direct oral anticoagulant] prophylaxis is very safe. The bleed rate of a low-dose DOAC might be the same as aspirin. The risk-benefit analysis clearly favors doing this, or at the very least favors studying it.

With higher-risk patients, I’m going to assess them. If they have concomitant cancer or if they have marked reduced mobility, maybe I’ll look at D-dimer at discharge. These are the patients I want to put on prophylaxis, but I hope we can study this too.

Absolutely. Alex, do we know what the rate is of VTE post discharge in patients with COVID-19? Or is it too soon to tell?

 That’s the other million-dollar question, isn’t it, Greg? What is the rate of VTE? One of the things we’ve done the last decade or so is have high-quality data to inform us on what a high-risk, medically ill patient looks like. We know that there’s a subset of these patients who are indeed at risk for developing severe VTE. Remember, in the medically ill population, when someone experiences a VTE event, it’s not the distal DVT that they get. It’s usually a central, major VTE, and an increased rate from the previous reports of VTE-related deaths. This is very different in terms of the first presentation of VTE and the surgical population, which tends to be younger and less comorbid and experience fewer cardiovascular morbidities. We know that the risk of VTE climbs dramatically in the first 3 weeks after hospital discharge and continues to about 6 weeks after discharge, and then afterward it plateaus. We’ve seen high-quality data from the MARINER trial. We’ve also seen high-quality data from the MAGELLAN and APEX trials. There are very consistent data.

Add COVID-19 on top of that, and in my view, a high VTE-risk population becomes a very high VTE-risk population, although we don’t have those data. We need those data desperately.

Something we’ve learned from key subanalyses is that we can also identify a low-bleed-risk population. If patients go through their hospitalization without a bleeding event, they are an excellent candidate for postdischarge prophylaxis. If they don’t have active cancer, if they don’t have bronchiectasis, if they’re not on dual antiplatelet therapies, if they don’t have other risk factors or a recent history of bleeding or active gastric duodenal disease, these all are appropriate patients.

Lastly, we just published a very important study with a lot of patients suggesting that certain key VTE risk factors—such as a history of VTE, a cancer history, history of thrombophilia, advanced age, and elevated D-dimer&mdash;are all very relevant in the COVID-19 population, either as separate risk factors or as part of a risk score, such as an IMPROVE [International Medical Prevention Registry on Venous Thromboembolism] score of 4 or more. These represent very high-risk patients who tend to have a 3-fold increased risk of VTE in the immediate postdischarge period and benefit from extended prophylaxis. In our institution, if you’re hospitalized with COVID-19 and you have an IMPROVE score of 4 or more or an elevated D-dimer or an advanced age, you’re getting postdischarge prophylaxis. Our regimen is what I call the MARINER regimen, which is 10 mg of rivaroxaban for up to about 4 to 6 weeks.

 That’s what I was going to ask you. That’s what you would do for your patients whom you feel are at high enough risk that they’re leaving the hospital and need something: 10 mg of rivaroxaban once daily.

 Yes. Of course, the minority of patients who are on certain antivirals that may interact with DOACs, and this is what Vic and you were saying earlier. There are some drug-drug interactions, so we have to be careful with the DOACs. Ritonavir or lopinavir, there are certain antivirals that would have some interactions that we’d have to be careful with. But the vast majority of patients can go home on a DOAC, as described.

I thought I’d mention that we need more data on JAK inhibitors, IL-6 inhibitors, baricitinib, tocilizumab. What are these drugs doing with this inflammatory state? Are they helping us with the thrombotic state? There’s more to learn here. Maybe we’ll find out in big studies if patients who’ve received some of these drugs and reduced their cytokine storm are going to do better and have less thrombosis.

 Vic, along similar lines, we talked about getting patients out of the hospital. What about our patients, especially these elderly patients, who end up in an acute rehab setting or skilled nursing facility? Same approach? Do you give them a DOAC or a low-molecular weight heparin or something else for extended prophylaxis while they’re there?

 I strongly suggest a DOAC and a careful risk assessment. If they’re older, they’re already right up there in risk. As Alex said, do the best risk assessment you can, but when you’re older, you’re already high risk. Risk-benefit analysis almost always favors anticoagulation protection versus the risk of bleeding, but that’s got to be assessed. Bleed risk always has to be assessed.

Venous Thromboembolism Postdischarge Risk 

What’s going to be the landscape following this coronavirus pandemic? Certainly, there’s renewed interest in extended thromboprophylaxis for the medically ill. Something that many of us were hoping for is that we would get people interested in medically ill patient prophylaxis, and now we have this pandemic doing that. How are things going to change? What do you foresee for extended-duration thromboprophylaxis in medically ill patients, even considering those who don’t have COVID-19 [coronavirus disease 2019]? Is this going to change? I’ll start with Alex and then go to Vic.

 Conceptually, I think of this as 3 phases. In the prehospital phase is a subset of high-risk patients who have never entered the hospital that we should study? Indeed, we’re trying to assess a randomized trial looking at this question, whether there’s a subset of the ill outpatient population with polyvascular disease or key risk factors that would benefit from a purely outpatient-based thromboprophylactic strategy. The majority of data in the medically ill prophylactic space in terms of a DOAC [direct oral anticoagulant] are with rivaroxaban [Xarelto]. The majority of data, especially data that are relevant to the COVID-19 crisis, relate to D-dimer elevation. We’ve seen that using D-dimer elevation and a thromboprophylactic strategy with low-dose rivaroxaban produces a net clinical benefit in terms of an appropriate risk reduction without an appropriate increased bleed risk. That’s the first phase. The second phase is the in-hospital phase that we just talked about. Is there a strategy that uses more aggressive, or even therapeutic, doses of anticoagulants that confer advantages?

And lastly, as you just said, there’s the posthospital discharge phase. In that sense, is there an optimal thromboprophylactic strategy? Can we interchange DOACs? Can we use apixaban versus rivaroxaban? I don’t know. A lot of the data are with rivaroxaban. It’s FDA approved. Apixaban is not. Can we use aspirin? I doubt it. Can we use long-term low molecular weight heparin, like enoxaparin 40mg? Yes, there’s the EXCLAIM trial that Vic and I were part of. But again, there was a lot of bleeding in EXCLAIM as well. If I had to guess, my sense is that maybe these patients would benefit from a dual-pathway inhibitory approach, something like low-dose anticoagulation plus an antiplatelet therapy. More and more data suggest that there is an antiplatelet-derived process, especially with oxidative stress, as well as the classic thrombin-derived process that is leading to the thrombosis. I would love to study a dual-pathway or COMPASS-type regimen in these patients with a low-dose anticoagulant and low-dose aspirin. Those are the 3 phases that I think should be evaluated.

We’re in the tabula rasa, right? This is brand-new territory. We can extrapolate only from what we know, but we know very little in terms of the thrombotic risk in the COVID-19 era. 

Alex covered it very well, Greg. I’ll echo what he said. I was thinking about COMPASS when Alex mentioned that. Maybe we need a COMPASS-type study for VTE [venous thromboembolism]. Maybe we need a low-dose Xarelto, a low-dose DOAC, together with an antiplatelet drug in everybody. Maybe we need a huge study of 50,000 patients. Maybe we’ll see that many of us in the population could really benefit from this, even if we don’t have any obvious or visible risk factors for VTE. Something like that should be done. If quarantining goes on or comes up again, we may have to look at patients who don’t have COVID-19: people who are getting around and pretty healthy but who are sitting around a lot more. We might look at telemedicine and things like that. How do we study those folks?

Alex covered those things well. I’m going to finish with an optimistic point and say there’s so much going on right now. Most of our protocols at Cedars Sinai have been cut off for COVID-19 protocols. We’re using COVID-19—VTE protocols and other COVID-19 protocols, using a lot of different drugs. What about BCG? What about an actual killer cell? There are a lot of things we’re studying right now. We’re going to get a handle on this, but it’s going to come around again. It took us millions of years to evolve as humans to get our genome where it is. It took this virus a couple of months, a few weeks, so it’s a scary time. We’ll get a lot of antiviral funding, and we’re going to learn a ton in the future here.

I agree. That brings us to the end of what’s really been a wonderful discussion with a lot of terrific insight to how we approach this pandemic. One of the things that lifts my spirits a little, as I try to keep my children from assaulting each other during this no-school period, is the fact that we have so many great minds working on this. Our research infrastructure is so strong that if there are inroads or handholds that we can find to boost ourselves up and improve care in this setting, we’re going to find them. I’m encouraged by that. Getting to work with you, Vic and Alex, on these things also makes me feel good, and we all need a little bit of that right now. I want to thank you both for being amazing friends and even better faculty. We reviewed great information.

I want to thank everybody who watched this 

® presentation. Hopefully this will entertain you for a little while at home as you’re staying locked down. We found this to be fantastic. I learned a lot, and hopefully we’ll do more of these 

® discussions focused on COVID-19 and how we can care for these patients better. I want to thank all of you.

Antithrombosis in Patients with COVID-19