Activity Transcript

Jean M. Connors, MD: Hello, and welcome to this Medscape education session. We are going to discuss monitoring and managing coagulopathy in patients with coronavirus disease 2019 (COVID-19). I'm Dr Jean Connors, Associate Professor of Medicine at Harvard Medical School and a hematologist at Brigham and Women's Hospital in Boston, Massachusetts. I'd like to introduce my esteemed colleague, Professor Hugo Ten Cate, who's Professor of Medicine in the Department of Internal Medicine at Maastricht University in the Netherlands.

Hugo ten Cate, MD, PhD: Thank you, Jean.

Dr Connors: Thank you, Hugo.

I'm going to give you a brief overview right now of the material that we're going to cover during this session. We're going to look at COVID-19 and coagulation dysfunction, particularly with an eye towards inappropriate microvascular thrombosis. We will discuss testing for the coagulation dysfunction, and we'll summarize what we are seeing in patients with COVID-19 and how we can approach their management.

We're going to start by demonstrating what we mean by coagulation dysfunction and inappropriate microvascular thrombosis. I'm going to ask Hugo to discuss what we know about COVID-19 and coagulation dysfunction and how does this work and how does this occur. Hugo?

Dr ten Cate: Thank you, Jean.

We have to quickly go to the start of the process with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus entering the lungs, which is the primary route of entry in many individuals, where it meets the alveoli and the alveoli cells -- the epithelial cells that are susceptible to infection with viruses such as this SARS one because they express the angiotensin-converting enzyme to a receptor that is able to interact with spike proteins at the surface of this virus that mediates the uptake of virus and the infiltration in the cells.

By doing that, a local inflammatory response will be induced that involves T-cells and alveolar macrophages as indicated in the left-hand side of this slide, and they try to limit the viral invasion. And in many subjects infected with the SARS virus, this will be successful because we know that, fortunately in a majority of cases, this infection goes rather mildly and passes without much symptoms. But in a minority of subjects that are susceptible to this infection (we are mostly talking about elderly individuals and males more than females, and in particular, the ones with co-morbidity)...in those subjects, the risk of dissemination of the virus but also the elicitation of systemic inflammatory response syndrome (SIRS) as indicated in the right-hand side is very likely to occur. This involves more massive cellular responses and the engagement of other types of cells including neutrophils, monocytes and macrophages releasing a number of pro-inflammatory cytokines. And in the worst case this may involve the release of what we call the cytokine storm, that can be an overwhelming inflammatory reaction.

In all cases on the right-hand side, and to a certain extent also in limited disease, inflammation goes hand in hand with coagulation activation. This is a principal that we refer to in the remainder of this conversation as "thrombo-inflammation." Because what we've learned from this viral infection is that it triggers mostly thrombotic phenomena due to a close interaction of inflammation and pro-coagulant mechanisms.

To summarize this in the next slide where we refer to it as "coagulopathy," I think it's important to note that during the course of this pandemic, we, as physicians, have learned quite a bit about the typical features of this coagulopathy, which is clearly different from what we knew in other severe diseases like sepsis. There is no clear indication of disseminated intravascular coagulation with consumption of platelets and coagulation factors, but there is a profound inflammatory response as illustrated by high IL-6 levels, strongly elevated fibrinogen levels, C-reactive protein and concurrent panel of changes in the coagulation status reflecting pro-thrombotic rather than bleeding or reentered changes. These are summarized in the bottom of the slide where you see the inflammatory-mediated responses in fibrinogen factor VIII, the endothelial release of von Willebrand factor, which can also be profound, the activation of the contact system and the intrinsic coagulation pathway through the release of NETs and through the interaction with platelets that activate factor 12 with polyphosphate as mediators, the activation of platelets and endothelial dysfunction -- it all leads to thrombo-inflammatory in a rather profound manner. And to look at some of these features in greater detail, I want to ask Jean to take us through some of the microvascular findings in autopsy studies. Jean.

Dr Connors: Thank you, Hugo. So, in the next few slides we'll look at some of the events demonstrated on autopsy findings that Hugo was discussing. We can actually see in this panel from Varga and colleagues in Lancet that not only do we see viral inclusion and viral infection in some organs, such as in the electron microscopy pictures on the far left of the kidney, but we also see invasion of both the small bowel and the vascular endothelium in the lung with mononuclear cell infiltrates causing inflammation and disruption of function that way, as well as frank apoptosis of the vascular endothelial cells indicated by caspsase 3 staining -- which means that we're losing the vascular endothelial cell protective effects against the formation of clot. The vascular endothelial cells contain a number of anti-coagulant mechanisms including nitric oxide, which dilates the vessels, thrombomodulin, protein C, tissue factor pathway inhibitor, and even the glycocalyx itself prevents the activation and formation of thrombus on the vascular epithelial cells. When these cells die or became inflamed or infected with either virus or infiltrated with the mononuclear cell response to inflammatory cytokines, we find that we do not have that protective effect.

As seen on the next slide, this endothelialitis manifests by interstitial and perivascular lymphocyte infiltrates (these blue dots throughout the left panel in the alveoli.) The alveolar walls are thickened and inflamed and edematous. But we also see the development of microvascular thrombi in the alveolar capillaries. And this is the combination of the events that Hugo described with the activation of intervascular coagulation as well as the loss of the protective function due to endothelialitis in the microvasculature of the alveoli, which hinders gas exchange in these patients and is responsible for the hypoxemic respiratory failure that we see in patients who have severe COVID-19 infection.

Now, Hugo, how do we recognize the inflammatory response in these patients? Are there biomarkers we can look at or how would we approach understanding the degree of inflammation in a patient with COVID-19 infection?

Dr ten Cate: Thank you, Jean. That's an important element and it has been addressed by a number of groups. This is shown in this slide from the Wuhan study that looked at the panel of pro-inflammatory cytokines and observed that they were all elevated. You have to be aware that these studies were mostly done...or all of them were done in patients in critical care units, so these are all very sick individuals. Distinctions can be made, for that reason also, in biomarkers in relation to outcome, specifically mortality. That's what you see in the right-hand side for a number of these biomarkers. On the far right side, you see the picture for IL-6, which is more elevated in the patients that did not survive (in orange) as compared with the survivors (in green.) In the same picture you'll observe in the next slide from another Chinese study. I would like to focus on the D-dimer in the left, upper part panel because in these serial measurements we observed that also clotting activity measured with the D-dimer fragment, which is an indicator of the presence of polymerized and cleaved fibrin molecules in the circulation. We will come back to discuss this a little later. You see marked rises in patients that did not survive (in orange) as compared with the blue curve in patients that survived.

In fact, in the next slide you see this other Chinese study published in the Lancet, that already at baseline, on admission, patients with an elevated D-dimer level had much greater odds ratio for mortality than the ones with not so much elevated D-dimer level. Also, the progression -- that's illustrated in the left upper part of this panel that shows the D-dimer course taken over subsequent days -- is much steeper in the non-survivors as compared with the survivors. In this panel otherwise, you see a number of other biomarkers -- IL-6, troponin, which is at the bottom and on the right-hand side. There are other markers including serum ferritin and LDH as markers of organ damage or acute phase reactants. So, if we come back then to discuss...and I introduced this slightly with the D-dimer fragments...but it's the tip of the iceberg, basically. Jean, would you guide us a bit more through the coagulopathy from the hematologist's perspective? What are we looking at?

Dr Connors: Thank you, Hugo. I think, we, in the United States could observe early on (at least in Boston), before we were hit a deluge of COVID-19 positive patients...that the coagulation changes that were seen based on the data coming out of Wuhan and then from Italy. Initially, D-dimers, as you discussed, were noted to be elevated and a marker for severity of illness and mortality.

But in this paper from Tang and colleagues in J Thromb Haemost, we can see that there can be a difference not only in the rise of D-dimer between patients who survive and those that don't. So if we look carefully at this slide, measurements of fibrinogen in concordance with D-dimer and prothrombin time (PT) in partial thromboplastin time (PTT) can distinguish over time, and changes over time, between those who were more severely ill and do not survive vs those that survive. If you look at the top left-hand panel, you can see that days after initial admission, patients whether they will survive or not, have similar elevations in fibrinogen that are well above the upper normal of limit. Over time, these will persist...although in some patients and in this series...in the patients that did not survive, it appears that they had a drop in fibrinogen, which may be associated with disseminated intravascular coagulopathy (DIC) in a very small subset, 21 non-survivors compared with 162 survivors.

We can also see, as Hugo has pointed out in the previous works, that the D-dimer level increases dramatically in those that do not survive compared with those that do survive. You can see from the bottom 2 smaller panels that the PT, although initially similar on presentation, will increase in those that are not doing well and are more severely affected and most succumb to COVID-19 compared with the PTT which shows very minimal elevations or changes. When we discuss D-dimers, I think it's important for many people to understand exactly what a D-dimer is. Because I know in dealing with the hospitalists and anesthesiologists at my institution, the D-dimer concept was not very well understood. We want to point out the D-dimer assays are not standardized. Unlike an INR, there is no one standard or correlation of tests across the world. So, when you read reports you will see different units and you need to know what the threshold cutoff is for normal, for a particular test. A D-dimer is produced after fibrin is crosslinked in clot formation by factor 13 and then cleaved by plasmin. So, it's a fragment of this crosslinked fibrin monomers that is only produced when there's active coagulation and active fibrinolysis. The point of this slide is to note that D-dimer can be elevated in conditions other than venous thromboembolism (VTE), and in this series from an Italian emergency room over 1600 patients who had D-dimers measured for presumed VTE. Only a small fraction ended up having VTE and patients had other diagnoses that you can see across the bottom on this busy bar graph. A significant proportion of patients had elevated D-dimers simply due to infection, and this does not have to be COVID-19 infection.

In the next slide, we can also see that the levels of D-dimer in intensive care unit (ICU) patients can vary. It appears higher in patients who do have positive pulmonary emboli imaging with computed tomography (CT) pulmonary angiograms. But sometimes, it can be difficult to distinguish whether or not the elevation in D-dimer is due to a macrovascular thrombosis such as pulmonary embolism (PE) or it's simply part of the thrombo-inflammatory microvascular process. So, it's important to image people if you can as that may affect your treatment. So, we're now going to move on to testing for coagulation dysfunction.

As you can see from this guidance from Marcel Levi and colleagues, they have outlined a diagnostic approach and therapeutic approach. At our institution we manage these patients, particularly those in the ICU, by measuring D-dimer, PT and platelet count every 3 days. We will also monitor fibrinogen in those patients on anticoagulation. We found this helpful not only to monitor for severity of disease, but also to potentially screen or scan patients for VTE. Hugo, any similar strategies at your institution with regard to managing these patients and their coagulopathy?

Dr ten Cate: I fully agree with this kind of monitoring strategy and I think all of us applied it to the same extent.

I was thinking about the situation of desaturation...that you also mentioned somewhere in the nice Blood paper that you published...as a trigger point for looking carefully into the possibility that an acute VTE may complicate a COVID infection.

In addition to a rise in changes D-dimers, also such subtle changes, in oxygen saturation for instance, turned out to be helpful triggers for thromboembolism. We did a small study with CT scan in patients admitted to the ward in which that was used as a trigger and we found in 30% of those venous thromboembolic events. We have to carefully consider these different biomarkers for their value and learn how to interpret it to manage our patients in time, when possible.

Dr Connors: Yes, Hugo. You bring up an excellent point about in how difficult it can be to identify macro-vascular thrombosis, pulmonary emboli, or even deep vein thrombosis. We need to have a heightened awareness that these can occur either due to clinical findings that change, such as oxygenation or tachycardia, right heart strains seen on echocardiogram.

I know some institutions use a D-dimer threshold to guide a dose of prophylactic anticoagulation, we felt it's important to make decisions based on clinical status outside of these coagulation tests. So, there are a wide variety of ways in which to approach these patients, and that is because we really have no data about how to manage these patients. It's very important that everybody who is involved in the care of these patients try to engage in randomized clinical trials of anticoagulation so that we can actually get the answer to what dose of anticoagulation is sufficient to prevent thrombosis, particularly in those that are critically ill in the ICU.

That brings us now to our summary and closing remarks. COVID-19 really results in multiple hits that result in the thrombotic pathology. It really hearkens back to the Virchow's triad. Although a lot of press is focused on thrombosis and arterial thrombosis, we're seeing also this micro-vascular thrombosis which we can only appreciate on autopsy. These multiple hits involve the marked elevation in pro-coagulant proteins that destructed and destroyed anti-thrombotic protection afforded by vascular endothelial cells that we demonstrated in a variety, in both diagrams and autopsy findings in our presentation today. We observed stasis due to the inflammatory infiltrates and the small capillaries of both the alveoli and other organs. ICU patients appear to be at higher risk because they are simply more infected. They have more inflammatory response and many of them are paralyzed and mechanically intubated leading to stasis. So, we have both macrovascular events and microvascular events occurring in these patients. Monitoring them is important as it can help shape care and determine the direction the patient is going. The coagulation tests can actually give you an indication of the severity of disease in any particular patient.

Hugo, I would like to thank you for joining me in this discussion today. I would like to thank the audience for listening and thank Medscape for inviting us. Thank you, Hugo.

Dr ten Cate: Thank you, Jean.

This is a verbatim transcript and has not been copyedited.