Thank you very much. Again, my name is Eric Schmidt. I am a physician scientist. I see patients at Denver Health. And I have a laboratory at the university.
And so today I am taking the con side of this argument that we should not be empirically and regulating patients. And rephrase it, empirically, systemically anti-coagulating patients who have severe COVID-19, even if they're in the ICU.
And upfront, I have no conflicts of interest. I pored over my cognitive biases, which of course we all have cognitive biases, but um, my lab is interested in heparin sulfate, which is a kind of heparin. We actually are designing structurally defined heparin treatments for organ injury. So that's a cognitive bias.
Also, one of several side investigators at Denver Health for a randomized study of salvage fibrinolysis. Now, both of these, I think, actually would, if anything, bias me against my point here. So I think we're in the safe area.
So what I want to do today is I'm going to start a little bit about what we know about coagulation as a contributor to ARDS. And there's been a lot of research in this, albeit in the pre-COVID-19 era. There's been animal models and there's been human studies.
And then I'll address some common arguments about how is COVID-19 somehow different? Is there a unique coagulopathy that's involved? Or is there some sort of unique vascular basis? Or maybe the risk of venous thromboembolism is so high, there's something unique about this disease compared to other pre-COVID-19 ARDS. And I try to go through this and suggest there's not a clear evidence that this may be all that substantially different.
And finally, I want to finish up with this concept of pandemic exceptionalism and basically describe how in our desperation and our valid desperation to treat this disease, we have to be careful not to do things that may hurt patients.
And how adherence to the scientific principles we always do, in fact, is the best way to save lives.
Okay, let's talk about coagulation. Coagulation, obviously, is really complicated. There's been a ton of work done. I'm going to try to summarize it using this schema of a lung. And it's not to scale, obviously. There's the alveolus for type 1 cells and type 2 cells.
There's a not to scale pulmonary capillary, although both Todd and I love the pulmonary capillary so much that's to scale in our minds.
During critical illness, you get activation of coagulation, specifically in the vasculature, and that's a problem. You actually get activation of the extrinsic cascade.
And tissue factor, it's often circulating as microparticles, basically binds to factor 7 and then activates thrombin, which leads to fibrin deposition. That leads to plugging of micro vessels, which is bad.
Furthermore, you get abnormalities in processes that govern coagulation. You get loss of tissue factor pathway inhibitor, loss of antithrombin 3, loss of activated protein C. You also get inhibitors that are produced that prevent breakdown of clot. So you get, once clot forms, you get impaired fibrinolysis. The clot sticks around and causes harm. So this is all a bad thing. It's been consistently shown to be a bad thing.
The challenge is the exact same process occurs within the airspace. It has a completely different effect.
It's been shown over and over again in animal models, of course, that it actually is protective.
This exact same process is occurring, and this has been shown, for example, in influenza models.
In influenza models, if you take a mouse and you give it influenza, the presence or absence of tissue factor in different compartments of that lung can influence the mouse's response.
If you take a mouse that the whole body, there's tissue factor deficiency, these mice hemorrhage like crazy into their lungs.
If you limit that tissue factor deficiency to the alveolar epithelium, they also hemorrhage into their lungs.
If you limit it to the myeloid cells or endothelial cells, there's no real harm seen. So there's this compartmentalization phenomenon.
Clot's good in some parts, clot's probably bad in other parts, and that makes it really difficult to treat, and that probably explains why we've had such trouble treating this in human studies.
And this is a complex slide, which means, of course, I shouldn't be showing it, but I'm trying to squish together every multi-center, randomized, controlled trial of coagulation in either ARDS or a critical illness that leads to ARDS.
It includes things like tissue factor pathway inhibitor, antithrombin 3, activated protein C, heparin, and rhombomodulin.
Now, each one of these studies improved coagulopathy. They got the coagulopathy better, but none of them, with the exception of one, I guess all of them except one, showed no difference.
There's no difference in patient outcomes. The only one that had a difference was the activated protein C, Zygres, the PROWIS study in 2001, which of course then was not replicated in 2012 in PROWIS shock.
So the challenge is, even though coagulopathy exists in ARDS, and we can make it look better with anticoagulants, it doesn't translate to better patient outcomes.
With that in mind, is it possible, the answer is yes, it is possible, but is it possible that COVID-19 is different?
COVID-19 is much worse than its previous ARDS, and we should expect that anticoagulant efforts may improve outcomes.
And so I'm going to address three arguments to that. I'm going to try to put those into a little bit of perspective.
One is people raise the fact that COVID-19 has this profound coagulopathy. It's different than ARDS. And so this initially comes from a lot of studies that report D-dimer.
D-dimer is a very primitive way of measuring coagulation, but it's a commonly used one. And so a lot of studies report D-dimer. So D-dimer is the breakdown product of cross-linked fiber.
It's commonly used to predict deep venous thrombosis or venous thromboembolism. If you're not a clinician, we'll use this DBT or VTE kind of interchangeable.
And so since people use it clinically, it's easy to report in case series. And one of the first big ones that came out was from Wuhan.
It's from Ning Tang and colleagues, and this group has actually published a lot on this from Wuhan looking at coagulation.
This study at the Journal of Thrombosis and Hemostasis had 449 patients with severe COVID-19. 350 of them had no heparin exposure at all, no prophylactic heparin at all.
That's actually not a normal. The risk of DBT is very low in Southeast Asia, and so it's not standard of care to do always to do subcutaneous heparin prophylaxis, although it's under some debate.
So all these 99 did receive some form of prophylactic heparin or low molecular weight heparin.
Now if you look at the whole group, 449, D-dimer at study entry or observation entry predicted worse outcomes. If you had a higher D-dimer, you were more likely to die at age 28.
And so that's interesting. It suggests maybe there is a coagulopathy that drives mortality. And furthermore, the study looked back retrospectively and thought if you had
screamingly high D-dimers, greater than six times up living a normal, the patients that happened to be on prophylactic heparin, which we tend to do for all of our patients in the United States, those patients did better.
And this was one of the studies that really prompted enthusiasm for understanding coagulation and COVID.
Now the problem with all these studies, you always need to ask what's D-dimer in normal ARDS? The answer is we don't quite know. We do know very well from sepsis studies.
From sepsis studies, including the ones I've shown you, this is medians and interquartile range in the prowess study upon entry. This is just the placebo group, but since it's randomized controlled study, there's the exact same thing in the intervention group.
But basically the medians anywhere from 3,000 to 4,000 interquartile range is vast. That's in sepsis. And one of these studies, the thromomagin study, which is scarlet, had no benefit. The other study prowess did have a benefit, but since been refuted.
Here's what it is in COVID-19, at least in three major studies. The Tang one is the one I already mentioned, which has a wide range, but there's one published in Jam Internal Medicine, and the one in New York City, again, the D-dimers were not significantly elevated.
So if there is a coagulopathy based upon D-dimer alone, it doesn't look much worse than what we see in sepsis. Now the problem with this is D-dimer is a crummy measure of coagulation.
A much better measure is thromboelastography. If you're not familiar with this, this is a way of really measuring clot and whole blood. He basically placed the blood in a cup. There's a spinner that spins over time.
Tension forms on this, and this tension basically is mirrored. You get this waveform, and looking at that waveform, you can ascertain many things about clot and connect.
So here's an example of thromboelastogram that was shared to me by Peter Moore, who's a fellow in our division and is at National Jewish, basically looking at a patient with COVID-19 and one of the thromboelastogram traces.
You can see you can learn a lot about clots. You can learn about clot initiation in the R-time, clot initiation in the K-time, propagation in the angle, max amplitude, clot strength, and then lysis at 30 minutes. If you wait long enough, you can see if it lysis or not.
And so what is known about thromboelastography in COVID-19? Well, to date, and as of this morning, there are only two, I believe, two studies. One was in Milan.
It's three patients, and you can see their clot initiation was normal. Their clot amplification was normal. The propagation was almost at the upper limit of normal, but the clot strength was really robust.
What's also kind of striking is these things lysed pretty aggressively. It was at the upper limit of normal, which is not seen by other people.
In fact, today, as of this morning, there was an article that was accepted for publication at the University of Colorado by Frank Wright, and Peter Moore was a co-author on this, which saw very similar things.
That the R-time was okay, the clot propagation was okay, but the clot strength was elevated. And in contrast, the clot does not lice easily.
You can't get less than zero, so it's at the lower limit of normal.
So this suggests the problem in COVID-19 is the clot, when it forms, is particularly robust, and that can be things like platelets are involved, the fibrinogen, and the clot does not really lice.
That's something that suggests fibrinolytics. None of those would be specifically targeted by heparin.
Now, one of the things I want to bring up, though, is how does this compare to critical illness?
And there's no thromboelastography study of ARDS, but people are doing these studies in sepsis, and I know that because we are.
We have myself and Nate Shapiro of an R01 looking at thromboelastography in sepsis, not designed to look at clotting, actually, but designed to look at endothelial function.
This will be found in sepsis. In sepsis, the R times relatively normalish, K times relatively normal, the clot propagates quickly, and clot strength is almost abnormal.
The point is, this isn't much different from COVID-19. In fact, actually, there's probably an impairment in clot lysis, too.
Now, there may be a chiropractic, but at least from this, to me, it's not 100% clear. It's much different from what happens in sepsis.
Of course, sepsis has never benefited from anticoagulant therapies.
So, again, coagulopathy probably does exist, but I don't think you can be convinced that it's much different from critical illnesses that we see based on what's been published.
Now, what about this vascular concept? A lot of people suggested this is a different kind of ARDS, and it basically involves vasculature.
And as somebody who loves the pulmonary vasculature, actually, I'm endeared to this concept, but I want to talk a little bit about how that does not necessarily prove that it's clot.
This comes from Gattinoni's numerous articles. There's two of them. He's actually had four so far, all based upon a case series of 16 patients that was in the Blue Journal.
And he highlights this L-type, L being low recruitability with PEEP and high compliance, versus an H-type, a high recruitability of PEEP with low compliance.
Now, people have been more fascinated with the L-type because that seems to contradict our vision of what ARDS is.
And furthermore, Gattinoni stressed that these patients were quite hypoxic.
And he estimated shunt fraction in these and found, although it's a little unclear because there's only 16 patients, there's more than 16 observations of this,
but that some of these patients had elevated right to left shunt fractions.
Hypoxemia was out of proportion to what you see in the world.
And that says maybe these patients are having clots, maybe they're in PEs on top of it.
And is that true? Well, there's a couple of things. First of all, it's important to emphasize that Gattinoni's L-type versus H-type is not new.
He described this in 2006. He basically had an L-type in usual ARDS, and an H-type in usual ARDS.
And the shunt fraction, while it's not as high as COVID-19, is still pretty substantial in both cases.
Furthermore, other people have not been able to replicate this L-type phenomenon.
There's the Boston study, which carefully looked at the rest of the dynamics.
You see a normal distribution of compliances that stays stable throughout the course. It doesn't evolve over time or change.
And the hypoxemia is consistent with what's seen in other ARDS studies.
In order to this observational study, this is regular ARDS.
Maybe there's suggesting that maybe this L-type could be exceptions.
So let's assume, though, there is a vascular basis, that there are some patients who have this unique hypoxemia that's so severe, something's going on in the vasculature.
That doesn't mean it's a clot.
And there's this really cool article, and I show it to only three patients, which shows you how low the threshold is to publish something right now.
Three patients who had CT scans for severe hypoxemia and COVID-19.
At Mass General, they do these dual-energy CT scans, which allows you to look for both clotting and also blood flow.
And what's interesting, they claim this, although I think it's open to debate.
None of these patients had PEs, but they had profound dysregulation in the way the blood was flowing.
You had dilation of vessels over here, which could just be hypoxic vasoconstriction. You massed the dilation away from it.
Always, that leads to shunt physiology.
So this is a vascular problem, but it's not a clot vascular problem.
That could explain the L-type.
Okay, the last one, and this one, I think, has gotten a lot of buzz in the past 72 hours.
And I think this actually is probably the best argument against me, which I shouldn't be doing, but is maybe something about venous thromboembolic risk is so sky high in COVID-19 that it's a different disease.
Now, to understand that, we need to frame a reference.
What is the medical disease we take care of in the ICU that has the highest DVT rate?
And believe it or not, it is stroke, I would argue, it's stroke.
There is a study called the CLABS III study in Lancet, which had two arms, roughly 1,400 patients.
So just in the placebo group, which had 1,438 patients, these patients had stroke that was bad enough that they couldn't get out of bed to go to the bathroom on their own.
So not terrible stroke, but just any stroke.
Despite subcutaneous heparin and other pharmacologic prophylaxis, these patients, at seven days, had a clot rate of 12.1%.
That's high, but we don't anticoagulate these patients.
And why not?
Well, the reason we don't anticoagulate them is because, I'm going to guess, it's the risk of hemorrhage, of hemorrhagic conversion.
So the risk-benefit ratio in stroke suggests we do not empirically, systemically anticoagulate all our patients.
What about COVID-19?
Well, I'm not going to go, there's a Journal of American College of Cardiology article, which I'm sure everyone's heard about.
There's a bunch of metallogic and potential bias issues in there.
I think it's a really interesting article.
We can probably talk about that later.
But I think a more representative and one that probably is more applicable to our experience was, again, that Boston study.
We're on the clot rate of 22.7%.
We're probably in the mid-20s, I would estimate, at the University of Colorado.
That's higher.
That's twice as much.
So is the risk-benefit ratio of anticoagulating, is it favoring anticoagulation, or is it not?
What's the risk of anticoagulating these COVID-19 patients?
And the answer is, we don't know.
In fact, most societies are silent on the issue.
They dodge it.
The Surviving Sepsis Campaign put out guidelines for COVID-19.
They don't mention it.
British Rassica Society says just do prophylactic heparin or noxaparin.
Journal of American College of Cardiology itself basically was really vague about it, but said the majority of people said just stick to prophylactic anticoagulation, like we usually do in our critically ill.
This is a dynamic thing, and it may change, but the point is we don't know.
We don't know the risk-benefit.
So what I would say right now, as of today, and this is an incredible time, so things may change tomorrow, I think we feel pretty comfortable that COVID-19 is associated with the coagulopathy.
However, other critical illnesses are associated with coagulopathy, and those have not gotten better with anticoagulation.
Some centers suggest this L-type phenomenon, but it's not universal, and it doesn't really mean that there's clotting going on.
Finally, I think we can agree that there's probably a higher venous thromboembolic risk.
We don't know, however, is clotting the lung protect us from viral infection?
Animal studies say it does.
We don't know if actually treatment dose heparin will accordingly improve outcomes at all.
Furthermore, we don't even know if heparin is the right anticoagulant.
Is the problem more platelet? Is the problem more fibrinolysis? Should we be doing fibrinolytics instead?
And finally, we have no idea what the bleeding risk is, and we're obviously swayed by anecdote heavily in that situation.
There are a number of, I'm looking at the data at the University of Harvard, there are a number of pretty big bleeding events that have happened.
So we just don't know. And so what we have here, I would say, we have equipoise.
We have equipoise in a disease that's common right now, it occupies our thoughts, and a pretty common event rate, which is DVT, let's say.
So this is prime for us to do clinical trials.
It's not prime for us to deviate from our standard of care, it's prime for us to do clinical trials.
Now the argument against that is this, is that you're in a crisis.
You don't have time for clinical trials.
And I'm a basic scientist, so I'm not a clinical trialist, but I'm going to defend the thought that I'm going to push back against this assertion.
And this comes from this concept of pandemic exceptionalism.
I take this from two different articles.
One has pandemic conceptualism in the title, one from Science, one from the New England Journal, that remind us the following.
In times like this, in pandemics and epidemics, what saves lives is doing good science, not opposite.
If we abandon our principles, we put people at risk.
And they give two, basically two examples.
One is in SARS, there's ribavirin, basically was very promising, but ended up hurting people.
Suggesting that what we think might work, may not.
That's a common theme in critical care.
The other thing you want to keep in mind is that if we do garbage, if we do garbage studies right now, it doesn't help us in the future.
And in SARS, and actually in Ebola, there are a ton of studies on steroids, but people were reluctant to randomize the placebo.
They thought it was unethical, and it led with a whole bunch of studies that were utterly uninterpretable.
We could use that knowledge right now.
In fact, if SARS-CoV-2 is seasonal, we really need to figure this out to save future lives.
So abandoning our scientific concepts now hurts people.
So that's my argument against empirically anticoagulating everybody with systemic anticoagulation.
One is it hasn't helped in non-COVID ARDS.
Two is I'm not 100% convinced by any argument that this is different than previous kinds of ARDS.
And the VTE risk, while substantial, we don't know the risk benefit yet.
And finally, what we should do is we should adhere to the scientific norms that have gotten us far.
And that's the best way we can help.
So that's my conclusion.
Great. Thank you so much.
And next we'll have Todd, who will be talking about PRO, arguing the PRO for anticoagulation.
Perfect. And I am unmuted.
Great. Well, I want to thank Amin for this kind invitation to speak on this highly interesting, one of my mind,
one of the more interesting topics that's coming across in the COVID realm because it involves the vasculature and particularly the pulmonary vasculature.
I share that aspect with Eric, my fascination in this particular area.
And I'm excited to sort of delve into it.
And then when Amin asked me to do this, it was a great, perfect, I love to talk about that, who would I be debating?
And then she mentioned it would be Dr. Schmidt and then immediately had some level of apprehension.
I think we all know Dr. Schmidt well.
Well, I know him. I know him to be a fantastic clinician, a fantastic teacher and also an excellent researcher.
A true triple threat within this group here at this point.
He is a recipient of multiple R01s. He has JCI papers and is interested in particular in this area.
He is an exceptionally well-trained, comes from Hopkins before he joined us.
He was, did his residency there as well as his chief residency.
And then I think what, so trained in the land of Osler, what many may not recognize during his time there.
He's also captain of the Hopkins polo team.
Hopkins being, again, one of the premier institutions in town or in the country, had a refined approach to it.
This is popular in inner city Baltimore.
Dr. Schmidt, top player on their polo group here at this point.
Myself, certainly am proud of my training here as well.
Though a little bit more casual in our approaches to horses there.
I trained at UT Southwestern, a very exceptional place to train.
Well, we did not have Osler. Chuck Norris is from Dallas.
So we had that going for us as well.
Again, didn't wear quite as tight of pants when riding horses around in Dallas.
And so we heard really exceptional arguments from Dr. Schmidt.
And while I did a great job and trained in great places, unfortunately, he's on the wrong side of this debate here at this time.
Which I think I'll show he pretty believes in.
So really, I kind of get that he laid great groundwork for all of this.
I think the question that comes up here at this point, and he mentioned it many times, is really is SARS-CoV-2 or COVID-19,
is this something new and different we're getting at?
Because I think, you know, if this was just another aspect of critical care of ARDS,
certainly this debate would probably not be as front and center as it has been.
And the argument I think we'll try to make is this different.
Really, you know, unfortunately, at this juncture have to come from our observations.
So these are observational data here at this point.
So is there, in fact, a higher incidence of venous thromboembolism in this disease state as compared to others that we've seen previously?
And this, again, we're reliant on observational information.
So here are reports.
And a myriad of them have come from initially out of Wuhan, but I've highlighted a couple of others here at this point
where we see a up to 27% incidence in patients.
This is still going on.
27% incidence in patients out of the Netherlands for VTE associated in their ICU.
In France, we see multiple reports of the same aspects of it.
20% incidence of VTE in patients prevalent in their ICU.
These are numbers in about the hundreds range when they're looking at them initially in terms of patient populations.
In the second study in France, they compared this against historical VTEs,
which compared against influenza where they expected around a 7.5% incidence in the preceding year for patients.
So are we looking, in fact, in the doubling of venous thromboembolism?
And Dr. Schmitt conceded this.
He did see a higher incidence of VTE within this group.
He mentioned this as well.
And this, I have to say, again, we're relying again on case series, but is striking, I think,
and speaks to the difference that we're dealing with in terms of this disease and COVID-19.
These are, again, a case series published in the New England Journal, again, looking at patients now with CBA.
And what we're essentially seeing, and I think many on the call who have served time in COVID units may have seen this as well,
I personally have, is young patients presenting with large vessel strokes.
And so this was a series of five patients published in the New England Journal, all less than 50 years.
And these patients aged, ranged from 30 to 50, so 30, 41, 47 years of age, again, presenting now with large vessel strokes.
Now, this was in a two-week interval that this was one of the author's attempts to define historical controls to this,
went back and looked at two-week intervals in the preceding year and expected to see an incidence of about 0.73 patients as opposed to this five incidents.
And this is, to me, again, clinically, or as a clinician aspect of this, we see a strike to see this in my own patient population.
In my last tour in the COVID ICU, we had two patients, both relatively young, with cerebrovascular accidents that occurred.
And of course, here is part of our conundrum because of the potentially devastating nature of this occurrence.
You know, the survey, many patients have left neurologically devastated versus death, which would be preferable.
We look at this in, for example, TPA studies. It is the former, which is the more abhorred outcome, if you will.
So the question that was raised, OK, are there other markers of coagulation? So this is, again, Dr. Schmidt addressed this,
the elevations of D-dimer that have been noted to the point we are now actually screening many of these patients because of early reports out of Wuhan,
D-dimer being a marker of adverse outcome in this patient group.
Here initial report that a D-dimer greater than one microgram per mil, that's about a twofold increase, was associated with an odds ratio of 18.4 in terms of mortality.
So, again, marked potential worsening of outcomes from this.
D-dimer levels, again, on admission out of Wuhan of greater than two in a retrospective study, again, showing marked increased mortality.
If you did an ROC curve on this from this group, we see a 93% sensitivity, 83% specificity for marking mortality.
Of course, all sorts of problems using this as a marker.
This is certainly we know clinically that D-dimer is nonspecific and is associated with many other adverse events.
But does this factor in as, for example, a surrogate, if you will, not a true biomarker, but a surrogate in terms of identifying who one might think about systemically anticoagulating going forward.
I can have one here. So Dr. Schmidt had turned back to history on these aspects of it.
And it has been recognized for decades now that ARDS is associated with abnormalities in the pulmonary vasculature.
You alluded to this in earlier studies, talked about other other, I was looked at it.
This is work from Zay Paul and Reid, essentially looking at autopsy studies we see in panel A, sort of a normal vasculature from an age match control, a patient panel B with an aspiration event that resulted in ARDS.
And again, you can see this marked pruning of the vasculature that had occurred, you know, raising questions.
Is ARDS inherently also a pulmonary vasculopathy, if you will?
Here they show the in situ thrombosis that can occur.
And of course, there's been great interest in pursuing this for some time.
Here we see work in the early 2000s from Dr. Mathay and his group looking at dead space fraction as a marker of mortality.
So again, as a surrogate for this pulmonary vascular arteriopathy.
And we see as the dead space fraction increases in even relatively small intervals, we see significant increase in mortality associated with it.
Work from our group here in Colorado similarly showed worsening of outcomes as for another potential marker of pulmonary vasculopathy.
That is when we looked at transpulmonary gradient from patients in the FACTS trial.
So the retrospective look at the FACTS trial, fluid and catheter treatment trial, a thousand patients who had PA catheters,
I assume a thousand patients who were enrolled in trial 500 with PA catheters measured their transpulmonary gradient.
And we see as that transpulmonary gradient increases, there are mortality increases.
So there was lots of sort of interest, certainly from us, from others, in thinking about pulmonary vascular bed as a potentially target for therapy, but a marker also of adverse outcomes in ARDS.
But as pointed out by Dr. Schmidt, is there something new and different going on here?
Because this, I think, becomes some of the crux of the discussion, if you will.
And he already went into some detail on Dr. Gattinone's work.
Here we see the impact of COVID-19 by these L and H phenotypes.
I have to say, my thoughts on this, we've discussed in other venues.
To me, the H phenotype is certainly something that we see, and this looks to me like more of a progression as opposed to two different disease states.
Though I find the L phenotype, again, to be quite fascinating, this concept that there is VQ mismatch, a move towards a shunt type physiology.
And again, now relying on sort of clinical observations of myself, but many others across the country, across the world, is that these patients present in an abnormal fashion with pretty profound hypoxia, but without sensing it.
When initially going on the event, their compliance does not seem particularly bad, or they seem relatively compliant, but then progress to a more typical ARDS fashion.
So what's happening early on? And so the question has been raised, is this something amiss with the pulmonary vascular bed?
And again, some potential evidence of that showed earlier here by Dr. Schmidt.
Other support of this, of course, is the spinotory ratio, another way of sort of getting at this question of VQ mismatch here, if you will, or shunt like physiology.
And so is this different and what could be the potential reasons for this? And this is where I think, again, things get quite interesting.
Early, certainly in the science of this, but here out of the Lancet published here this year, is evidence that COVID-19 can in fact, in impact endothelial cells.
It, of course, uses ACE2 as its entry point, which is prevalent on endothelial cells, most prevalent certainly in our epithelial cells from the lung, hence its entry point in the nose and oropharynx, but pleasant in other organs as well, endothelial cells being those as well.
And this is at least autopsy evidence that there can be active infection.
So these are samples from our renal biopsy electron microscopy showing viral inclusions in panel A, again, evidence of infection here at this point.
B is, in fact, evidence from a GI tract inflammation, and then C, and then D, we see the lung itself with small vessel showing mostly inflammatory cells accumulating, but then with caspase staining endothelial cell death in that region.
So, albeit observational, but potentially evidence here that something different is occurring from an infection standpoint, that these, this virus can get into the endothelial cells.
Here, perhaps, better evidence along these ways, really interesting work published by, excuse me, Montelli et al., which, in fact, in a, so this was a group that was actually interested in SARS-CoV-1, the initial SARS epidemic,
and had shown, again, the ability of the virus to infect organelles.
And so they developed endothelial cell organelles, so vessel mimics, if you will, applied this to SARS-CoV-2, demonstrated the ability of the organelle to be infected by virus.
So here we see platforming units, and then showing three days post-infection, six days post-infection, increase in viral RNA, sort of similar amounts from these different amounts of platforming units, and then were able to sort of harvest that virus and re-infect their more typical cell for evidence that this could be infected.
And then, really, the point of this was a study, it's a pretty interesting study, looking at recombinant ACE2, clinical grade recombinant ACE2, to essentially decrease the risk of infection, which is shown in D, applying this as a potentially therapeutic option.
But once again, getting at, I think, the point being this virus can, in fact, get into endothelial cells, and might that be why we're dealing with a different entity here in terms of VTE stroke, D-dimer levels that are rising precipitously, and even, again, I think, fascinating to this VQ mismatch problem here.
So in review of this, we seem to have, albeit in the form of retrospective looks, case series, clotting that seems over the top from what one would expect.
At least we can agree on the VTE aspect of it. I suspect we will also see this along the CVA, a potential devastating portion of it at this point.
So do we have a possible mechanism, albeit not proven, but do we have endothelial cell infection as our points here? And I think the question becoming, should we act on this, or do we wait and watch at this juncture?
So already, and this was mentioned, that many societies are being quiet, but a number are starting to imply that we should act. And I can tell you here, at least at the university, we are using D-dimer to adjust at least our prophylactic dose of anticoagulation.
Certainly we can all agree that prophylaxis is important. Should we, in fact, push that prophylactic dose further? And these are recommendations that we're looking at.
Do we have evidence that anticoagulating patients is beneficial? And so this is one of the first looks at this. This was, again, out of the Wuhan province. This was anticoagulant treatments associated with decreased mortality.
Again, not prospective. This is retrospective data of 449 patients. These were 99 patients who received anticoagulation for seven days based this on their SI SCI score, so their sepsis induced coagulopathy score.
Now, if we just looked at all comers, no difference in mortality between those that received it or not. But those with marked elevation, they're D-dimers. There was a statistically significant improvement in mortality if they received anticoagulation.
Take that data then for sort of what it's worth there.
So this is coming through. So this, more hot off the presses, this is published in Jack study out of Mount Sinai and their health system, a much larger population. This is the largest one I've seen to date of 2773 patients.
They look back now to see who received systemic anticoagulation in this group. So this is a first look at systemic anticoagulation versus prophylaxis. 786 patients received systemic anticoagulation.
Of those patients on mechanical ventilation, again, we see a marked mortality benefit here in this group. So mortalities of 29% in patients who received systemic anticoagulation versus 62% versus those that did not.
And here we see the Kaplan-Meier curves looking at those two differences. So in your sicker patient populations, I think if you make that assumption from patients who are on mechanical ventilation, then this is where our potential mortality benefit could come.
And again, the risk, you know, one would conceive potential rate of ETE, though this is data that we still need to parse out of this, though these were relatively carefully matched as could be done in a scenario like this.
Of course, you know, that data can be criticized, but here again, large patient population, significant mortality difference with anticoagulation.
So what we're really getting at, too, is what we always do in medicine, right, which is this, and let me mention on this last study, no difference in bleeding.