Uncertainties about the Roles of Anticoagulation and Microclots in Post-acute Sequelae of SARS-CoV-2 Infection, 2023, Connors and Ariëns

[SNT Gatchaman]

Uncertainties about the Roles of Anticoagulation and Microclots in Post-acute Sequelae of SARS-CoV-2 Infection
Jean M. Connors; Robert A.S. Ariëns

Three years after the start of the pandemic, approaches to the prevention and treatment of acute Covid-19 have been established. However, post-acute sequelae of SARS-CoV-2 infection (PASC) is now the focus of current investigations searching for the etiology of the symptoms and diagnostic strategies and treatments, as up to 10% of those with acute Covid-19 will go on to develop PASC, a significant public health burden.

Some have suggested that ongoing microvascular thrombosis and microclots may play a role in the persistent sequelae of Covid-19 infection. Treatments are being given to address the presumed role of thrombosis in PASC and some suggest that randomized controlled trials of anticoagulants in patients with PASC should be performed. In this Forum article, we focus on findings from patients with PASC that have led to the suggestion of using anticoagulants and discuss alternative considerations.

Link | PDF (Journal of Thrombosis and Haemostasis)

 

[SNT Gatchaman]

I can't access this currently, but presume it will be a strong rebuttal. I hope it will discuss mechanisms of hypercoagulability as a downstream effect.

 

[EndME]

I also don’t have access to it, but I always find it is a bit strange to write about a topic your colleagues next door have and currently are doing research on, without supposedly even reaching out to them or even being aware of their work.

 

[SNT Gatchaman]

I think hypercoagulability in LC (and ME) will be a part of the story, so thought it was worth spending some CME funds to access this journal for a year. (Hopefully our university will add it back into the portfolio, after it became MIA since the Jan move to Elsevier). Anyhow, I have this paper now and it says many things that we have discussed here. I think it's worth quoting some of the points.

Introduction

Suggestions for using anticoagulants to treat PASC are based on investigations of biomarkers of coagulation, endothelial cell activation or disruption, and the findings of microclots

These analyses are limited by small sample sizes, assessments performed at only one time-point, and wide variability in patient symptoms with variable timing following acute Covid-19. No individual patient baseline values before infection are available; patients without a history of Covid-19, or other convalescent patients, are frequently used for comparison.

 

[SNT Gatchaman]

Macrovascular Thrombosis

Macrovascular thrombosis does not occur at high rates in patients following acute Covid-19 with large database analyses finding a marked reduction in risk of thrombosis after the first 2-3 months.

Ref: Cardiovascular disease and mortality sequelae of COVID-19 in the UK Biobank (2023, Heart)

also dependent on severity of acute Covid-19, with those admitted to the ICU having greater risk of VTE post-hospitalization, suggesting alternative explanations.

Ref: Long-term cardiovascular outcomes of COVID-19 (2022, Nature Medicine)

Even those with VTE from acute Covid-19 hospitalization have no increased risk of recurrent thrombosis compared to hospitalized patients without Covid-19.

Ref: Venous thromboembolism secondary to hospitalization for COVID-19: patient management and long-term outcomes (2023, Research and Practice in Thrombosis and Haemostasis)

These findings suggest no role for VTE prophylaxis in patients with PASC.

 

[SNT Gatchaman]

Coagulation and endothelial abnormalities

In one study of 276 patients [...] 30% had an elevated VWF antigen/ADAMTS13 ratio, an indicator of endothelial cell damage. Of the 20% that had impaired exercise capacity, over half showed elevated VWF antigen/ADAMTS13. In all patients D-dimer levels were normal.

Ref: Impaired exercise capacity in post–COVID-19 syndrome: the role of VWF-ADAMTS13 axis (2022, Blood Advances)

Another study comparing coagulation biomarkers in 36 patients with acute Covid-19, 50 with PASC, and 20 matched controls, also found abnormal VWF antigen/ADAMTS13 ratios in the PASC patients. Although the ratios were lower than in the cohort with acute Covid-19, they were higher than control values. D-dimer, while elevated in the acute Covid-19 population, was in the normal range in those with PASC.

also correlated activated phenotypes of monocytes and T- and B-lymphocytes [...] findings suggest that ongoing endothelial cell activation is present in some with PASC, and is tied to changes in leukocyte phenotypes, but activation of coagulation and fibrinolysis is not, given normal D-dimer levels.

Ref: Sustained VWF-ADAMTS-13 axis imbalance and endotheliopathy in long COVID syndrome is related to immune dysfunction (2022, Journal of Thrombosis and Haemostasis)

elevated D-dimers have been found in some patients with PASC, levels are much lower than during acute infection.

Ref: Serum Metabolic Profile in Patients With Long-Covid PASC Syndrome: Clinical Implications (2021, Frontiers in Medicine)

a recent study found that plasmas from 46 patients with >1 symptom continuing >1 month postrecovery showed increased capability of triggering neutrophil extracellular traps (NETs) in healthy neutrophils, suggesting a persistent thromboinflammatory response.

Ref: NETosis induction reflects COVID-19 severity and long COVID: insights from a 2-center patient cohort study in Israel (2023, Journal of Thrombosis and Haemostasis)

A study of 21 patients with PASC [...] showed increased thrombogenicity of blood samples in microfluidic chamber assays, correlating with increased vWF levels compared to healthy controls.

Ref: Analysis of thrombogenicity under flow reveals new insights into the prothrombotic state of patients with post-COVID syndrome (2022, Journal of Thrombosis and Haemostasis)

While these findings suggest a vascular component to PASC, they are most likely due to ongoing vascular remodeling and neovascularization following microvascular thrombosis as result of acute Covid-19, or ongoing inflammatory effects on the endothelium, and not active coagulation.

Ref: Elevated vascular transformation blood biomarkers in Long-COVID indicate angiogenesis as a key pathophysiological mechanism (2022, Molecular Medicine)

 

[SNT Gatchaman]

Microclots

Most references are in S4ME under the microclot tag.

PASC is a complex disorder associated with a range of conditions such as chronic fatigue, cognitive disorder, and dizziness, depression and anxiety, some of which are unlikely caused by microclots.

Microclots have been found in patients with different diseases, and are not specific for PASC.

The term microclots requires some discussion. While microclots are microscopic in nature, their composition is variable and does not indicate a clear mechanism of formation. It is not clear whether microclots are indeed the result of blood clotting. Microclots are detected using an unconventional technique of staining plasma samples with amyloid dyes, which shows positively staining particles, which some authors then called microclots. The origin of these particles and role in causing disease requires further investigation.

 

[SNT Gatchaman]

Baker and colleagues identified fibrin and platelet aggregates in plasma of patients with pulmonary embolism that appear similar to the reported PASC microclots. Importantly, these samples were collected and analysed before the Covid-19 pandemic. Although initially thought to represent a potential artefact of sample preparation, the investigators recently published these findings with a critical discussion regarding the nature of microclots and mechanistic links with disease. It currently remains unknown if similar microclots occur in plasma samples of patients with other thrombotic disorders.

found that treatment with low-molecular weight heparin appeared to reduce the presence of microclots in patients with pulmonary embolism.

Ref: Plasma from patients with pulmonary embolism show aggregates that reduce after anticoagulation (2023, Nature Communications Medicine)

Senior author is this article's second author. It sounds as if that paper is worth posting separately.

ETA: Thread now here.

 

[SNT Gatchaman]

The next section basically re-states what @Jonathan Edwards has consistently written here.

The observation of microclots raise a number of key scientific questions. Reported diameters of microclots range from 8.8–62.0 µm, significantly larger than erythrocytes which measure 6-8 µm, yet these microclots do not appear to precipitate alongside blood cells during centrifugation when plasma is prepared. It is possible that microclots have lower density, however, an alternative explanation is that they form after the plasma is taken [...] double centrifugation of plasma did not reduce microclots.

Whether microclots are present in the circulation in-vivo or are the result of post-draw artefacts is unknown. The molecular or cellular mechanisms involved in the formation of microclots are also unknown.

A previous proteomics study reported the presence of many different proteins [...] microclots could be the product of activation of coagulation, platelets, or complement, or potentially all three [...] also been termed amyloid microclots, suggesting they could be the result of a different process such as amyloidosis. Deciphering the mechanisms by which microclots are formed is an important step towards future prevention or reduction of microclot formation.

Even if microclots are confirmed to occur in-vivo, the question remains whether they are cause or effect of disease. It is plausible that microclots represent fragments of existing thrombi undergoing remodeling.

 

[SNT Gatchaman]

They conclude —

In-vitro and in-vivo data demonstrating that anticoagulant treatments and plasma exchange reduce the presence or formation of microclots, and that doing so benefits patients with PASC, are needed from well controlled studies before inferences are made regarding possible treatments aimed at reducing microclots.

Based on the collective data discussed here, anticoagulation likely has little role in the treatment of PASC.

More plausible etiologies for PASC include ongoing inflammation from reaction of latent viruses such as Epstein Barr virus [...] or from persistent SARS-CoV-2 infection [...] Similarities between PASC and myalgic encephalomyelitis/chronic fatigue syndrome which can occur following acute infections have been made, suggesting similar etiologies.

 

[Michelle]

I know I've already mentioned my own experience with clotting and anticoagulants here on the forum, but just in case anybody new is reading this, here's my N=1 experience with ME/CFS and Enoxaparin sodium.

 

[EndME]

Iwasaki was also very recently asked about microclots since her team are investigating them as well .

I thought her opinion would be a decent reflection of their current research as she has a better tone than some of the “Twitter scientists”. The gist of it (or my interpretation of it) is that Akiko confirms that they find microclots and platelet hyperactivation in people after Covid (both Long-Covid and non-Long-Covid) as well as some HC. It's still ongoing work, they still need more controls which they are busy looking for. They are still cautious about this as it’s very early days, especially when it comes to a new theory for a new disease. It sounds like it will still take longer than this till the end of this year for first results to reappear. Naturally she also didn’t speak too highly of the current testing which is of course all not standardised or performed in some solid laboratory setting.

For me the takeaway still is though that at least they’ve being seeing something which she currently doesn't interpret to be an artefact of the methods.

 

[SNT Gatchaman]

Letter to the editor from Kell, Pretorius et al. Response from authors Connors and Ariëns. (Both are paywalled)

As there is an open study into microclots at another centre, it's probably helpful to record some of the letter's arguments.

In the letter, Kell et al discuss in two sections: incorrect citations; coagulation and endothelial dysfunction. In the first section, they conclude with —

We note too that these were all studies of single anticoagulants in acute COVID, which is not in fact what we have been studying (anti-platelet therapy is also required to deal with platelet hyperactivation) and whose demographics (especially regarding gender) are quite different from those of long COVID. Given the well-established thrombotic elements of acute COVID (also noted by the first author) elsewhere in the article, it is hard to know what the studies cited as references 2 to 13 of the study by Connors and Ariëns [1] are supposed to tell us about the microclot phenomena that also demonstrably occur in acute COVID. The authors fail to cite any of the many other and large-scale studies that nonetheless showed a clear benefit of preexisting or applied anticoagulation in acute COVID, such as [15–20] and others. Equally, we do note that occasional studies found no effect.

A few of the points made in the section on coagulation and endothelial dysfunction —

This lowered fibrinolysis explains entirely why raised D-dimer levels are rarely seen, and are thus irrelevant, in long COVID; only fibrinolysis generates D-dimer, and this process is inhibited. Note too that early variants of the spike protein are themselves unusually amyloidogenic, while the omicron version of spike is both far less amyloidogenic and far less virulent; this is an important “control” showing that microclots are on the disease pathway.

The authors note that “Microclots have been found in patients with different diseases, and are not specific for PASC.” We are unsure what the issue is here. As a parallel, inflammatory markers go up in all kinds of inflammatory diseases, as does ferritin, including in all those where we observe microclots.

“While microclots are microscopic in nature, their composition is variable and does not indicate a clear mechanism of formation. It is not clear whether microclots are indeed the result of blood clotting.” We have performed proteomics to elucidate the composition of microclots. We are also unsure why the variable composition of microclots should be problematic. The transition to a thermodynamically stabler amyloid form requires only a tiny amount of catalyst in the form of an amyloid conformer itself. This is well established with prions and prionoids.

“Microclots are detected using an unconventional technique of staining plasma samples with amyloid dyes.” The stains for amyloid structures are well established. In addition to thioflavin T (the classical stain), we have also used oligothiophene “Amytracker” (Ebba Biotech) dyes that have other desirable properties and are well established as stains for amyloid structures. We have also correlated the staining with the microclot structures seen in both electron microscopy and in flow cytometry.

“Microclot aggregates have also been termed amyloid microclots, suggesting they could be the result of a different process such as amyloidosis.” With 1 rare hereditary exception, the established amyloidoses do not include normal fibrin(ogen), which is what we discovered in fact becomes amyloid. The term “amyloid” simply refers to a kind of protein structure which we know to involve crossed-beta sheets, and this is what is stained by the amyloid stains.

The author response includes —

However, the discovery of microclots in several other diseases defies the specificity of microclots for long COVID and its complications. After all, patients with diabetes and pulmonary embolism, who also show microclots in plasma samples, do not experience fatigue or cognitive disorder such as brain fog associated with long COVID. Thus, it remains unclear how microclots may cause these symptoms. Microclots cannot be compared with inflammation, as suggested by Kell et al., since inflammation has clearly been demonstrated to play a mechanistic role in a variety of diseases including atherosclerosis, thrombosis, and other disorders in numerous clinical, in vitro, and in vivo studies, including animal studies, while such studies are lacking for microclots.

We are still surprised that microclots, which are larger than red blood cells and platelets, do not precipitate with these cells when platelet-poor plasma is prepared by centrifugation. This remains scientifically unexplained and raises questions on why such microclots are present in the plasma. One explanation is that they form after the blood is withdrawn. However, until we have evidence showing that this is not the case and that there is a good scientific rationale for their nonprecipitation during centrifugation, questions remain regarding where and when they are formed.

 

[rvallee]

We note too that these were all studies of single anticoagulants in acute COVID, which is not in fact what we have been studying (anti-platelet therapy is also required to deal with platelet hyperactivation) and whose demographics (especially regarding gender) are quite different from those of long COVID.

Reminds me of the people who trash recipes by commenting that they substituted 4 ingredients, halved one, doubled another, and the recipe is not good!

I thought MDs were better than this. It's been a while since I thought that, but I used to think that.

Then again when the standard approach to generic psychosocial rehabilitation doesn't care about substance, it seems natural that they would apply the same non-thinking to biomedical treatments, since the problem isn't with the treatments themselves but rather the patients, whom they see as not worth bothering to do anything.