Preprint Increased fibrinaloid microclot counts in platelet-poor plasma are associated with Long COVID, 2024, Dalton et al.

[RaviHVJ]

Abstract

Outcomes following SARS-CoV-2 infection are variable; whilst the majority of patients recover without serious complications, a subset of patients develop prolonged illness termed Long COVID or post-acute sequelae of SARS-CoV-2 infection (PASC). The pathophysiology underlying Long COVID remains unclear but appears to involve multiple mechanisms including persistent inflammation, coagulopathy, autoimmunity, and organ damage. Studies suggest that microclots, also known as fibrinaloids, play a role in Long COVID. In this context, we developed a method to quantify microclots and investigated the relationship between microclot counts and Long COVID. We show that as a cohort, platelet-poor plasma from Long COVID samples had a higher microclot count compared to control groups but retained a wide distribution of counts. Recent COVID-19 infections were also seen to be associated with microclot counts higher than the control groups and equivalent to the Long COVID cohort, with a subsequent time-dependent reduction of counts. Our findings suggest that microclots could be a potential biomarker of disease and/or a treatment target in some Long COVID patients.

https://www.medrxiv.org/content/10.1101/2024.04.04.24305318v1

(This is an attempted replication of the microclots finding by a UK team)

 

[Jonathan Edwards]

The paper seems well written.

They note problems with artefacts with prior methodology and cleaned that up. They also not problems with leaving samples on the bench.

The structures they show are extremely sparse and don't look a lot like some previous reports.

The data on numbers of the fluorescent blobs for Long Covid and controls does look plausible. There is too much overlap for the origin of he blobs to be a major factor in whether you have LC or not, as they note, but it does suggest some sort of trend in plasma protein behaviour.

The thing I would be most worried about is that this could be due to use of anti-inflammatory pain killers, which have major effects on coagulation-related proteins and their activation.

The us of thioflavine T puts the emphasis on amyloid and I wonder if there is a shift in amyloid A or P levels. It may be mentioned. I have not read through everything in detail.

If this turns out to be just a very roundabout way to show that production or activation of certain acute phase proteins, or perhaps proteins that are not classic acute phase reactants but may go up with viral infection, is shifted in LC then it would be very interesting. The values only give statistical patterns, so this might not be that good for confirming anything in an individual but it would be objective evidence of a prolonged immune response.

As I say, my only concern is that this might be secondary to common drugs.

 

[SNT Gatchaman]

Methodology improvements they noted were —

Existing assays have used a blood smear approach whereby PPP is mixed with ThT and incubated before analysing on glass slides. During the development of our method, we identified several processing issues that affected the observed microclot counts with this method. Firstly, we noticed that auto-fluorescent material was visible on untreated glass slides, comparable in size to previously reported microclots. We determined that this material was particulate matter from the laboratory environment or imperfections within the slide's glass. To counter any false positive results, we established that it was essential to handle all samples in a sterile environment and ensure that all laboratory equipment (tubes, pipette tips, slides) was clean and had not been left outside this environment at any point in the process.

We established that when whole blood tubes were left standing for more than an hour before centrifugation, the resulting PPP showed reduced or no microclot count. After observing this effect, to ensure the accuracy of our analysis, we only used PPP samples that were processed within 30 minutes of collection. Additionally, repeated freeze-thaw cycles of the PPP following processing also resulted in alterations in microclot counts. As such, PPP was stored in aliquots, and analysis was carried out on aliquots that had only been thawed once.

we used 15-well µ-Slides that come pre-sealed and were handled in a dust-free flow cabinet. We also used low protein binding plasticware at all stages. Each PPP sample was imaged directly in the well of the slide, allowing us to obtain 3D images within approximately 10 minutes per well. To minimise variability, we took triplicate technical repeats and analysis parameters were set to automatically identify the boundaries of the well to remove edge effects.

Finding that —

Control experiments and repeat data collection on the same slide gave consistent results up to six hours after sample handling, allowing the use of an autosampler for medium throughput. The researcher who performed the assay was blinded to the analysed PPP category, and each slide contained samples from both the Long COVID and control groups. Our assay provides an unbiased methodology for quantifying microclot counts within PPP samples.

 

[FMMM1]

RELEVANT?
I recall that Maureen Hanson highlighted that Andrew Grimson found platelets to be dysregulated** in ME/CFS potentially related to clotting?
So maybe there is something in micro clot findings.

**NIH Immunology webinar - https://www.ninds.nih.gov/news-events/events/mecfs-research-roadmap-webinar-series-immune-system

 

[EndME]

If this turns out to be just a very roundabout way to show that production or activation of certain acute phase proteins, or perhaps proteins that are not classic acute phase reactants but may go up with viral infection, is shifted in LC then it would be very interesting. The values only give statistical patterns, so this might not be that good for confirming anything in an individual but it would be objective evidence of a prolonged immune response.

As far as I can tell they just look at the LC cohort without looking at how long their illness duration has been or when their last known infection was (for the Covid cohorts they look at microclots w.r.t. last known acute infection but they don't present this data for LC). As such I don't see how it's possible to differentiate anything, especially as they don't mention the mean illness duration of the LC group (only that the minimum illness duration is 12 weeks). Is the LC group possibly experiencing the same clearance of microclots as the other groups?

 

[Jonathan Edwards]

Is the LC group possibly experiencing the same clearance of microclots as the other groups?

I am not sure what you mean by clearance of micro clots?

I think it swarth remembering that we have no evidence that these small structures exit in live people. They seem to occur in plasma incubated after centrifugation. The numbers of particles seem to be tiny. We see a few tiny dots. If you look at unseen blood it is more like looking at a brick wall. It is 50% solid particles roughly. I don't think these bits would do anything even if they were present in live people.

They still might be interesting.

 

[EndME]

I am not sure what you mean by clearance of micro clots?

I think it swarth remembering that we have no evidence that these small structures exit in live people. They seem to occur in plasma incubated after centrifugation. The numbers of particles seem to be tiny. We see a few tiny dots. If you look at unseen blood it is more like looking at a brick wall. It is 50% solid particles roughly. I don't think these bits would do anything even if they were present in live people.

They still might be interesting.

The suggestion is that these microclots could be a marker of something happening during an acute Covid infection and that this process, whatever it may be, doesn't resolve after the acute phase in a subset of LC patients (the authors state "microclots are cleared over time"). However, as far as I can see the authors have failed to look at whether this process really resolved less in LC patients than in the general population as they didn't look at time since last known infection in this group. It might just be that nothing is happening apart from the fact that the LC group had more recent infections.

 

[Jonathan Edwards]

The suggestion is that these microclots could be a marker of something happening during an acute Covid infection and that this process, whatever it may be, doesn't resolve after the acute phase in a subset of LC patients (the authors state "microclots are cleared over time").

But marker of what?

Actual clots in circulation would be expected to clear within about 1 minute max I think.
If they are too small to embolise they would probably be cleared by spleen at least in half an hour.
So I am unclear what this is supposed to be about.

 

[ME/CFS Skeptic]

Also from the paper:

"Using this symptom scoring system we did not see clear relationships between microclot counts and the presence of symptoms. Around half of the participants had microclot counts similar to the controls but reported the same symptom patterns as those with raised microclots, indicating that the presence of elevated microclot counts does not always appear to be a pre-requisite of Long COVID."​

 

[ME/CFS Skeptic]

These seem to be the main results:

 

[chillier]

As far as I can tell they just look at the LC cohort without looking at how long their illness duration has been or when their last known infection was (for the Covid cohorts they look at microclots w.r.t. last known acute infection but they don't present this data for LC). As such I don't see how it's possible to differentiate anything, especially as they don't mention the mean illness duration of the LC group (only that the minimum illness duration is 12 weeks). Is the LC group possibly experiencing the same clearance of microclots as the other groups?

This is a good point. They have this plot showing the relationship of microclot counts with time since infection in the people who had covid but don't have long covid:



Ideally we would also see the same plot for people with long covid.

The potential problem is that if patients with LC had covid more recently than the the covid (but not long covid) group, then part of the difference they described could just be due to microclot clearance with time.

The authors specify the long covid group had experienced LC symptoms for at least 12 weeks, so if for arguments sake we make a conservative assumption that every LC patient had symptoms for exactly (and no more than) 12 weeks (84 days), then we'd need to compare the LC group values (as shown in the plot in @ME/CFS Skeptic 's post) to the average from that plot above at 84 days, which is probably around/just below 40. It's not totally clear but I think the LC group would probably still be higher than that.

Overall I like this paper, I think the fact that they see a relationship with time after infection in their control cohort suggests that these do reflect an actual biological thing - not just some noise - even if it's unclear whether the clots themselves are present in vivo. The suggestion that they indirectly reflect some acute phase proteins seems like an interesting idea.