Aphaeresis/ Apheresis (for removal of microclots)

(I know I'm going to regret trying to answer Jo's questions! Realistically they're unanswerable at this time - which is, I guess, his point. Apologies for the rehashing, I think we've taken this subject about as far as we can until there are other publications that support or deafening silence refutes.)

I'm enthusiastic about the idea of amyloid-form clotting, with abnormal fibrinogen, small fibrin monomers and larger fibrin aggregates, being resistant to fibrinolysis. I'm not enthusiastic about apheresis, which does not appear to have usefully held up over time. I was more interested in it at the beginning, although thought it would be difficult to subject to an RCT. Many commentators here warned back then that it would not pass scrutiny.

As Jo says, no-one has yet reported these in living people. The closest related report I came across was descriptions of microvascular thrombosis appearing during clot-retrieval interventions. As I recall this was said to be a new phenomenon, not previously noted by experienced operators. I can't recall whether that was a Twitter post or elsewhere. Of course that only describes de novo observed microvascular clotting, it doesn't say they were due to microclots in circulation or that they were amyloid in nature.

Microclots have been described in postmortem studies, but needless to say, those patients are dead, so quite different and of limited study value. The other related data point I guess is the multiple reports from embalmers that postmortem clots, with which they are very experienced, had become very abnormal in extent and consistency. This was said to be a new finding, shown in people who had died of COVID-19 (or, more contentiously, following recent vaccination).

None of that is particularly helpful for the questions though.

How could apheresis help? (Or: how might it have helped?). The hypothesis is that the microclots they were seeing in vitro, were present in vivo. In their papers they have stated that the microclots they show must have already been in circulation, because no thrombin was subsequently added. The logic I presume being that thrombin is required to turn soluble fibrinogen into insoluble fibrin. I don't know that that follows exclusively. As far as I understand fibrin exists on a continuum that as a small fibrin monomer (that has not yet cross-linked) can still be soluble. I imagine there is opportunity for soluble or near-soluble fibrin to form larger aggregates in vitro, such that microclots could simply be an artefact (as Jo has always maintained).

Assuming they were circulating in vivo though, the hypothesis was that they were abnormally formed: not tightly compacted as clots should be, but instead nebulous and deformable enough to pass through small capillaries even if they were themselves quite large. (c.f. RBCs turning into tacos in small capillaries). In this way they would continue to circulate, not be filtered in lung or kidney and persist. The in vitro microclot analysis showed they contained a2-antiplasmin which would tip the balance away from fibrinolysis and help explain their longevity.

As Jo says, no-one appears to have replicated this in a publication. What has been shown recently is the mechanism for how spike protein causes amyloid fibrin(-ogen), again in vitro (also in silico). See this thread detailing that basic science work by an unrelated Swedish group. That demonstration goes lower level, whereas we need to go "up" and show circulating amyloid microclots. The Swedish group did demonstrate two PET tracers that tag the amyloid fibrin. They suggest this could be used in patient investigations.

Presumably if a patient had lots of amyloid fibrin (small) in blood, then the blood pool would be generally "hot" on PET. I'm surmising that if there were larger circulating microclots, then, notwithstanding the deformability mentioned above, they might be expected to pass much more slowly through some organs with obligate capillary networks, compared to organs with opportunity for arteriovenous anastomotic bypass, which could have the effect of appearing to accumulate tracer preferentially in eg lungs or kidney, despite not ultimately being filtered out there, as above.

 

Yes, agree (including your point about liver, spleen etc). I'm pretty keen on the possibility of abnormal (small) fibrin, but would like replication. I don't see proof of circulating microclots, but think it remains a possibility worth exploring.

Isn't that what's being shown? Retinal changes. Covid toes/nails. Higher thromboembolic risk. Often young, previously healthy/fit people down the line from Covid having the later thromboembolic events of DVT, PE, MI and stroke.

• Retinal Microcirculation as a Correlate of a Systemic Capillary Impairment After [SARS-CoV-2] Infection (2021)
• Retinal vessels modifications in acute and post-COVID-19 (Nature, 2021)
• A review of nail findings associated with COVID-19 infection (2021)
• Long-term cardiovascular outcomes of COVID-19 (Nature, Feb 22)
• EDIT: See also today's CDC figures to Nov 2021

Also common to see individual reports on Twitter.

https://twitter.com/user/status/1489221020837888005

https://twitter.com/user/status/1502994935326388226


Additionally, V/Q scans appear to be coming back into vogue to assess for subsegmental PE, which is below the resolution of CT pulmonary angiography.
https://twitter.com/user/status/1528016436135964672

 

Yes, the theory is a combination of platelet hyperactivation, endotheliitis and amyloid long-lasting fibrin(-ogen) are interacting with each other to perpetuate both all of these pathological processes and also downstream pathology.

I don't, but endotheliitis / endothelial dysfunction would likely be associated with degradation of BBB integrity.

Standard and even advanced neuroimaging by MRI has been limited. The Xenon MRI lung studies were definitely an advance. I think a promising line is specific PET tracers. These might be able to assess for amyloid fibrin in the same way that Aβ is assessed in Alzheimer's. (I appreciate the amyloid in AD is probably a secondary phenomenon.)

I don't know, but there's always the problem that postmortem studies have limitations. By definition, the patient died so has had a more severe disease pathway than living patients. But I guess more importantly blood likes to clot once it's out of circulation (or not circulating). So would be prone to artefact.

I'm not keen on HELP apheresis, until it can be shown to be effective, following good replicating evidence that this is the mechanism at play. But possibly a use case would be a one-off for severe symptoms, and then proceed with medical therapy. Also, that medical therapy may not be anticoagulation, but MAbs or some other more tightly targeted approach. Unclear if the coagulation pathology is a specific, additional feature of this virus which is unrelated or obliquely related to more typical postviral ME pathogenesis.

Paucity of publications so far. Anecdotally, people maintain they have not had adverse bleeding events, but the researchers are at pains to demand this is only attempted under proper medical supervision, with demonstrated hypercoagulability (and the doctor has to agree this is reasonable). Asad Khan maintains that 3x anticoagulation is much safer than you might think, because it brings the patient down from hypercoagulable to normal, rather than making a normal patient hypocoagulable and prone to adverse bleeding events. For what it's worth, I haven't seen any Twitter reports of bleeding.

Those advocating claim it is very safe with a 40 year track record. Unclear if 1 or 2 treatments have the same risk profile as 8+ for this LC treatment. I would have concerns that it's removing lots of good stuff on the way, and this could be harmful. I would prefer that we can understood the potential mechanisms and could then target with RCT-able drugs.

We have good quality PET scanners, but the trick is to be able to manufacture the novel tracers. They may be challenging and expensive to make and often only available to research institutes. Also, if the half-life is short it needs to be made close to the scanner, just before the patient appointment. In NZ standard 18FDG is made in Wellington and flown up or down, which is fine given its 2hr half-life. We can even fly it in from Oz at times.

 

Thank you, I very much appreciate your review. Yes absolutely agree with your point that PE etc implies large thrombo-emboli and wouldn't be explained by "microclots". Given these ongoing increased T-E reports, to me that supports the idea that there's persisting hypercoagulability in many people. It's something I've subjectively observed in myself (frequently clotting tubing to the surprise and lack-of-delight of phlebotomists).

I had thought that Covid toes were a more prolonged phenomenon (per patients' postings on Twitter) but interesting to see that reported evidence indicates a shorter timeframe in reality.

Does seem a shame that no-one has read the Pretorius publications. While this current work may ultimately not amount to anything, I still think the underlying ideas warrant a second look by other groups. Appreciate this was likely a Medical Grand Round, but I imagine her co-author, Prof. Doug Kell would be known to one or two researchers at UCL, though perhaps they wouldn't be audience members.

He has commented on Twitter that previously there was a clear distinction between their in vitro micro-clot finding being present in 100% of LC and not (or minimally) in healthy controls. Now they observe that it's no longer quite so minimal in HCs and that even commercial plasma shows more activity than previously. He ascribed this to the fact that many more donors have been exposed to spike protein (virus and/or vaccine). I wonder if this could be regarded as representative of a dose-response relationship.

https://twitter.com/user/status/1523802460258816002

 

Just to follow on this part of the discussion as it relates to acute COVID disease, there is a new publication: Sex Differences in Clinical Characteristics, Management Strategies, and Outcomes of STEMI With COVID-19: NACMI Registry, looking at MI in acute C19 patients.

It discusses the findings in ST segment elevation myocardial infarction (STEMI) in those hospitalised for this, who are also recently positive for SARS-CoV-2 (out to four weeks). They also include those having MIs while in hospital for any other reason, who are similarly recently C19 +ve. It notes a 30% in-hospital mortality rate and found a relatively high rate of no identifiable coronary artery lesion, moreso in women. 585 patients.

In women, chest pain presentations without angiographically demonstrated obstructive coronary artery lesions have been challenging. Women have often been treated as if the chest pain is non-cardiac. This is despite heart disease being the leading cause of death in women across all ages. Discussed more in Gender and Microvascular Angina (2011).

 

This seems to me to be an important and well balanced article. I think one of the important things it highlights is the different approaches to evidence and risk among clinicians - those providing the treatment, those urging caution until a clinical trial has been done, and those who are both clinicians and patients and want to get on with trying anything that might work, and arguing that clinicians shouldn't wait for clinical trials.