Muscle abnormalities worsen after post-exertional malaise in long COVID, 2023/4, Wüst, van Vugt, Appelman et al

[Jonathan Edwards]

Good luck to Carson if his last hope for him not to be absolutely wrong is DOMS… I would love the sweet feeling of DOMS instead of PEM, not least being able to exercise in a way that would provoke DOMS.

Indeed, but I worry that the interpretation of this recent paper supports that view - it suggests that PEM is all about inflamed muscle.

My limited experience of Long-Covid/PASC in the sense of being kiboshed for a couple of months or more each time has nothing to do with local muscle symptoms suggesting inflammation. It is more like feeling the life has been sucked out of one by a Dementor. There seems to be weakness in the sense that going upstairs is like climbing a mountain but no hint of muscle pain or soreness.

 

[Sid]

Wüst: “Also, it’s important to remember that people undergoing bed rest get better when they start exercising again, but this is not the case in long COVID patients. They get worse…”

Carson: “Look forward to reading it. As you note delayed onset muscle soreness has been recognised for at least 120 years and must ov course have a mechanism - i think thats where the issues of inference become key- but of course maybe your other data speaks to that

What nonsense. He's going to continue telling his patients their PEM is actually DOMS and they should just push through with "rehabilitation".

 

[Simon M]

Interesting comment from Alan Carson, President of the FND Society, asking for a bedrest comparison.

He should read the analysis here. someone here (@ME/CFS Skeptic ?) has already pointed out the long Covid patients averaged 4000 steps a day: bedrest is not a relevant comparison. Plus, as Rob Wust, Jonathan Edwards and others here have pointed out, PEM is not the same as delayed onset muscle soreness.

Maybe I’m overgeneralising, but psychosocial enthusiasts always seem to respond with generic, knee-jerk criticisms. I don’t understand why they don’t just read the biomedical research that makes them uncomfortable: they might find real reasons to criticise the methodology (or might not). It’s almost as if they’re not able to operate outside their own bunker, or just don’t feel the need to engage properly with other scientists – never mind patients.

I’m a big fan of scientific debate and criticism, but it's only useful if it's thorough and thoughtful, not regurgitating vague criticisms in the hope that one will stick.

Anyway, I suspect tthat the RW bedrest study will settle Alan Carson's concerns on the bedrest point.

 

[FMMM1]

For that it is crucially important that someone actually tries to replicate already existing research, which requires the research to be methodologically strong since the replication crisis is to a very large extent a methodological crisis. It's equally important that if findings can't be replicated that negative results are published as well, since negative findings are just as valuable as positive findings. Unfortunately, some research groups seem to be just jumping from hypothesis to hypothesis, without ever providing much evidence of the hypothesis being true or why they abandoned it.

A couple of weeks ago there was comment that alluded to the fact that muscle cells appear to look normal in patients with ME/CFS. Not knowing nearly enough about past research I was of the impression that Rob Wüst was one of the first people to have looked at this. @Andy was quickly able to change my mind as he posted an abundance of studies that had looked at similar things in the past. Unfortunately, after having a looked at these studies I'm none the wiser. They all varied extremely heavily in the methodology as well as in the recruitment criteria they used (unfortunately very often Fukuda).

In my eyes the work by Wüst et al carries many properties, most prominently w.r.t. to cohort selection in LC patients, that are needed for others to see if they can replicate these findings and if not what phenomena could explain these anomalies. Apart from Wüst conducting the same research in ME/CFS, I hope someone picks this up in a larger cohort (including bed-bound patients as well as bed-bound healthy controls) and possibly even examines connections to the work of other groups (for example the muscle findings by Hwang or PEM w.r.t. cognitive exertion).

Thanks for your reply; unfortunately mine won't be of the same quality!
I recall the Group at Griffith's University publishing that they'd found certain genetic differences in a receptor. Wengzhong Xiao [OMF Harvard] review the data, in one of the Stanford Symposiums, and said that it didn't add up.
I'm impressed by the group at Griffith's University and, to me, they could do some good/great research. So GWAS data might just save a whole lot of --- around. I think Jonathan may have insightfully pointed out recently --- people in labs are desperate to find something i.e. after all that hard work. GWAS might just provide a reality check --- abandon this -- and thereby an opportunity to the re-focus on more promising research areas.

 

[dave30th]

Maybe I’m overgeneralising, but psychosocial enthusiasts always seem to respond with generic, knee-jerk criticisms. I don’t understand why they don’t just read the biomedical research that makes them uncomfortable: they might find real reasons to criticise the methodology (or might not). It’s almost as if they’re not able to operate outside their own bunker, or just don’t feel the need to engage properly with other scientists – never mind patients.

I don't think you're over-generalizing. They are incapable of accommodating any new information that challenges their biases. Period.

 

[Andy]

Maybe I’m overgeneralising, but psychosocial enthusiasts always seem to respond with generic, knee-jerk criticisms. I don’t understand why they don’t just read the biomedical research that makes them uncomfortable: they might find real reasons to criticise the methodology (or might not). It’s almost as if they’re not able to operate outside their own bunker, or just don’t feel the need to engage properly with other scientists – never mind patients.

They rely on poor research to support their own theories, which logically must mean that either they can't do better, so would be unable to critique other research, or they are aware of the poor nature of their own research and don't want to point out flaws in research that most likely already exist in their own.

 

[Kitty]

They are incapable of accommodating any new information that challenges their biases.

I wonder if it's even worse than that, with some of them not actually invested in their theories at all. They're invested in their ability to hoover up grant funding by telling policymakers what they want to hear.

 

[Mij]

This is receiving a lot of traction on X which is fantastic, but it's also getting mixed up in the messaging. I've been reading from researchers who are responding to the study that 'injury' is caused from 'intense exercise' when it can be brought on by walking.

 

[Mij]

The small cumulative factors must also be clearly stated.

 

[Amw66]

Todd Davenport commenting re Carson's tweet too

Twitter/X exchange between Alan Carson and Wüst:


Carson: “Interesting and longitudinal design helpful - but small so needs replication but also needs a bed rest or similar comparison - are we seeing the cause or the effect?”

Wüst: “Dear Alan, please bear with me for 2 weeks, when we publish our results of a bed rest study in Cell Reports Medicine. I am also excited to see this work replicated elsewhere, but this work is in line with previous studies on similar topics (even one dating back from 1980s)”

Wüst: “Also, it’s important to remember that people undergoing bed rest get better when they start exercising again, but this is not the case in long COVID patients. They get worse…”

Carson: “Look forward to reading it. As you note delayed onset muscle soreness has been recognised for at least 120 years and must ov course have a mechanism - i think thats where the issues of inference become key- but of course maybe your other data speaks to that

 

[EndME]

Todd Davenport commenting re Carson's tweet too

In my eyes the problem is less that "this isn't what you see in healthy or deconditioned people", but rather that an inactive control group (and also a bed-bound group of PEM patients) provides a further level of control and helps you further analyse the existing data (that doesn't mean the group by Wüst didn't do excellently well with the extremely limited means given in LC research funding and others can now work on replication). Having a study of "LC with PEM vs. healthy controls vs. inactive controls" can in that sense be more robust than one study of "LC with PEM vs. healthy controls" and one study of "inactive controls" due to small differences occuring in the lab that could introduce some noise, because nobody sensible could anyways think that PEM is related to deconditioning.

I've also seen some responses to Carson's ridiculous tweets (not sure if it makes sense to respond to someone about metholodogy when said person clearly doesn't care about methodology to begin with and is only driven by his own beliefs rather than any evidence) which then responded with claims that this study proves causality. Unfortunately, even with replication we'd still not be there yet since the study only shows a strong correlation between PEM and these muscular abnormalities, which however is different from causality. As an example it may for instance be that autoimmunity causes both PEM as well as exercise induced muscular abnormalities via different mechanisms and that the findings are linked via autoimmunity rather than the one causing the other or vice versa.

 

[SNT Gatchaman]

Neither did we observe an increase in muscle breakdown products, such as creatinine and creatine kinase, in the plasma of both groups.

In the supplementary data —

At baseline (day 1), venous blood in the fasted state, and a muscle biopsy of the vastus lateralis muscle was taken, alongside questionnaires (see text) about long COVID symptomatology. Six days later (day 7), a cardiopulmonary exercise test (CPET) was performed to evaluate exercise tolerance, and to induce post-exertional malaise in patients with long COVID. One day later, fasted venous blood and another skeletal muscle biopsy from the vastus lateralis muscle was taken, and questionnaires confirmed the worsening of long COVID symptoms, typical for post-exertional malaise. On day 15, 8 days after the CPET, another fasted venous blood sample was drawn, and questionnaires were filled in. During the entire study period, participants wore an Actigraph to register the total steps taken.

So the blood samples relevant to CK changes from baseline were obtained at day 1 and day 8, relative to CPET.

From Eccentrically Induced Skeletal Muscle Damage in Patients With Chronic Fatigue Syndrome CFS, With Reference to Overtrained Athletes (1995) it can be seen that in that study CK peaked at day 4 but is very close to baseline at day 1 and close to baseline at day 8. (The testing regimen was days 1, 2, 4, 6, 8, 12, 16, 20, 24, 28.)



Looking at healthy / athletes Creatine kinase monitoring in sport medicine (2007, British Medical Bulletin) —

The time of CK release into and clearance from plasma depends on the level of training, type, intensity and duration of exercise. Peak serum CK levels of about 2-fold above baseline occur 8 h after strength training. Increased CK levels after eccentric exercise are associated with muscle injury, with a pronounced increase between 2 and 7 days after exercise.

After prolonged exercise, total serum CK activity is markedly elevated for 24 h after the exercise bout when subjects rest and remains elevated for 48 h when subjects train in the first week post-exercise. The release of CK following eccentric exercise peaked 96 h after the exercise bout, and an additional bout of exercise produces only small increases, probably from accelerated enzymatic clearance. More intense activity, such as a twice daily football training, leads to significant increase of CK during the fourth day of training. CK levels decrease between days 4 and 10, probably an adaptation to training. A bout of exercise performed 48 h after an initial bout does not change the time course of the CK leakage.

The decrease in the serum enzyme levels depends on the period of rest after exercise, as short-term physical inactivity may reduce both the lymphatic transport of CK and the release of the enzyme from the muscle fibres. Manual lymph drainage after treadmill exercise is associated with faster decrease in the serum levels of muscle enzymes. Another factor that may reduce muscle damage and serum concentrations of CK following prolonged exercise is supplementation with branched-chain amino acids, often used in sports.

See also Creatine-Kinase- and Exercise-Related Muscle Damage Implications for Muscle Performance and Recovery (2012, Journal of Nutrition and Metabolism) which also discusses sex differences.

 

[Sid]

The responses to Carson's uninformed tweet are just great. He's really getting roasted. It's clear he didn't look at the study and tweeted that to try and dismiss it out of hand. These patients were not on bedrest. I'm sitting here with a similar step count as the people in this study and I'm profoundly sick compared to controls who walk only 1000-2000 steps more. Absurd life-destroying symptoms after exertion like sore throat, pain everywhere, flu-like malaise, heart issues, insomnia...

 

[Sid]

https://twitter.com/user/status/1743241329738395749


https://twitter.com/user/status/1743286301053358339

 

[dave30th]

I wonder if it's even worse than that, with some of them not actually invested in their theories at all.

That is of course always possible or maybe likely in some guess, but I guess I prefer in general to think that people are deluding themselves rather than that they're being consciously venal and deceptive. that could be a naive perspective, however.

 

[LarsSG]

If amyloid deposits were a problem in Long Covid then one would expect to see them before exercise and be very lucky to see any visible increase after.

They did see the amyloid deposits (or whatever they might be staining with Thioflavin T) in LC before exercise. Not really a huge change after either (excepting those three outliers with very high densities).



I suppose the baseline high amyloid deposits density in LC would be pretty easy to try to replicate for anyone who has frozen LC muscle biopsies, which I think we've seen a few studies of. Replication could involve a different stain (seems like Congo Red is the older standard). I wonder if that's something that the Patient-Led Research Collaborative would fund. Is anyone connected with them?

 

[dave30th]

Can someone explain what he means by: "...must of course have a mechanism - i think thats where the issues of inference become key- but of course maybe your other data speaks to that" What inference is he talking about here? Can someone clarify? Is he suggesting these findings are similar to what happens in DOMS?

 

[LarsSG]

This paper seems like it might be relevant to the amyloid in muscle question: TDP-43 and RNA form amyloid-like myo-granules in regenerating muscle

Abstract: A dominant histopathological feature in neuromuscular diseases, including amyotrophic lateral sclerosis and inclusion body myopathy, is cytoplasmic aggregation of the RNA-binding protein TDP-43. Although rare mutations in TARDBP—the gene that encodes TDP-43—that lead to protein misfolding often cause protein aggregation, most patients do not have any mutations in TARDBP. Therefore, aggregates of wild-type TDP-43 arise in most patients by an unknown mechanism. Here we show that TDP-43 is an essential protein for normal skeletal muscle formation that unexpectedly forms cytoplasmic, amyloid-like oligomeric assemblies, which we call myo-granules, during regeneration of skeletal muscle in mice and humans. Myo-granules bind to mRNAs that encode sarcomeric proteins and are cleared as myofibres mature. Although myo-granules occur during normal skeletal-muscle regeneration, myo-granules can seed TDP-43 amyloid fibrils in vitro and are increased in a mouse model of inclusion body myopathy. Therefore, increased assembly or decreased clearance of functionally normal myo-granules could be the source of cytoplasmic TDP-43 aggregates that commonly occur in neuromuscular disease.

 

[EndME]

They did see the amyloid deposits (or whatever they might be staining with Thioflavin T) in LC before exercise. Not really a huge change after either (excepting those three outliers with very high densities).

Indeed. At least this data seems like something that could also be explained by variations occuring in the lab (taking a biopsy from a different location, some slight differences in applying the staining etc) rather than necessarily a faulty exercise response. On the other hand it might also be that exactly those 3 outliners are exactly those people you want to be studying in such a heterogenous condition. Typically I'm far more impressed by large difference in a subset of people, than small differences in everyone often established by some of the AI/ML classifiers.

I suppose the baseline high amyloid deposits density in LC would be pretty easy to try to replicate for anyone who has frozen LC muscle biopsies, which I think we've seen a few studies of. Replication could involve a different stain (seems like Congo Red is the older standard). I wonder if that's something that the Patient-Led Research Collaborative would fund. Is anyone connected with them?

Indeed, seems like an excellent idea. The main problem will probably be that most of those studies wouldn't have assessed for PEM and given the heterogeneity of LC and the many patients with a short duration of self-resolvent symptoms, very mild symptoms or only a cough, one might need very large sample sizes to detect meaningful differences (and controlling for a recent Covid infection seems to be a larger problem in the microclot studies). Technically some of the PolyBio studies could address this (this Polybio study is using some kind of in-vivo imaging technique to look at something similar https://polybio.org/projects/evalua...fibrin-pet-imaging-peripheral-blood-analysis/, whilst this study is collecting biopsies and then staining them https://polybio.org/projects/immune-activity-and-microclotting-in-long-covid-peripheral-neuropathy/).

 

[Andy]

Can someone explain what he means by: "...must of course have a mechanism - i think thats where the issues of inference become key- but of course maybe your other data speaks to that" What inference is he talking about here? Can someone clarify? Is he suggesting these findings are similar to what happens in DOMS?

I would guess he is trying to lay the ground for the "of course there is something going on in the body but it's how the brain interprets it that is the problem" line of psychobabble hardware/software, 'definitely not cartesian dualism' waffle.