Long COVID and chronic fatigue syndrome/myalgic encephalitis share similar pathophysiologic mechanisms of exercise limitation, 2025, Jothi et al

forestglip

forestglip

Moderator
Staff member
Long COVID and chronic fatigue syndrome/myalgic encephalitis share similar pathophysiologic mechanisms of exercise limitation

Swathi Jothi, Michael Insel, Guido Claessen, Saad Kubba, Erin J. Howden, Sergio Ruiz-Carmona, Todd Levine, Franz P. Rischard

[Line breaks added]

Abstract
Post-acute sequelae of SARS-CoV-2 (PASC or “long COVID”) and chronic fatigue syndrome/myalgic encephalitis (CFS/ME) share symptoms such as exertional dyspnea.

We used exercise oxygen pathway analysis, comprising six parameters of oxygen transport and utilization, to identify limiting mechanisms in both conditions. Invasive cardiopulmonary exercise testing was performed on 15 PASC patients, 11 CFS/ME patients, and 11 controls.

We evaluated the contributions of alveolar ventilation (V̇a), lung diffusion capacity (DL ), cardiac output (Q̇), skeletal muscle diffusion capacity (DM ), hemoglobin (Hb), and mitochondrial oxidative phosphorylation (Vmax) to peak oxygen consumption (V̇O2peak). To simulate targeted interventions, each variable was sequentially normalized to assess its impact on V̇O2peak.

V̇O2peak was significantly reduced in both PASC and CFS/ME compared to controls. Skeletal muscle O2 diffusion (DM ) was the most impaired parameter in both patient groups (p = 0.01).

Correcting DM alone improved V̇O2 by 66% in PASC (p = 0.008) and 34.7% in CFS/ME (p = 0.06), suggesting a dominant role for peripheral O2 extraction in exercise limitation.

Impaired skeletal muscle oxygen diffusion (DM ) is a shared mechanism of exercise intolerance in PASC and CFS/ME and may represent a therapeutic target. However, our findings are limited by small sample size.

Web | PDF | Physiological Reports | Open Access
 
They seem to say that they found something similar to the preload failure of Systrom's group:
When examining standard iCPET measurements outside of the O2 pathway, CFS/ME and PASC demonstrate reduced cardiac index relative to controls. This is in the absence of elevated cardiac filling pressures or pulmonary artery pressure indicating heart failure is an unlikely cause. In keeping with the lack of a rise in right atrial pressure with exercise, this may be considered “preload insufficiency” (Joseph et al., 2023)
Click to expand...

The novel aspect of the paper seems to be that they used O2 pathway analysis, which tries to break down oxygen transport into different aspects. In ME/CFS and Long Covid patients, the problem seems to lie in the Dm parameter, which stands for muscle diffusion capacity.

1756914673069.png
 
The presentation of some results doesn't seem very convincing to me despite being a layman. For example they took skin biopsies of 6 people and then later present the image of the biopsy for one person and claim for 4 other people look sorta the same. Why is it not standard to just present all the data you've got (at least in the supplementary materials)? If I recall the diagnosis of SFN is pretty much biased entirely by sample extraction, handling and how you're counting fibers and more standardised efforts have failed to show anything consistent.

It also appears to be the case that the experiments for controls might have been performed quite differently as some experiments were performed in Arizona whilst others in Belgium but it isn't clear whether the results might have differed and whether that introduces problems. I don't know much about iCPET but I know that professional athletes undergoing CPET for training optimisation try to always do so at the same facilitates because results tend to be different in different facilities.

It's unfortunate that they didn't try to replicate some of the results by Wüst et al given similar setups.
 
EndME said:
If I recall the diagnosis of SFN is pretty much biased entirely by sample extraction, handling and how you're counting fibers and more standardised efforts have failed to show anything consistent.
Click to expand...
Can you remember any details of that, and does it refer to SFN testing in general? I had always wondered how reliable the findings were, but hadn't seen anything looking at the methodology?
 
Simon M said:
Can you remember any details of that, and does it refer to SFN testing in general? I had always wondered how reliable the findings were, but hadn't seen anything looking at the methodology?
Click to expand...
I haven't seen anything about SFN in general (or for diabetes or something else) but only in the context of ME/CFS, LC and FM. I think there's generally 2 tests that are used for SFN: A certain type of microscopy of the eye where a combination of fibre measures is taken and this nerve counting exercise following tissue extraction where different protocols are often used. I recall reading studies suggesting outcomes in one experiment don't predict anything for the next (the same applies to the Goebel study mentioned below) but I don't know where I've read it.

There's a somewhat sizeable study with controls by Goebel on SFN in FM, who is a big advocate for SFN pathology in FM, and in that study it's reported that there are no statistically significant differences between controls and FM patients and mean IENFD (intraepidermal nerve fibre density) values for the 3 different regions are presented for both groups but none of results are significant, despite there being a mismatch in BMI matching (and higher BMI has been shown to be associated with lower IENFD in some studies). The corneal results are also not statistically significant. All of that is even without multiplicity correction. Funnily enough the authors still write "In keeping with previous studies, approximately half (48.4%) of the individuals with FMS had IENFD below the normative range for age and sex in at least 1 location" but since they appear no different to controls, to me that can only mean that previous studies have also showed nothing significant.

Whilst I'm generally sceptical of AI this should probably be an area where it should be able to at least help with image classification quite easily.
 
EndME said:
It's unfortunate that they didn't try to replicate some of the results by Wüst et al given similar setups.
Click to expand...
@EndME Do you mind if I ask how this group would do that?

Asking because I've been messaging with the PI of this study (Franz P. Rischard), who is a RECOVER PI with Univ. of Arizona - we've been discussing next steps from this study and he's seeking advice from the patient community.

If any here have any advice or best recommendations to make, feel free to let me know or share here/DM, and I'll try to pass along
 
Dakota15 said:
Asking because I've been messaging with the PI of this study (Franz P. Rischard), who is a RECOVER PI with Univ. of Arizona - we've been discussing next steps from this study and he's seeking advice from the patient community.
Click to expand...
Would he be willing to join here for public or private conversations with more members? If I’m not mistaken, the offer still stands for private invite only subforums if researchers are concerned about having their ideas out in the open. Other members will probably be able to suggest suitable invitees.
 
@Utsikt That's a fair proposal, but since I don't really have a longstanding relationship with him I feel that might overwhelm him off the jump if I offered that to him, assuming he doesn't have that type of experience with patient advocacy. We just started conversing today. I can definitely keep in mind here though
 
EndME said:
I haven't seen anything about SFN in general (or for diabetes or something else) but only in the context of ME/CFS, LC and FM. I think there's generally 2 tests that are used for SFN: A certain type of microscopy of the eye where a combination of fibre measures is taken and this nerve counting exercise following tissue extraction where different protocols are often used. I recall reading studies suggesting outcomes in one experiment don't predict anything for the next (the same applies to the Goebel study mentioned below) but I don't know where I've read it.

There's a somewhat sizeable study with controls by Goebel on SFN in FM, who is a big advocate for SFN pathology in FM, and in that study it's reported that there are no statistically significant differences between controls and FM patients and mean IENFD (intraepidermal nerve fibre density) values for the 3 different regions are presented for both groups but none of results are significant, despite there being a mismatch in BMI matching (and higher BMI has been shown to be associated with lower IENFD in some studies). The corneal results are also not statistically significant. All of that is even without multiplicity correction. Funnily enough the authors still write "In keeping with previous studies, approximately half (48.4%) of the individuals with FMS had IENFD below the normative range for age and sex in at least 1 location" but since they appear no different to controls, to me that can only mean that previous studies have also showed nothing significant.

Whilst I'm generally sceptical of AI this should probably be an area where it should be able to at least help with image classification quite easily.
Click to expand...
Thanks. That's more information that I was expecting. Yes, maybe AI will help standardise results.
 
Dakota15 said:
@EndME Do you mind if I ask how this group would do that?

Asking because I've been messaging with the PI of this study (Franz P. Rischard), who is a RECOVER PI with Univ. of Arizona - we've been discussing next steps from this study and he's seeking advice from the patient community.

If any here have any advice or best recommendations to make, feel free to let me know or share here/DM, and I'll try to pass along
Click to expand...
The experiments performed by Wüst et al are all published in their paper. There's a slight difference in that Wüst took muscle samples before and after exercise, but some of the results in his study were seemingly not driven by this.

If discussing next steps for the study then all of those things can anyways be adapted. The most obvious problem is of course: Nobody wants to publish just a replication attempt because that's hard to publish and not good for your academic career, even in cases where it is the most valuable scientific contribution. So that likely means he will have add something that Wüst et al didn't do, to give his work the stamp of being "novel". In their newer studies Wüst et al are planning to use NIRS and other things already but maybe there's something that fits into Franz P. Rischard's expertise that could be done so that he can do a "replication study + xyz"? Wüst is not doing iCPET so I guess that would already be the advantage here, if it does offer valuable information.

On a little side possibility: If people are collecting muscle samples it would be fairly easy to see if the results by Hwang et al on WASF3 can replicate, but of course that would only be another replication study and not something novel.

Tagging @SNT Gatchaman in case of additional ideas. @Jonathan Edwards asked above whether the diffusion capacity was measured or inferred, maybe that would be something to explore in a follow-up?

Tagging @jnmaciuch as she wanted to do something with muscle biopsy samples which this group seems to have access to.

Regarding a follow-up of the SFN findings: I think one will have to follow an established protocol, there should be blinding and there will have to be sufficient samples sizes in controls and ill people and all data should be made part of the results.
 
EndME said:
If people are collecting muscle samples it would be fairly easy to see if the results by Hwang et al on WASF3 can replicate, but of course that would only be another replication study and not something novel.
Click to expand...
It is absolutely mind boggling to me that scientific research works this way. That study needs replicating yesterday. Replication studies should be seen as doing your scientific duty and boost your career in a sane world, because how else does anything ever get proven?

But we live in a mad world driven by profit and publish or perish and careerism etc etc. Alas.
 
ME/CFS Science Blog said:
The novel aspect of the paper seems to be that they used O2 pathway analysis, which tries to break down oxygen transport into different aspects. In ME/CFS and Long Covid patients, the problem seems to lie in the Dm parameter, which stands for muscle diffusion capacity.

View attachment 28216
Click to expand...

Besides deconditioning, and microvascular unit regulation failure (https://journals.physiology.org/doi/full/10.1152/ajpheart.00278.2003), there is another possibility for reduced skeletal muscle O2 diffusion.

Note that VO2Max occurs far below maximal motor unit usage.

The reduced skeletal muscle diffusion at Peak could potentially be explained by quite different motor unit activation patterns, - muscle afferents act on the motor cortex and also spinal efferents, requiring increased (top down) effort and altering the balance of motor unit activation leading to muscle fibres with poorer O2 diffusion capacity being activated.

A motor unit-based model of muscle fatigue

A motor unit-based model of muscle fatigue - PMC

Muscle fatigue is a temporary decline in the force and power capacity of skeletal muscle resulting from muscle activity. Because control of muscle is realized at the level of the motor unit (MU), it seems important to consider the physiological ...
pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov
 
If I'm understanding the results correctly, does this study ascribe a relatively smaller role to oxidative phosphorylation? This slightly older study from 2010 found no dysfunction in oxidative phosphorylation capacity among CFS participants, but the 2024 long COVID paper from Appelman, et al. regarding muscle abnormalities did find significant diminishment in capacity. I'm approaching this subject with no formal research training, so I'm trying to figure out how to think about these disparate results. Do we have a good sense of the pathology here?
 
One small note about Franz Rischard after messaging with him - he's very close with David Systrom, which I didn't know. Franz said that "We placed two grants to the NIH in the early phase of RECOVER on iCPET and disease mechanisms which were both overlooked. iCPET was then adopted by the RECOVER at large but fizzled out given its lack of resources and feasibility across sites. There are only about 5 or so institutions in the US that have enough volume to be truly feasible. I feel like the best route is to look at non-invasive CPET which is part of the RECOVER at large and see if we can gleam best practices for diagnosing Long COVID, biomarker development for diagnostic purposes, and eventually more targeted interventions."
 
You must log in or register to reply here.
Back
Top Bottom