Post-exertional malaise and the myth of cardiac deconditioning: rethinking the pathophysiology of long covid, 2026, Charlton, Wüst et al

[ScoutB]

My muscles feel extremely tense all the time, especially in PEM.

I’ve seen others mention this and I think I have it too, to a lesser degree. It’s really noticeable if I try to do one of those mindfulness “relax your body parts one by one” exercises — it just doesn’t work, it’s like properly relaxing the muscle is just not an option. It’s so strange.

I agree it seems like it could be related to the apparent lack of deconditioning.

 

[Evergreen]

In the paper, they cite a pre-preprint Wearable heart rate variability monitoring identifies autonomic dysfunction and thresholds for post-exertional malaise in Long COVID co-authored by Wurst.



I haven't read the manuscript but this doesn't hold for cognitive exertion for those of us who don't experience increase in the HR during cognitive activity.

Great. Before I say anything else, I want to say that I am delighted that this team has joined the field and is doing these studies. They are gathering data that is really valuable, and they're listening to patients and taking them seriously. I really hope they keep going. For me, the issue is just that clinical recommendations are being made prematurely.

I do appreciate that they're trying to save people with long covid who have PEM from "normal" rehab that would not be appropriate for them - and there may well be considerable pressure for that in the Netherlands, since the particular experience of ME/CFS patients with exercise did not reach the wider rehab or sports medicine worlds prior to the pandemic. But I worry that they're landing well within the borders of GET.

In the Ruligt study, they did not check whether patients were actually experiencing PEM or not. So they identified lower HRV in patients compared to controls after exercise, but we don't know whether that correlated with symptoms.

I think that's important because in Charlton et al. 2026, clinicians are being told to screen for PEM, currently defined by symptoms rather than by an objective measurement, and if someone doesn't report PEM, they can safely be rehabbed in the usual way, whereas if they do, they're rehabbed differently. I think it's possible that some people who don't report PEM actually have post-exercise reduced HRV, and that some people who do report PEM actually don't have post-exercise reduced HRV. That has some big implications, where some people who shouldn't be rehabbed intensively would be, and some people who would be well able for rehab would miss out.

I also think it's possible that some with perfectly fine HRV after one submaximal exercise test, could, when rehabbed through low-intensity activities on a regular basis, be not fine, i.e. experience deterioration.

And all of this is based on hypotheses, which, while they seem plausible, have not been tested. So I like Ruligt et al.'s future research section:

Future research
Controlled trials with PEM induction and continuous follow-up of heart rate, heart rate variability and other parameters related to the autonomous nervous system would provide a more mechanistic insight into the underlying pathophysiological mechanisms underlying autonomous and endocrine dysregulation during exercise in patients with long COVID, and the putative link with the threshold for post-exertional malaise. A second line of research relates to the practicability of HRV for pacing. Clinical trials on the (lack of) effectiveness of exercise training, pacing and use of wearables in patients with long COVID are urgently required.

But I'd like them to follow people being rehabbed over time and compare them to a control group that is also getting plausible and equal input to actually validate all of this. Otherwise we haven't moved forward from Workwell's similar ideas from a long time ago. Because the clinical recommendation that actually should be made might be (and I think is likely to be) completely different and more akin to a warning.

The amount and intensity of exercise Ruligt et al. mention makes me nervous too:

Exercise intensities were categorised into three levels: mild (80-90% of HR at VT1), moderate (90-100% of HR at VT1), and intense (HR above value at VT1). For every hour after exercise completion (up to 24 hrs), a 15-minute average HRV was computed. Exercise was defined as maintaining a heart rate above the predefined exercise intensities for at least 20 minutes. Because not all healthy controls underwent a submaximal exercise test, heart rate at VT1 was predicted using 75% of the age-predicted maximum heart rate in a subset of the healthy controls.

Currently, no curative treatment for PEM is available, and patients are advised to limit intense exercise above the PEM-threshold. [this is defined as >100% VT1]

This intensity of the PEM-threshold is currently unknown, but as HRV recovery responses were particularly disrupted with exercise of a longer duration (>60 min) and at intensities near VT1, we hypothesize that the PEM threshold occurs near VT1.

I think plenty of GET was well under these threshholds, yet it still made plenty of people worse over time. Mine, for example, did not provoke PEM, it started at 20 mins total, was never >60 minutes (though it was 45 mins some weeks and I did not like those weeks as it felt much harder), and my guess is that it was below or in the mild category above, though I cannot be sure.

I just want to see a lot more data and more circumspection.

Absolutely shattered so won't be back in any substantial way today.

 

[Evergreen]

Quite possibly and I find it possible that breathing exercises also lower heart rate upon HUT or other scenarios but is there evidence to suggest that it does more than that?

I don't know about evidence, but see:

There has been a hype around "calming the nervous system" in LC by doing breathing exercises. Breathing exercises are supposed to improve autonomic function and increase HRV.

Putrino was involved in Meo Health (ex Stasis) programme which is a paid 12-week resonant breathing programme on a phone app. I couldn't be bothered to read their paper on symptom improvement following the programme.

 

[InitialConditions]

I have just subitted the following rapid response. Not sure how often this journal publishes RRs, and this might be too off-topic for a sports medicine journal, so I'm posting here. It was helpful to get my thoughts in order, but now I need a break.

--
Post-exertional malaise cannot be reduced to exercise intolerance and often does not respond well to traditional rehabilitation approaches

I applaud Charlton et al. [1] for challenging the assertion that cardiovascular deconditioning plays a significant role in the development and persistence of long covid and that patients can be rehabilitated through appropriately tailored exercise training [2].

Charlton et al. instead argue that deconditioning alone is not sufficient to explain known cardiac alterations in all long covid patients, and that a growing body of evidence of biological abnormalities and pathophysiological findings in long covid (some of which replicate previous findings in ME/CFS) point to different mechanisms. This, they argue, has implications for the treatment and care of long covid patients, particularly in how rehabilitative strategies are designed and implemented.

Interestingly, despite seeing only a limited role for deconditioning, Charlton et al. reach similar conclusions to the AHA statement and recommend individually tailored rehabilitation strategies, albeit with the addition of certain safeguards to mitigate the harms (i.e., symptom exacerbation and further loss of function) associated with post-exertional malaise (PEM).

The focus placed by Charlton et al on PEM — the cardinal and defining feature of ME/CFS — is warranted, because it is found in a sizeable subset of long covid patients. The authors provide a prevalence rate of 80%, but this is considerably higher than other estimates, including those from systematic reviews and meta-analyses, [3, 4] and likely reflects unrepresentative clinical data.

Presence of PEM, much like absence of deconditioning, has important implications for how long covid is understood and treated. Presence of PEM in long covid helps define a specific ME/CFS(-like) phenotype describing patients with a highly debilitating, perverse reaction to exertion that is not seen in other patients.

Reading the long covid literature, I have noticed that PEM is still widely misunderstood and too frequently mischaracterised. Charlton et al. present a better understanding of PEM than most, but simplistic models based on ‘baselines’ and ‘thresholds’ do not fully capture PEM's inexorability and unpredictability. Further, the incessant focus on physical fitness/activity minimises the role of mental and cognitive exertion in triggering PEM and the significant role that ‘cognitive/neurological’ and ‘immune’ symptoms play in determining overall symptom burden and function capacity in these patients. The presence of these symptoms is generally independent of cardiovascular fitness or physical activity; this in itself should provide a clear rebuttal of the deconditioning myth.

Matters are further complicated by conflation of PEM and exercise intolerance, including in many long covid studies. The AHA long covid statement is really a statement about identifying and treating exercise intolerance (as per its title), not PEM. Confusingly, exercise intolerance might be present in long covid patients in the absence of PEM, and, in some patients, might well be explained by deconditioning and therefore respond well to physical rehabilitation.

I am concerned that many of the mistakes made in ME/CFS research are being repeated in long covid patient populations, particularly in trials of exercise and rehabilitation. Some of these mistakes include: complete ignorance of PEM [5]; poor identification and monitoring of PEM status (and associated harms) in patients; and/or exclusion of the most severely ill patients from research and clinical trials. Because of these mistakes, there remains an absence of evidence that these interventions improve patient outcomes and do not lead to deterioration or other harms, particularly in the most severely ill patients.

I therefore have doubts about the promotion of ‘tailored rehabilitation plans’ for patients with (correctly identified) PEM, particularly if there is no evidence of deconditioning. After all, reversing apparent deconditioning has always been the primary justification for exercise-based programmes in post-infectious chronic conditions. PEM and growing evidence of associated pathophysiological mechanisms can help explain why rehabilitative approaches very rarely ‘shift the dial’ in these conditions, and why many patients report them to be inaccessible, inefficacious, and counterproductive.

Acknowledgments: Comments from members of the Science for ME (S4ME) forum helped shape this response.

[1] Charlton BT, Janssen K, Systrom DM, et al. Post-exertional malaise and the myth of cardiac deconditioning: rethinking the pathophysiology of long covid. British Journal of Sports Medicine Published Online First: 10 February 2026. doi:10.1136/bjsports-2025-111387

[2] Cornwell III WK, Levine BD, Baptiste D, et al. Exercise Intolerance and Response to Training in Patients With Postacute Sequelae of SARS-CoV2 (Long COVID): A Scientific Statement From the American Heart Association. Circulation. 2025 Aug 5;152(5):e50-62. doi:10.1161/CIR.0000000000001348

[3] Li X, Zhang Y, Li J, et al. Prevalence and measurement of post-exertional malaise in post-acute COVID-19 syndrome: a systematic review and meta-analysis. Gen Hosp Psychiatry 2025;92:1-10. doi:10.1016/j.genhosppsych.2024.10.011

[4] Pouliopoulou DV, Hawthorne M, MacDermid JC, et al. Prevalence and impact of postexertional malaise on recovery in adults with post-COVID-19 condition: a systematic review with meta-analysis. Arch Phys Med Rehabil 2025;106(8):1267-1278. doi:10.1016/j.apmr.2025.01.471

[5] Spichak S. Less than 20% of Long COVID trials involving exercise even mention post-exertional malaise. The Sick Times. 2025 Nov 21. Available from: https://thesicktimes.org/2025/11/21...xercise-even-mention-post-exertional-malaise/.

 

[Sean]

That's an interesting hypothesis. How would you show that we're less deconditioned than expected, I wonder?

I think any deconditioning is simply neither here nor there because

• it's not what makes us unable to maintain normal levels of activity in the first place
• for people who are not already improving naturally, reconditioning of any deconditioning without messing up overall function is not possible

Or to put it another way, it's inevitable and it's irrelevant.

Stop trying to rev our broken engines. Figure out what's going on with the bloody engine.

Indeed. The correct response to the oil pressure warning light coming on and the temperature gauge starting to rise is not to press the accelerator harder.

I just hope that Appelman, Wüst & co follow these pwLC being rehabbed and actually start figuring it out. Other publications of theirs that I've read are much more promising. They don't overreach.

I hope this paper is an anomaly.

I would say I am nowhere near as deconditioned as I "ought to be". Some days my muscles are quite wobbly, on better days they are near normal for limited activity. In terms of cardiac deconditioning, just looking at my recent data, my HR still routinely reaches the mid 40s overnight. (Prior to illness it went to low 40s or even high 30s such that I had to turn off low HR notifications.) But if I had significant cardiac / cardiovascular deconditioning in the context of average ~1000 steps per day housebound (following bedbound) for 5 years now, I would have expected a much higher resting overnight HR. I think you'd expect to be >60 during sleep with any meaningful degree of cardiac deconditioning.

A (maybe even the) critical piece of evidence for lack of deconditioning (at least substantial enough to be a major limiting factor in activity) is that on their good days patients are able to immediately be substantially more active, in every sense, sometimes to near the level of pre-illness, without any reconditioning at all, let alone an extended multi-disciplinary course of it from experts.

The classical deconditioning/reconditioning paradigm cannot explain that fact. Indeed, it contradicts that paradigm.

The only possible conclusion being that patients are not deconditioned enough to make much practical difference, and hence are unlikely to significantly benefit from the re-conditioning approach.

Which raises the question of why. Either patients:

1) are not sufficiently inactive for deconditioning to happen, for which we have good evidence from various activity tracking and deconditioning studies over the years, including one of Wust's own recent studies on deconditioning, or

2) are sufficiently inactive to induce clinically significant classical deconditioning but something is preventing or at least substantially mitigating that deconditioning, or

3) possess some miraculous ability to instantly reverse serious deconditioning without any accompanying active rehabilitation program.

Incidentally the implications of that second possibility are enormous, and go way past ME/CFS. It opens up the possibility of treatments for (temporarily or permanently) bed bound non-ME/CFS patients being able to avoid or mitigate deconditioning. It may even help deal with one of the most serious problems facing long term stays in low or zero gravity - extreme general deconditioning.

To quote @ahimsa's signature: It might look like I'm doing nothing ... but on a cellular level I'm really quite busy

 

[Sean]

Very good letter, @InitialConditions. Thank you.

 

[V.R.T.]

Incidentally the implications of that second possibility are enormous, and go way past ME/CFS. It opens up the possibility of treatments for (temporarily or permanently) bed bound non-ME/CFS patients being able to avoid or mitigate deconditioning. It may even help deal with one of the most serious problems facing long term stays in low or zero gravity - extreme general deconditioning.

I think if we get a proper clinical service that does research it should be a priority to investigate this

 

[EndME]

I have just subitted the following rapid response. Not sure how often this journal publishes RRs, so I'm posting here. It was helpful to get my thoughts in order, but now I need a break.
....

Hi @InitialConditions,

thanks for this piece and I appreciate that you kept the tone friendly. Thanks for calling out the prevalence rate and the lack of robust understanding of PEM in the LC field. I can understand that you need a break now (in particular don't feel the need to respond)!

Something I don’t quite understand is that you seem to be debating whether or not patients are deconditioned. I think all severe and very severe patients will be deconditioned to certain degrees and the CPET studies which are almost exclusively conducted on mild and moderate patients seem to all show mild deconditioning if I’m reading them correctly. I don’t think it should be surprising that if someone is unable to walk more than 100 steps a day that they would be deconditioned. However, as the phrases by @Evergreen taken from the PACE manual show is that even in those situations a tailored approach (even if the tailoring is arbitrarily different) does not lead to a higher well-being or more activity in the future. PACE will have included people that are severely deconditioned and people that are not deconditioned, as far as I’m aware nobody showed a difference in outcomes between those groups, nor in the PEM vs no-PEM groups.

Plently of people with Alzheimers, MS and Parkinsons are also deconditioned but without this being a primary focus of anything.

 

[InitialConditions]

the CPET studies which are almost exclusively conducted on mild and moderate patients seem to all show mild deconditioning if I’m reading them correctly.

The crux is what is driving these abnormalities. You could get similar results, for example in CPET, from very different mechanisms, which is sort of what Charlton et al are arguing. I don't intend to argue that deconditioning isn't present in some patients, only that we should challenge the assumption that it is the primary driving factor and that it is reversible through so-called reconditioning.

Perhaps CPETs are revealing similar abnormalities to those found in deconditioning, which might be explained by other factos unrelated to physical inactivity and bedrest?

 

[EndME]

The crux is what is driving these abnormalities. You could get similar results, for example in CPET, from very different mechanisms, which is sort of what Charlton et al are arguing. I don't intend to argue that deconditioning isn't present, only that we should challenge the assumption that it is the primary driving factor and that it is reversible through so-called reconditioning.

I think that would be the consensus of most people here on S4ME. I don't think anybody on S4ME is suggesting that deconditioning is the primary driving factor or even a worthwhile driving factor. I think I got a bit thrown off by things such as "much like absence of deconditioning" and "if there is no evidence of deconditioning". I think we can assume deconditioning to often be the case without much relevance.

 

[Sean]

3) possess some miraculous ability to instantly reverse serious deconditioning without any accompanying active rehabilitation program.

This possibility was, of course, a bit of snark.

If it were true then we would be self-healing instantly all the time and hence would not be sick!

 

[InitialConditions]

I think I got a bit thrown off by things such as "much like absence of deconditioning" and "if there is no evidence of deconditioning"

Ah ok. Yes, these are not blanket statements — they are refering to the individual patient, but perhaps that's not clear. "if there is no evidence of deconditioning in a patient" night have been better.

 

[EndME]

Ah ok. Yes, these are not blanket statements — they are refering to the individual patient, but perhaps that's not clear. "if there is no evidence of deconditioning in a patient" night have been better.

I see. At least for me that would have made it a bit clearer. Of course even all the CBT/GET people sort of realise that pure deconditiong alone is not sufficient to explain ME/CFS because even they admit that you have some fairly conditioned people and you can develop ME/CFS overnight but not decondition overnight.

 

[Utsikt]

They are gathering data that is really valuable, and they're listening to patients and taking them seriously.

I disagree. They can’t be listening if they still believe that an individually tailored rehab approach is suitable. Or that PEM is triggered by going past a certain cardiopulmonary threshold (or something of the like).

They are letting their own preconceptions dominate their work.

 

[poetinsf]

There seems to be an element of randomness and it's very easy for patients to fall into the trap of over-attributing PEM and crashes to exertion/activity alone, even though roughly the same level of exertion/activty can have wildly different outcomes.

There probably are factors other than exertion for the trigger. But it often entails lack of experience when patients in early stage say PEM happens without reason or rhyme. And the signal (known exertion) to noise (unknown exertions and factors) ratio is low in severely ill patients. In most mild or recovered cases only with PEM as the symptom, however, the relationship between exertion and PEM is absolutely unmistakable. Tracking all activities could clarify the relationship (including the accumulation effect) in some cases. The problem is that logging and analyzing takes effort, especially for the severely ill, so that the establishing the relationship is not that easy.

 

[rvallee]

This is a very good point- it would be so ironic if a vital clue was missed because doctors had been confidently stating it was all deconditioning when pwME for some reason don't become as deconditioned as they would br expected to.

There could be some of that, but also decades of exercise studies have pretty much added up to the fact that it takes very little exercise to maintain a healthy body. Not fit, just healthy. Of course the recommendations vary wildly, based on which cherries get picked, but in recent years the shift has been noticeable, making this to be good news, that it pretty much only takes a little movement here and there during the day, that it's actually staying immobile for long periods of time that is the main concern.

Ironically, it could be that it's pacing that helps the most here, as pretty much all of us divide our activities in chunks during the day. I rarely sit down for more than 30 minutes at a time, my very repetitive days are a back-and-forth between sitting down, or taking a nap in late afternoon, and doing a bit here and there, including some light exercises and short walks, basically meeting most of the recommendations for light physical activity. Most people do, unless they are actively lazy, such as always avoiding stairs, always driving to the closest parking spot and just disliking walking or doing anything more exerting than getting up from the couch to grab a snack and back.

And when looking at exercise studies, the amount of exercise that is needed for a healthy level of fitness is also pretty low, anything above it will mainly increase markers of fitness, with little impact outside of showing visible muscle or being better able to compete physically. It's still pretty good to do regular exercise, but the benefits have been wildly exaggerated. The big tell is that people who work physically demanding jobs don't seem to do any better, if they did it would be promoted like a god appearing on toast. Instead, it's all about recreation, hence the shift in recent years from recreational exercise to recreational light activities.

Ironically, by promoting the lie that work is good for health (for most people, the thing they hate the most in their life is work, and the relationship is obviously backwards: being healthy makes it easier to work), the medical profession may have actually made things worse since most people work too much, sitting down for long hours without taking breaks. Plus if exercise were as good as falsely advertised, governments would have long have been subsidizing it to make it free. But not a single one does, because the math doesn't actually math here.

 

[EndME]

The problem is that logging and analyzing takes effort, especially for the severely ill, so that the establishing the relationship is not that easy.

I doubt this is the problem.

And the signal (known exertion) to noise (unknown exertions and factors) ratio is low in severely ill patients.

I'm not sure how one can conclude this without any data and patient reports being fairly mixed.

I suspect it is much more as @InitialConditions describes with randomness and non-linear relationships making the whole thing more complicated then a simple cause and effect language implies. From what I've seen that is typically also how the processes that one speculates underly ME/CFS are typically modelled and if the brain is crucially involved things are likely going to be more complex than the opposite. One can introduce dynamic thresholds, dynamic baselines and accumulation effects, but in the end that only strengthens that things are complicated.

If we had long-term activity and symptom tracking studies we'd probably know more.

 

[rvallee]

There has been a hype around "calming the nervous system" in LC by doing breathing exercises. Breathing exercises are supposed to improve autonomic function and increase HRV.

Yeah, there's that. Ugh. That's one of the rare woo mind-body things that I see being repeated in the LC community. No one knows what it means, but damn does it sound great. Must be like finding nirvana, I guess.

 

[poetinsf]

I think the key to understanding a threshold for ME/CFS is that the effort-response relationship is non-linear, dynamic, and difficult to predict.

It's extremely non-linear in my experience: about 5% increase in intensity/effort requires 50% cut in duration. I'm not sure if the threshold is dynamic though. The inability to measure and predict correctly could make it look dynamic when it is static.

 

[rvallee]

My muscles feel extremely tense all the time, especially in PEM. I can't really relax them without effort. Like they are naturally just tense now. Could this have something to do with that?

Very little has been found wrong in our muscles, but this is a common problem, although so minor it's rarely mentioned, but it is notable that, unless I got it wrong, muscles need energy to relax, not to contract (hence rigor mortis?). So the default state is tense, and something is making it harder than it should for us to return to a relaxed state. Could be a signalling problem.

Of course it's lost in all the 'stress' nonsense, tense in the sense of being anxious, not physically rigid. So much useful information that never gets looked at because it's buried under a pile of garbage.