UK: Aberdeen Uni: Major new study aims to increase understanding of fatigue

For sure there's no stopping a disingenuous interpretation of the results. Obviously there will be a correlation between inactivity and fatigue and it should be blindingly obvious that you're doing less because you're tired and not the other way around - but as we've seen it's possible to spin any result into a prefab conclusion of your choosing.

If good data is collected though you could have a sincere exploration of possible causal relationships by looking at delays between activity levels and fatigue worsening. For example; given you've seen in the data an increase in steps in this time period relative to a global average baseline, do you see an increase in reported symptom severity in 10 days from now? I would hypothesize that when look over many time periods and individuals the answer to that would be yes, and that the same question phrased as a reduction in activity would not lead to a worsening of fatigue.

 

I'm no expert - I assume that ME versus CFS is based on the outcome of a questionnaire re "PEM"? If you respond in a way that indicates you have "PEM" then you're allocated to the "ME" category; if not then you're in the CFS category - correct?

 

I have not seen more on this study than is post in the thread but I think people are going off on tangents a bit.

PEM is not a specific symptom. It is a temporal pattern for some symptoms that can occur with various conditions - a temporal pattern of 'post-exertional exacerbation'. But I am quite sceptical that we have a clear idea of how that pattern should be defined. I had lunch with Travis Craddock, from the Klimas group this week. We both noted that we had dreadful post-Covid 'fatigue' - which included all sorts of symptoms including nausea and orthostatic intolerance - for a few weeks. Travis said he had PEM, but he meant feeling even worse after gym work - which I take to be standard with any viral infection and not what discriminates ME/CFS. He agreed on reflection. But there is nothing specific about this in any written definition of PEM.

I would much prefer a project like this to forget any preconceived idea of PEM and just gather data and look at the 'granularity' - i.e. see what the patterns really are. It may well be that PEM is a slightly off target description of something that can be defined quite precisely, but rather differently. In the evolution of our understanding of forms of arthritis there are several examples of concepts that were no quite on target that got sorted out with detailed analysis.

And if this is primarily looking at Post-covid fatigue in comparison to specific illnesses like heart failure then its value will be in showing that patterns can be found, not how it handles ME. To me this is exactly how the Covid epidemic should be exploited - as a way to develop techniques that might be relevant to ME/CFS in the longer term.

 

AI, if it is well programmed and fed valid data, is potentially better at spotting patterns than humans.

This looks to me like a promising avenue IMHO, to provide unbiased empirical assessment of physiological patterns in patients and characterise them for report to human medics and suggest therapeutic recommendations. If its done properly it should learn what works and what doesn't and remove prejudice from the battlefield, I mean doctors surgery, forever!

For CFS/ME patients, the devil is in the detail.

Currently it does not address ME at all. If it can be made to work effectively with its target patient groups, I would like to see it using more data types which can be added to a wrist device, such as pulse oximetry and maybe additional monitoring like urine analysis etc and expanded to include other conditions, eventually including ME patients to try to discover characteristic patterns of readings.

Whereby it might have better luck discerning subtypes of ME than human observers have had to date. Though I suspect it will need a larger range of data capture to achieve that.

I think after a few chats with GPT, that we are not far off some useful general intelligence language models and would hope that more focussed data driven AI like this might also be truely useful in a few more iterations. So I think its an investment in the long term development of medical AI.

As it stands this is not going to unravel the mechanisms of ME. Though by spotting subtypes and patterns of associated physiological phenomena it might assist that process by allowing more carefully defined cohorts.

I for one welcome our new overlords!

 

I saw a comment on twitter from a LC-aware MD who mentioned a colleague who noted that the people who recover from LC are the ones who exercise. So close to getting it. But completely unable to. They don't hear anything the patients are telling them about it. The dogma is too rigid, the need for this pseudoscience to be true is too strong, the alternative too damning.

This is the biggest problem we face until there is a biomarker that ends all this absurd controversy. They will believe what they want to believe, even if evidence otherwise is right in front of their faces. And whether or not they actually perceive this, rather than simply comply with what's expected by their employer, that's just a distinction without a difference.

We're seeing the same problem of meh intentions going badly with the JAMA RECOVER paper, covered exactly the way the authors didn't want to, but was guaranteed to happen because of the way they framed it: those are the key symptoms of LC. And then there's the extra bad faith people that will distort everything even when it's indisputable, because when an issue is political, all the facts are fluid and open to arbitrary interpretation.

The problem isn't technical or has anything to do with us. It's medicine and its culture and traditions. We can do everything right and still fail. In fact we do most things right and still failed every single time. So it's critical that any effort takes into account the reality of this issue being politicized and where messaging is far more important than evidence, until evidence makes all the messaging obsolete anyway.

 

It's as funny as them believing that the patients who recover are those who never see a doctor or go to a hospital so that must be the cure and treatment isn't it. SO in effect the existence of medicine itself by that theory would be what causes illness.

Yet they never fall into that trap, so it is chosen selective thinking rather than complete warpedness which makes it even harder to believe it isn't seeing what they want to see.

 

I wonder if it truly is something that occurs with any viral infection or something that can occur in some viral infections in some or all people. Could it represent a brief period of latent viral reactivation, that causes symptoms that are limited in time and scope (a very light and transient form of ME)? The difference being that in a postviral fatigue syndrome of, say a few weeks or months, it persists for a bit longer but does spontaneously resolve; but for ME/CFS that resolution does not occur and the latent viral reactivation persists and is in fact reinforced. In the extreme situation, the reinforcement is much more marked, the situation rapidly worsens, symptoms are very much worse and you have "severe ME/CFS".

I've had in my mind the concept of an "arrow of recovery" which can be pointing sharply or shallowly upward, flat, or shallow or sharply downward. This would depend on physiological factors, possibly including things like genetics, or maybe factors relating to the specifics of viral latency. For those whose arrow is pointing upward, they would recover unless they do something unwise. For those whose arrow is pointing very high, they have more license to exert and push-through - and it may be that in those, the benefits of exercise are gained and outweigh any penalty; and recovery could then be accelerated. Eg in the Garner situation, he may have been inevitably recovering, due to a favourable set of physiological circumstances, although that's not how he saw it as we've subsequently heard ad nauseam.

For long-term pwME, perhaps the arrow is flat for many years, but something else happens (eg a new viral infection) and tips the arrow downward and symptoms worsen.

When I initially read your comment, which used the word "included", I wondered if we ought to draw a distinction between fatigue and other things like nausea and OI, ie as separate symptoms. But I think the way you phrased it could be right. Ie if fatigue (in the sense of exhaustion and reduced energy, as opposed to the feeling of being tired and wanting to sleep as per most media) underlies these other symptoms. A hypothetical explanation would be: OI being caused by inadequate ATP supply to endothelial cells leading to endothelial dysfunction; nausea due to inadequate ATP in gut mucosal, smooth muscle and nerve cells with resulting poor and uncoordinated motility and increased gut wall permeability and inflammation.

Or it could be that fatigue and PEM are actually separate but related effects, where PEM relates to the sort of biochemical changes (or lack thereof) that Maureen Hanson's team are uncovering. Ie does lack of energy production capability underlie the inability to prevent PEM? Or is it more related to changes in the immune system such that what should be done to clear up the toxic effects of exertion can no longer be? Either way I would imagine dysfunctional* mitochondria could be at the centre.

---
* Dysfunctional might not be the right term, see Multifaceted mitochondria: moving mitochondrial science beyond function and dysfunction (2023, Nature Metabolism)

 

Indeed and I think the pattern-seeking being WITHIN individuals is going to be vital given the latency + size/type of symptoms (and mapping patterns in those and longevity) needs to relate to the ‘relative magnitude’ of that activity (vs that persons’ norm) for that person. THEN compared to others to show this same pattern happens in x%.


If it is done other other way around and data averaged across persons for size of activity, latency before symptoms etc then you are taking away that ‘bodily environment’ as well as that individual being the one ‘scale’ ie people might be pretty accurate at in themselves describing severe x symptom from moderate, even if one persons’ severe can’t be as easily objectively mapped to be the same ‘severe’ as the next persons.

Plus of course we hear some get PEM consistently hitting at 36hrs, others 24 but also where PEM starts and ends is still up for debate because there is tge ‘over threshold’ and the tired after that and then the PEM hits like a lovely additional reaction on top and fo we say it has ended when you can sit up again or function a bit or are definitely back to where you were but given how many of us live in non ideal circumstances that will rarely happen. So rolling PEM (which most of us will have lived in for years at some point and I imagine many won’t get the ability to be out of OEM completely much for years when they are first wondering what is wrong with their bodies) I get as a concept of ‘fatigue/fatiguability’ it just needs a heck of a lot more granularity within it to steal Jonathan’s term. I

I would like it to include the bone-type pain though as that certainly accumulates from not being able to rest it/having to consistently go over limits on walking for example and even when you finally rest it away ut is back from feet on the floor for maybe just an hour. Before that it was calves just from driving or sitting. I could have been thinking I’m not ‘fatigued’ if someone described tiredness stuff and I was comparing one week to past weeks and yet my legs and other physical indicators could be worse than ever yet certain questions make my replies fubctionedly-fixed to exclude such.

 

Apparently Dr. Todd Davenport is developing an instrument to measure post-exertional malaise. Anyone here (e.g. from Canada) who knows about this? Perhaps we are, at last, seeing research aimed at applying actimetry [Fitbit type devices] to fatigue.

 

I’ve long wondered about exactly this too. But I didn’t have an arrow metaphor to divide up my groups so it took me ages every time I wanted to refer to it. I’ll save your post and save time next time!

Most grateful.
Yours a falling arrow.