What are the necessary conditions and criteria for a theoretical model of ME/CFS?

[poetinsf]

I'd like it to account for the fact that some people feel better with an infection—and for a minority, very much better for a short time after some vaccines.

That reminds me, the alcohol and flu-shot intolerances may need to be accounted for, in case they haven't been mentioned yet. Those seem fairly common.

That said, I think we need to be careful not to become overly inclusive. Given all the quirks of ME/CFS, the list will become unwieldy, and therefore useless, if we throw in the kitchen sink.

 

[Creekside]

That reminds me, the alcohol and flu-shot intolerances may need to be accounted for, in case they haven't been mentioned yet. Those seem fairly common.

These and the lymph node swelling someone else mentioned are some of those "some people show this, some don't" responses and can be explained by "our bodies aren't functioning 100% correctly". My ME gives me double-vision, but I attribute that to some brain cells not functioning properly, so it's not definitive of ME. To be useful, the model needs to be simple, so we should leave out the individual-specific downstream effects and abnormal responses to inputs.

Cases where a PWME has a body function removed or suppressed are useful. If a patient had both kidneys removed for some period of time, yet still suffered ME symptoms, that would be evidence against theories involving kidneys. Blind people with ME rule out they eyes being involved. That sort of data helps simplify the model. Patients with destroyed immune systems (cancer treatment, transplants, etc) might be useful if any could be found with ME.

The lack of a reliable test for ME is a real problem, since without that, we don't know which data comes from people who actually do have ME, rather than just a similar presentation of some symptoms.

@rvalee , are you going to maintain a list of conditions and criteria that we have agreed on, and another of ones up for debate, and maybe one for suggestions that have been rejected? That might help keep the discussion on track.

 

[Kitty]

I'm losing track a bit, so I've started a list of some of the things that have been highlighted for the shortlist of factors to be accounted for.

Not all of them will be necessary or even a priority, I've probably missed off blindingly obvious stuff, and the wording is clumsy (still in PEM). But anyway:

ONSET

• Variety of trigger events
• Sudden or gradual
• Onset appears to have little if any relationship with hormonal shifts
  
• Severity of initial insult has no correlation with ME/CFS risk
• Very wide range of severity levels
• Initial pattern of symptoms often inconsistent between individuals
• No abnormalities usually found on routine screening and bloods
• Affects all ages and sexes, though people assigned female at birth (AFB) are more likely to develop ME/CFS, and there may be age-at-onset peaks

COMMON CORE SYMPTOMS

• PEM (see below)
• Intolerance of physical, cognitive, emotional exertion, with rapid onset of fatigue
• Diminished cognitive function with a wide range of severities
• Intolerance of sensory input
• Rapid muscle fatiguability, but underlying muscle function not usually diminished
• In some, muscles fatigued close to or at failure point may recover within minutes to permit further activity, but fatigue usually recurs after increasingly short times
• Poor, unrefreshing, or inadequate sleep, unusual sleep schedules (early hours to mid-morning, complete day/night reversal)
• Variety of pain types, with muscle pain sometimes significantly worse in certain areas (thighs, neck and shoulders).
• In a minority, pain is largely absent
• Disabling orthostatic intolerance, may occur without marked changes in heart rate or blood pressure
• POTs
• Migraine, persistent headache
• Intolerances of food and medicines, with substantial impact on function and QoL
• Minority feel better during infections or after some vaccines, others see a significant worsening

PEM

• Onset usually delayed
• May be triggered by trivial exertion
• Can be ongoing (‘rolling PEM’), and is compounded by further exertion
• Can be prevented in some by careful pacing, but some have too low a threshold even to tolerate light and sound
• Core ME/CFS symptoms are aggravated, plus additional symptoms that may not otherwise be frequent, including:
  • Dysfunctional sleep (complete insomnia, light/fitful sleep, sleep pattern reversal, hypersomnia)
  • Heightened sensory sensitivities
  • Severely diminished cognitive function
  • Poor motor control, marked clumsiness
  • Sore throat with swollen lymph nodes
  • Adrenaline surges
  • Digestive upset
  • Increased pain
  • Frequent urination
  • Body feels ‘poisoned’
  • Neurological oddities—tics, tingling, muscle twitches, ‘brain zaps’
  • Sensory oddities—phantom sounds and smells, visual disturbances

FLUCTUATIONS

• Medium and long-term fluctuations in severity may be associated with external events, but can also occur spontaneously
• Remission, recovery and worsening can all be spontaneous, and manifest gradually or within as little as 24 hours
• Trigger for PEM not restricted to physical activity in all pwME/CFS
• In people AFB, hormonal shifts frequently aggravate symptoms

RECOVERY

• Complete or near-recovery occurs in a minority
• Recovery seems rare after the initial two to three years of illness
• Some appear to experience complete or near-recovery but eventually relapse (remissions may last several years), and this may happen more than once

 

[Ravn]

Highlighting some quotes I think are important - but also most at risk of being forgotten when considering models and unifying hypotheses. We can get carried away with trying to include everything and the kitchen sink in a model and that's rarely helpful, especially when much of our data limited in many ways (too little, contradictory, anecdotal etc etc)

Basically it would serve as a checklist for any theoretical mode

A model does not have to be complete or explain everything. If a model leads to working treatments it has value. Measurable beneficial outcomes, not just claims based on unreliable data. Such a model might be expanded or even replaced as things progress.

Something that could still be relevant is that we cannot be certain to be working under the assumption that ME/CFS is indeed one "thing".

There may be a root with many smaller branches. Or there may be more than one root branch,

What's more important about a good model isn't so much that it accounts for all things, but allows for them, makes sense of them.

A model accounting for every detail would probably be too complex, it has to be distilled into its most fundamental components. But as long as it makes sense of the data, certainly as long as it doesn't contradict the data, then it's probably good enough. At least as a starting point.

 

[Ravn]

This thread is a valuable thought exercise even if it turns out to be impossible to agree on the ultimate list at this stage due to too little reliable data. Anything to help sharpen reasoning.

A major issue is the difficulty in distinguishing what is core and what is secondary. I think we too often try to squeeze too much into a single or unifying hypothesis and then can't see the forest for the trees

The branching tree analogy is useful to keep in mind here. We all have a unique combination of genetic predispositions, epigenetic profiles, microbiomes, immune systems etc etc etc which means even if we postulate a single disease entity - a big if for starters - and no comorbidities - an even bigger if - we would expect different phenotypes, different tree-branching patterns, growing out of the same type of trunk

Two lists could well be helpful, as a reminder of things to consider when you're in danger of getting carried away with one particular train of thought while cobbling together your hypothesis.

List 1 would be very short and only contain the corest of core and most specific to ME observations, the must-explains.

List 2 would be longer and contain all the other frequently reported signs and symptoms (but not every single symptom ever reported), these would only be should-be-considereds. Still, any proposed hypothesis shouldn't contradict the possibility of them occuring, or at least provide a clear explanation of why the hypothesis disregards a particular item on the list

As we learn more some items may end up moving between the lists, not least because of the risk of false attributions and insufficient or low quality data leading us astray at times. Two examples of where this risk may be at play

1) The observation that more women than men are affected:

This looks to be some of the most solid data we have and, for the record, I'm strongly inclined to believe it is correct. However, before MRI it was thought MS was primarily a male disease, women with the same symptoms were being diagnosed with hysteria instead. Best to keep an open mind to the tiny possibility that we're dealing with differences in rates of diagnosis rather than differences in prevalence. Or to the possibility that male and female ME are different entities that resemble each other only superficially

2) The debate about whether all ME is postinfectious:

This debate can't be settled for now because we have no way of testing for recent asymptomatic or near asymptomatic infections when people first turn up with ME-like symptoms. If they didn't notice or can't remember such an infection but something else major or at least more memorable happened in their life around the time they'll naturally link the onset of ME to that. It remains an open question despite all the stats regularly and confidently thrown about on how many % are supposedly postinfectious versus how many are not

[There are dedicated threads to discuss the pros and cons of the views in the examples so no need to derail this thread by arguing about them here, I just wanted to demonstrate how some of the data we take for granted may yet turn out to be unreliable, or be further strengthened]

Right now I'd put #1 on the core list and #2 on the secondary list, but with a mental footnote for both reading *subject to change

There's also an argument for a third list to record things that don't need to be (re)considered in any new hypothesis because they have been investigated enough to establish they are just an epiphenomenon or otherwise irrelevant (the deconditioning hypothesis springs to mind, obvious to us but not necessarily to anyone new to the field)

 

[Creekside]

In my experience, PEM symptoms are the same whether triggered by physical or cognitive exertion, so I'd model it as a black box that outputs symptoms, and either two input ports, or just one port with a separate box fed by two input ports. A possible biological model might be that cognitive exertion produces some output from certain brain cells, which triggers the black box, but communication molecules from the rest of the body, produced by physical exertion, also trigger those certain brain cells the same way. That model fits the different delays for the triggers.

In my case, I've lost or blocked or broken the physical triggering pathway. Cognitive exertion can still trigger PEM.

 

[Creekside]

Is it too early to propose models? I propose an abnormal immune response in the brain as the root cause, involving a feedback loop that is normally negative, but in ME, it goes positive, locking us into this abnormal state. The vagus nerve could be part of the feedback loop.

This model fits any initial trigger that would activate or involve the immune systems. The reason why these normal events push the feedback loop positive might be genetic, developmental, environmental, or some other reason.

If the mechanism in the brain depends on specific characteristics of brain cells, it could affect different parts of the brain to different degrees in different individuals, resulting in the wide range of symptoms and sensitivities. I don't think any of Kitty's core symptoms can't be explained by this model. For example, in one person, the brain cells connected to the gut are affected, resulting in digestive symptoms. In someone else, the cells processing vision are affected, resulting in light hypersensitivity. Studying the brain structure and comparing with symptom frequency (how many people have it) might reveal something. For example, is there something unique about the brain cells processing pain signals front thigh muscles? Maybe it's the largest grouping, or is closest to some other specific part of the brain, or has more/less blood supply? That's the muscle that is the first for me to have ME pain, and maybe it's just an artifact of artwork, but the Mayo Clinic's diagram of symptoms has "Pain" pointing at that muscle group, so maybe that's common among PWME, and thus indicating something.

Changes in severity can be explained by slight changes in the cells involved in the feedback loop or cells receiving communication from these. For example, changing the ratio of fatty acids in the diet could affect the structure of the cell membranes, changing the "equation" of chemical processing in the cell.

I like the model of a feedback loop because the abrupt switching of state (full ME to full non-ME) is what I would expect from an amplifier that changes the sign of the feedback loop factor. In electronics, making the feedback loop even slightly positive causes the output to snap into the maximum output voltage, whether positive or negative. For brain cells, the output could be production of some protein or change of membrane transport of certain molecules, or some other such thing.

As for why this model doesn't show obvious markers, the changes in the brain cells don't need to be dramatic. What percentage change in protein production, number of a certain type of transport channels, or rate of glial process growth would be required to cause brainfog, lethargy, malaise, etc? Some of these factors probably can't be measured in dead tissue, and we might still not have the technology to measure them in-situ. Experimenting with (safe) brain cell altering chemicals might be a way of testing that model.

I include the vagus nerve as a possible part of the model because it's bi-directional, including immune system control signals, connects to most if not all of the parts of the body that show ME symptoms, and involves the gut microbiome.


So, go ahead and take your shots at this model. That's what it's there for.

 

[Jonathan Edwards]

However, before MRI it was thought MS was primarily a male disease, women with the same symptoms were being diagnosed with hysteria instead.

This isn't right. I was taught that MS was commoner in women as a medical student, long before MRI became clinically available. Men got dismissed as imagining things just as much as women, in my experience.

 

[Jonathan Edwards]

Is it too early to propose models? I propose an abnormal immune response in the brain as the root cause, involving a feedback loop that is normally negative, but in ME, it goes positive, locking us into this abnormal state. The vagus nerve could be part of the feedback loop.

I think this is a major option. The question is whether we need to propose positive feedback in the brain or maybe between brain and immune system or just immune system.

The immune response is generally thought of in terms of white cell like lymphocytes and macrophages, although it would be legitimate to include neural control pathways. I think it is very unlikely that white cells actually in the brain are involved in any feedback loop. Assuming there is no infiltration of lymphocytes as in MS the 'white cells' in the brain are the microglia. These may be recruited to clean up damaged nerve cells but I don't see that as likely to involve any 'flip switch' feedback.

A feedback loop involving vagal signals and lymphocyte responses outside the brain would make sense but that raises the question as to whether the flip to the switch is in brain or outside.

If there is a reversal of feedback in brain it is much most likely that it will involve the sort of feedback/regulatory processes that the brain does very well - but note that these processes are completely invisible to pathological analysis. We have absolutely no way of telling neurons are firing in a positive loop that should be negative because we have no real idea what the firing of any particular brain cell contributes to brain function. In an epileptic seizure negative feedback is override by positive feedback to the extent that the entire brain fires off. But we don't see any pathological changes.

 

[Jonathan Edwards]

I suspect that ideas of loops within brains are not so popular with a number of members because in a sense they imply that ME/CFS is indeed a 'functional neurological disorder' if we take that to mean a disorder driven by disturbed regulation in the brain without any structural pathological change.

I don't think such an idea should worry people, although I understand the concern.

The weird thing about the neurologists and psychiatrist who try to build feedback theories in brain in 'FND' terms is that they get it back to front. Because they are stuck with this idea that people with ME/CFS feel ill because they are expecting to feel ill they come up with this 'predictive coding' idea that PWME feel ill because their brains predict feeling ill and then when signals come in they really do feel ill.

And of course, this is back to front because if your brain predicts feeling ill and normal signals come in you ought to feel really well in comparison to expectation. So they have got the feedback idea wrong. Just as I think Walitt has got it wrong with effort preference.

 

[poetinsf]

2) The debate about whether all ME is postinfectious:

This debate can't be settled for now because we have no way of testing for recent asymptomatic or near asymptomatic infections when people first turn up with ME-like symptoms. If they didn't notice or can't remember such an infection but something else major or at least more memorable happened in their life around the time they'll naturally link the onset of ME to that. It remains an open question despite all the stats regularly and confidently thrown about on how many % are supposedly postinfectious versus how many are not

If you were diagnosed with infection, then you had an infection. If you didn't, then you probably didn't. (Unless it was something obvious like COVID). Some people say that "you could've had an infection and didn't know about it". Well, anything is possible. But saying anything is possible is not exactly a science. We should accept the percentages (something like 30%, I think?) since that is based on the diagnosis.

I suspect that there are more people assuming that they had an infection when they did not, than people who had infection and didn't know about it, because sudden onset so much resembles an infection. I keep using Bruce Campbell as an example. He and his doctor assumed that he had some kind of infection. But he showed signs of ME/CFS prior to it, which he managed to recover from by reducing his load by 10%.

 

[poetinsf]

So, go ahead and take your shots at this model. That's what it's there for.

I thought @rvallee was asking about litmus tests to reject a given model. Are we now working on a model instead?

I still think primed microglia is the best model that explains everything. It explains PEM, all kinds of neural hypersensitivities including brain fog, and all the triggers leading to it. It probably explains the sex difference too since neuroimmunology is often sex-sensitive. The only problem: we don't have an evidence yet! The annals of science is full of hypotheses that make perfect sense that turned out to be not true. If it doesn't pan out though, I don't know what else we can fall back on: there is no other theory that passes (my) litmus tests. It'll be a dark age for the rest of our remaining lives.

 

[rvallee]

@rvalee , are you going to maintain a list of conditions and criteria that we have agreed on, and another of ones up for debate, and maybe one for suggestions that have been rejected? That might help keep the discussion on track.

I'll see where the discussion goes and try to keep track of what comes out. I can't really do this on my own, just hoping this generates enough discussion and refinement to have a few items no one can poke holes in.

 

[rvallee]

I thought @rvallee was asking about litmus tests to reject a given model. Are we now working on a model instead?

I think this is the more useful exercise. Not a model per se, but a list of what can exclude potential models because they contradict reliable data. And hopefully not a done-soon thought experiment, but one that is refined with time as we get more data. We still haven't seen any of the RECOVER studies yet, there will be much more and better data soon to work with.

For sure those conditions can be positive or negative, what a model can be or what it can't be, but it has to be as fundamental as possible, a full model will have a much more thorough set of criteria, unless it happens to be simple, just incredibly hard to find.

 

[JemPD]

I suspect that ideas of loops within brains are not so popular with a number of members because in a sense they imply that ME/CFS is indeed a 'functional neurological disorder' if we take that to mean a disorder driven by disturbed regulation in the brain without any structural pathological change.

I don't think such an idea should worry people, although I understand the concern.

i hope people wouldnt be worried about it, i mean i know what you mean but i think most of us couldnt give a toss whats causing it, frankly i really wish it were dysregulation or whatever gumff BACME likes to spout, i dont think we care if it does turn out to be FND, if there is decent evidence that it is, but its not FND for the reasons they claim.

Same as id be quite happy if it turned out to be psychiatric, couldnt care less, all i know is that its not 'wrong thinking > too much rest > anxiety & deconditioning > symptoms & misinterpretation of normal bodily sensation as sensations of illness'.

Its not that, but I would have no issue with it coming from some other psych issue. I mean didnt i read schizophrenia is immune related? or am i imagining that?

The weird thing about the neurologists and psychiatrist who try to build feedback theories in brain in 'FND' terms is that they get it back to front. Because they are stuck with this idea that people with ME/CFS feel ill because they are expecting to feel ill they come up with this 'predictive coding' idea that PWME feel ill because their brains predict feeling ill and then when signals come in they really do feel ill.

And of course, this is back to front because if your brain predicts feeling ill and normal signals come in you ought to feel really well in comparison to expectation. So they have got the feedback idea wrong. Just as I think Walitt has got it wrong with effort preference.

thank you for putting that so succinctly.

I am always amazed that most of the time i stupidly expect that this time i do 'X' i wont get PEM in the way i have the previous 1000 times, because my body feels like it can do it 'this time'... so PEM is always, without fail, a bit of a shock to me.
Skinner would consider me dumber than pigeons but i just never expect to feel as bad as i do, its always a bit of a shock somehow. I have to re-learn it each time, which is, i suppose why i am so crap at pacing

 

[bobbler]

I suspect that ideas of loops within brains are not so popular with a number of members because in a sense they imply that ME/CFS is indeed a 'functional neurological disorder' if we take that to mean a disorder driven by disturbed regulation in the brain without any structural pathological change.

I don't think such an idea should worry people, although I understand the concern.

The weird thing about the neurologists and psychiatrist who try to build feedback theories in brain in 'FND' terms is that they get it back to front. Because they are stuck with this idea that people with ME/CFS feel ill because they are expecting to feel ill they come up with this 'predictive coding' idea that PWME feel ill because their brains predict feeling ill and then when signals come in they really do feel ill.

And of course, this is back to front because if your brain predicts feeling ill and normal signals come in you ought to feel really well in comparison to expectation. So they have got the feedback idea wrong. Just as I think Walitt has got it wrong with effort preference.

I'm hoping this isn't too much of a sidetrack but this comment made me think of having recently read this (as I thought of clock genes / clock cells and this came up) and thought if it is correct then it begins to undermine the simplistic 'hardware-software / brain controls the body one-way analogy' so popularly used by BPS etc : Liver cells control our biological clock | CNRS

I know ... it is on a mouse model sooo... , and I haven't been able to dissect in depth, but here is an excerpt:

"Scientists from CNRS, Université Paris Cité1 and the University of Queensland2 , however, in the framework of a joint EU endeavour3 , have just shown that the liver also influences peripheral clocks. Teams studied a chimeric mouse model with a liver containing human hepatocytes and observed that the daily cycle of these usually nocturnal animals had advanced by two hours.

The mice became active and began feeding two hours before nightfall, thus becoming partly diurnal. The researchers believe this shift comes from the mice’s central clock being taken over by the human liver cells in this chimeric animal model. These cells can thus affect the entire rhythmic physiology of the animals, including the clocks of the peripheral organs.

Findings suggest that a change in liver clock — for example in pathological condition such as cirrhosis — could affect the synchronisation function of the central clock. This in turn could affect the entire circadian physiology, including the sleep/wake cycle, and contribute to the development of metabolic disease. It also suggests that restoring disrupted liver biological rythm could benefit the entire body metabolism. The hormonal and nervous mechanisms driving this dialogue between the brain, liver and biological clock remain to be identified."

 

[Kitty]

I am always amazed that most of the time i stupidly expect that this time i do 'X' i wont get PEM in the way i have the previous 1000 times, because my body feels like it can do it 'this time'...

But that is the experience for some of us. Maybe quite a lot of us. It's not that we rashly carry on beyond our capacity, because we need someone in a clinic to put us through a course—we can feel as if we have the capacity when we don't.

 

[JemPD]

But that is the experience for some of us. Maybe quite a lot of us. It's not that we rashly carry on beyond our capacity, because we need someone in a clinic to put us through a course—we can feel as if we have the capacity when we don't.

sprry am exhausted & addled... your first sentence is throwing me off...
do you mean that you have the same experience as me & that quite a lot of people do (ie that many of us dont expect to feel ill), despite having had multiple previous experiences of 'X' making us feel ill in the past - & therefore contradicting the BPS notion that we expect to feel ill after activity?

Or do you mean something else? i'm confused sorry

 

[Kitty]

do you mean that you have the same experience as me

Yep, exactly that.

 

[SNT Gatchaman]

I am always amazed that most of the time i stupidly expect that this time i do 'X' i wont get PEM in the way i have the previous 1000 times, because my body feels like it can do it 'this time'... so PEM is always, without fail, a bit of a shock to me.

But that is the experience for some of us. Maybe quite a lot of us. It's not that we rashly carry on beyond our capacity, because we need someone in a clinic to put us through a course—we can feel as if we have the capacity when we don't.

I am one of the "maybe quite a lot of us" who has/does this too. Kiwipom and I discuss this and she's the same. Rather than trying to understand the psychology of pacing (we're describing over-optimism not negativity or negative interoception/predictive coding here), I'd want to add in to the postulates the unpredictability of PEM and maybe the idea that the deficit/threshold can also be reached cumulatively as well as a single over-exertion.