PEM discussion thread - post-exertional malaise

[poetinsf]

I haven't felt the need to go back and try to recreate it since I felt like I got all the insight I was going to get out of it. But I was very surprised at how good the predictive accuracy was when I fed it new data from a few weeks of recording. I don’t remember the exact number unfortunately, but I think it was in the range of 0.85–are you getting something similar?

0.85 is absolutely phenomenal. I was getting more like 10-40% range on average. But that was still a lot compared to other model including heart rates and plain calorie expenditures. They sometimes gave me negative correlation, lol. That's why I said it stuck out like a sore thumb despite the low correlation. The inaccuracies of fitbit minute-by-minute data contributed to the low score, I'm sure. I was able to raise it to as high as 80% by removing outliers, but that felt like cheating because there usually were more than a few outliers.

The performance of model is likely to vary from people to people. We need to get the model out there for other people to try and see how it performs. A few weeks is probably too short though. I ran mine for almost 2 years, both on monthly basis and in rolling windows. Some windows would come up occasionally with higher correlations of 50% or more. But they were fairly consistent in 10-40% range while other models were about zero.

 

[poetinsf]

Thanks @poetinsf. Clearly predicting PEM/a crash is really complicated, but I agree it's a puzzle no more complicated than weather forecasting. If we could crack it, if we could make fairly good predictions, I think the model would give us some clues about the pathology. It just takes a lot of data and insight.

Yep, that's what I've been saying: a black box approach can provide clues about what's inside the box.

Do you think that novelty or a time sensitive activity can initially improve capacity but contribute to PEM?

Actually, that hasn't been my experience. When I'm on the road living in my car, my tolerance boost stays up indefinitely. It's when I return home that I crash before returning to steady state. My guess is that some brain chemicals are protecting me by downregulate brain immune system. Then, when that chemical dips after returning, the bottom falls off. I think it's a similar process as post-trip blue. Dopamine dips when people return from trips and they feel depressed for a few days.

Same for caffeine, nicotine, other stimulants, things that are exciting in a good or bad way, being upright, having an infection?

3.5mg of nicotine boosts me without any adverse effect the next day. Caffeine and Sudafed also works for me without adverse effect. But 7mg nicotine did crash me badly the next day. So, maybe it's about the dosage? Higher dosage could be depleting the natural chemicals leading to crash.

 

[poetinsf]

If I was trying now to quantify activity to predict PEM I would need a measure of time spent upright as orthostatic intolerance has become much more significant over the last ten years when overall I am much worse.

I use 24hr timer that I got from Amazon for $3. I have one next to my bed, and I punch it whenever I lie down/get up. I'm hoping to automate it with fitbit or other wearables.

However, your tool is something that could be used by others who may need to tweak it for the aspects important to them, such as OI or sensory intolerances.

Plug-ins sound like a good idea.

 

[jnmaciuch]

0.85 is absolutely phenomenal. I was getting more like 10-40% range on average. But that was still a lot compared to other model including heart rates and plain calorie expenditures. They sometimes gave me negative correlation, lol. That's why I said it stuck out like a sore thumb despite the low correlation. The inaccuracies of fitbit minute-by-minute data contributed to the low score, I'm sure. I was able to raise it to as high as 80% by removing outliers, but that felt like cheating because there usually were more than a few outliers.

There definitely was a lot of fine-tuning to change how variables were calculated, which probably contributed to getting such a high accuracy. I played around with some random forest models as well to try to work out the best way to encode different variables—that’s how I landed on things like differentiating heart rate spikes from “amount of time spent with heart rate above a certain threshold”, for example.

All those little fine tunings probably massively overfit the model to me specifically. I also think that being a stimulant for most of the trial and not having common ME issues like orthostatic intolerance probably made the model very specific to my own biology.

Though in some ways I think that might be the case for any model that one was to build for these purposes. Just thinking about how one could possibly set this up at a larger scale with any hope of getting a good result gives me a massive headache.

Like @Peter T said, it seems like an exercise that would be useful for the individual to tease out patterns at the beginning of their illness, provided that all of that tracking isnt too costly. Which will probably only apply to a small portion of pwME, though I won’t knock the potential usefulness even for a small group.

So maybe it's possible that norepinephrine/dopamine act like fire retardant against PEM fire. Just something I've been musing about for a while.

That’s my exact experience as well. Once I started taking a stimulant regularly, my PEM changed so that it started as a more immediate build up of pain/muscle stiffness/feverishness. On days where I had to commute downtown, I would feel the symptoms start creeping up a few hours in and would know that I ought to head back home. But as long as I was out of the house, things would stay suppressed to an extent.

It was only once I got home and was able to fully collapse and close my eyes in a quiet room that the floodgates would open and I would end up in so much pain/fatigue/etc that I could no longer push through (and would stay like that for several days).

It very much feels like trying to keep certain types of pain under control with NSAIDs—it will work so long as you’re taking it both preemptively and consistently, but if you allow the NSAIDs to lapse, things spiral out of control and can’t be brought back under control with the same dosage.

I think it’s also likely for the “raging fire” of PEM to further trigger something having to do with adrenaline as a compensatory response (even if it’s one that ultimately fails), if my PEM insomnia is anything to go by.

And same as you, “pre-emptive” caffeine/stimulants or stressful situations seem to allow me to keep going for longer.

 

[Hutan]

I think I've said this before somewhere, but adrenergic or dopaminergic environment seems to protect me from PEM. Whatever wakes my brain up, including pseudoephedrine and caffeine, seems to help.

And I've said the same - excitement, novelty, stimulants, even stress seem to protect me from PEM. At least in the short term and until it doesn't. When my son returned to school a couple of years into the illness, we thought he was going well for a while, he was happy and enjoying things, but he was in fact slowly physically deteriorating. That process took months but led to him being bedbound for a short time and much worse than he had been prior to the school return for a long time.

I can't think what the mechanism could be, what could create that initial protection but only for a certain amount of time.

I do like the idea of interferons accounting for PEM symptoms, and activated t-cells producing the interferons. And adrenalin being one of a number of things that activates t-cells. Perhaps there is some mechanism running counter to that can make adrenalin release protective, but that mechanism is progressively worn down with constant use?

Sorry, I'm not expressing that well.

 

[Jonathan Edwards]

I do like the idea of interferons accounting for PEM symptoms, and activated t-cells producing the interferons. And adrenalin being one of a number of things that activates t-cells.

It's an interesting paradox.

One way out might be that we are dealing with 'no' T cells rather than 'yes' T cells. In more technical speak we may be dealing with T cells that have variously been called suppressor or regulatory or other things.

There are a number of things that suggest that ME/CFS is about activating an off switch where diseases with a more established immune basis tend to be about activating an on switch. The most obvious is that there is so little to find in terms of inflammation or damage.

The immune system is full of paradoxes. TGF beta mostly switches off inflammation but one way it does it is to down-regulate MHC Class I molecules like HLA-B27 that interact with T cells. But absence of MHC Class I can also be pro-inflammatory because if NK cells do not see it with their killing inhibitory receptors they will kill the cell and likely generate inflammation. Which may be why ankylosing spondylitis is a disease that affects the 5 places where you would expect TGF beta to be most active - sites of tissue tension.

 

[jnmaciuch]

I can't think what the mechanism could be, what could create that initial protection but only for a certain amount of time.

I think that would be explained by any agent with a moderate "immunosuppressive" effect (I put this in quotes since what counts as immunosuppressive in what particular context is not straightforward by any means). Same as in the NSAID example--it's possible for a systemic response to be suppressed but only if things are at a manageable level to begin with. Otherwise it's like using a tiny fire extinguisher for a raging house fire. Especially if the underlying effect is a positive feedback loop, like the type I interferon response.

Additionally, I think that the interaction between interferon and adrenaline would extend well beyond T cells in potentially interesting ways. I don't know of anyone who has looked at the effect of adrenaline on tissue-specific interferon production but it's something I've been interested in

 

[Sean]

And I've said the same - excitement, novelty, stimulants, even stress seem to protect me from PEM. At least in the short term and until it doesn't.

Yes, at least sometimes it can be overridden for a bit, but then comes payback time.

 

[Jesse]

And I've said the same - excitement, novelty, stimulants, even stress seem to protect me from PEM. At least in the short term and until it doesn't

I will add that this is the same for me. To the point that excessive resting and lack of novelty/stimulation actually makes me worse in the long run. It's a delicate balance.

On dopamine it's also interesting to note that dopamine is a precursor to adrenaline so there is a relationship there.

I think that would be explained by any agent with a moderate "immunosuppressive" effect

I know I might sound like a broken record but isn't cortisol considered fairly immunosuppressive? I've also read some anecdotes on Phoenix Rising, Reddit (and here too now that I looked) of corticosteroids being an effective PEM-blocker for some. As well as improving symptoms in general, but mainly in the short-term. I've had similar periods of running on stress for longer periods and then crashing hard. I haven't tried corticosteroids myself but I could totally see it having such an effect. Now whether it "blocks/prevents" PEM or just "masks symptoms" I don't know.. maybe those are even the same.

I do like the idea of interferons accounting for PEM symptoms, and activated t-cells producing the interferons. And adrenalin being one of a number of things that activates t-cells. Perhaps there is some mechanism running counter to that can make adrenalin release protective, but that mechanism is progressively worn down with constant use?

So I wonder if cortisol could be the protective mechanism? Adrenaline's effects are rather short-lived but does also trigger the release of cortisol which can remain elevated long-term and suppress the immune system. Then over time chronic high levels of cortisol might downregulate the release of more cortisol.. (as the body seeks homeostasis) and then the interferons take over... --> PEM time.

I could also see this explaining why PEM can be delayed. My experience is that when I don't overdo it too much and can relax afterwards I only get mild symptoms and get them fairly fast. While if I overdo it big time and have some stress/excitement symptom onset is delayed till my body gets in resting mode.

Just spitballing here.. I don't know much but stress has been a topic of great interest to me since getting ME/CFS and struggling with excessive feelings of stress and anxiety the first few years.
On that note: Why Zebras Don't Get Ulcers by Sapolsky is a great read on the biology of stress.

 

[Hutan]

So I wonder if cortisol could be the protective mechanism? Adrenaline's effects are rather short-lived but does also trigger the release of cortisol which can remain elevated long-term and suppress the immune system. Then over time chronic high levels of cortisol might downregulate the release of more cortisol.. (as the body seeks homeostasis) and then the interferons take over... --> PEM time.

Except that people with ME/CFS tend to have normal levels of cortisol. I don't think there is anything particularly unusual about the levels of adrenalin and cortisol that we have. If they are part of the story, it's their effect on other things.

I have this paper in my many tabs to have a look at, I've only read the abstract
Cortisol and epinephrine control opposing circadian rhythms in T cell subsets
It probably doesn't have much to do with what we are talking about, but, it does illustrate the multiple axes of variation going on in biology - it's not just levels of a substance, but where in the body it is, often precisely where in the cell, and the ratios of other substances and so many more factors. And this suggests that the time of day can also have an impact on the effect of the substance on various sorts of t-cells.

It's a 2009 paper, so presumably the understanding of this has improved since.

Another interesting point is that this paper reports that cell counts of specific T-cells vary with the time of day and exposure to cortisol and adrenalin. That could be a confounder in immune cell count studies, especially given people with ME/CFS participating in trials tend to have different waking times to healthy people.

Abstract

Pronounced circadian rhythms in numbers of circulating T cells reflect a systemic control of adaptive immunity whose mechanisms are obscure. Here, we show that circadian variations in T cell subpopulations in human blood are differentially regulated via release of cortisol and catecholamines.

Within the CD4+ and CD8+ T cell subsets, naive cells show pronounced circadian rhythms with a daytime nadir, whereas (terminally differentiated) effector CD8+ T cell counts peak during daytime. Naive T cells were negatively correlated with cortisol rhythms, decreased after low-dose cortisol infusion, and showed highest expression of CXCR4, which was up-regulated by cortisol.

Effector CD8+ T cells were positively correlated with epinephrine rhythms, increased after low-dose epinephrine infusion, and showed highest expression of β-adrenergic and fractalkine receptors (CX3CR1).

Daytime increases in cortisol via CXCR4 probably act to redistribute naive T cells to bone marrow, whereas daytime increases in catecholamines via β-adrenoceptors and, possibly, a suppression of fractalkine signaling promote mobilization of effector CD8+ T cells from the marginal pool. Thus, activation of the major stress hormones during daytime favor immediate effector defense but diminish capabilities for initiating adaptive immune responses.

 

[Jesse]

Except that people with ME/CFS tend to have normal levels of cortisol. I don't think there is anything particularly unusual about the levels of adrenalin and cortisol that we have. If they are part of the story, it's their effect on other things.

I don't mean to say they're abnormally high like we see in Cushing Syndrome but they could still be elevated temporarily (within normal ranges). Thus leading to immunosupression. Or maybe it blocks/mask symptoms through a neural pathway. But to me it does seem the stress system is involved in delaying / blocking / masking PEM and possibly also triggering PEM.

But like I said I'm spitballing, maybe there's nothing there. It just seems to fit with what I know and experience!

 

[jnmaciuch]

I know I might sound like a broken record but isn't cortisol considered fairly immunosuppressive? I've also read some anecdotes on Phoenix Rising, Reddit (and here too now that I looked) of corticosteroids being an effective PEM-blocker for some. As well as improving symptoms in general, but mainly in the short-term. I've had similar periods of running on stress for longer periods and then crashing hard. I haven't tried corticosteroids myself but I could totally see it having such an effect. Now whether it "blocks/prevents" PEM or just "masks symptoms" I don't know.. maybe those are even the same

It can be variable based on the cell type you’re looking at in what circumstances, but in most cases the answer is yes. corticosteroid injections are commonly used in treatment of RA (for example) for that reason if joint inflammation is not well controlled by other therapies. There are just problems that happen with chronic use