Characterization of Cortisol Dysregulation in Fibromyalgia and Chronic Fatigue Syndromes: A State-Space Approach, 2020, Pednekar et al

I'm just catching up with this now.
Cort has written about the related studies from this team today:
https://www.healthrising.org/blog/2...a-chronic-fatigue-syndrome-neuroinflammation/

My initial reaction is 'bloody cortisol again'. BPS researchers are so convinced that CFS is all to do with stress and 'fight or flight', so cortisol has to be involved somehow. So we get seemingly endless studies of cortisol in CFS. And those studies typically show total cortisol and cortisol taken at points at the day from people with ME/CFS is within normal ranges - although the abstracts often give different impressions. Different sleep-wake cycles and different life-styles are confounders.

I'm not sure I'm making much sense out of those statements from the abstract.

 

Sci-hub link to the Characterisation of Cortisol Dysregulation paper

 

Physiological stress - like pain? But surely if cortisol is a result of pain, then cortisol levels won't tell us much about the causes?

Oh, for goodness sake - they have managed to draw all those conclusions about fibromyalgia on the basis of 3 participants with pure fibromyalgia? And, out of 36 pairs, they discarded 5 pairs because of problems* with the data.

*Edit - given they discarded 5 pairs because the data was 'highly corrupted', presumably 'somewhat corrupted' was fine.

 

I think it gets worse.



So this is Figure 3 - data for one of those three matched FMS/healthy pairs. In the top boxes, the little pluses are the actual data points - cortisol levels in blood as measured every 10 minutes over 24 hours. That's all good. And there's a fitting of the black line to the data points - there's some interpretation there, but still, fine enough.

It's in the bottom two boxes where things get loose. Have a look at the cortisol levels in the top boxes from around the 5 am mark (that is, in the shaded 'Sleep' areas). In Patient 2, there is a moderate spike up to about 15, followed by a high spike of 23 or 25 and another spike of about 15. Now look at the top box of Healthy Subject 2. There is a moderate spike up to about 15, followed by a high spike of around 20 and another spike of about 15. Really, the data in that range looks very very similar.

So, now look at how that data is interpreted and presented in the two bottom boxes. For the Fibromyalgia Patient 2, there's one medium spike, then nothing, then the high spike and then multiple small ones. For the matched healthy control, there are multiple moderate and small spikes through out that period.

From the methods section, it seemed as though the authors were applying different algorithms to the interpretation of the data from each cohort.

I haven't read all of the paper yet, maybe there's something to explain how the different interpretation is justified. But, at this point, it's looking like they have taken data from samples that are way too small and then interpreted it with wild abandon. And, this is just the data from a single pair - the pair they chose out of all the others to illustrate their approach. Perhaps the interpretations of the others look even less valid? Or not. But who knows?

 

I probably should stop there, as with that seemingly random interpretation of the "secretory events" (the spikes), their subsequent conclusions about differences between the numbers and magnitudes of "secretory events" is nonsense. And conclusions about "cortisol clearance rates" and "cortisol infusion rates" is therefore just nonsense on top of nonsense.

But for the fun of it, here's Figure 4:



To be honest, I don't know and don't particularly care how they calculated the upper envelopes and lower envelopes of blood cortisol for each participant. The details are there and it seems to involve a lot of sin and cos. Whatever they did, those curves for the matched pair look amazingly similar. Bearing in mind we are talking about biological systems, where all sorts of factors that weren't controlled for could be influencing cortisol levels, I don't think we could expect those lines for Patient 2 and Healthy Control 2 to be any more similar than they are. And yet this is the pair (out of the 36 pairs) that the authors chose to show us the data for.

The version of the paper I looked at (probably not the final version) seems to me to be a muddled paper that reaches the usual conclusion of 'more research on cortisol in fibromyalgia and CFS is needed'. It's a shame, I'm sure there is useful information in the data, but I don't think this paper reveals it.

 

This study, related to cortisol, has been moved to its own thread:
Epigenetic Modifications and Glucocorticoid Sensitivity in ME/CFS, 2017, de Vega, Vernon et al