Hypothalamus volumes in adolescent [ME/CFS]: Impact of self-reported fatigue and illness duration, 2023, Byrne et al.

I think that's called 'grasping at straws'. Or 'grasping at bits of anatomy'.

So, no difference in hypothalamus volumes, but they found one statistically significant relationship between two of the many things they measured, and there was another relationship that wasn't statistically significant but they still wanted to mention it.

There's no mention in the abstract of the possibility that any change in brain part size might be related to the change of lifestyle that ME/CFS requires, rather than an intrinsic part of the pathology.

This team is from the Melbourne Royal Children's Hospital. In relation to ME/CFS, Katherine Rowe is pretty sensible but Lionel Lubitz is diabolical, a dyed-in-the-wool BPS proponent with power over vulnerable children and their families. Sarah Knight has done some ok work but definitely hovers in the BPS realm. We've seen some of the others in ME/CFS research before too.

 

I think it's important to note that they aren't saying that the ME/CFS group had smaller right anterior superior volumes than the healthy controls. I haven't looked at the data yet, but it's likely that the size of the right anterior superior volumes in the ME/CFS and healthy controls overlap to a great degree, as otherwise there would be a statistical difference between the two groups. What they are saying is that there was a relationship between the fatigue level reported by the 25 young people with ME/CFS and the size of this particular part of the hypothalamus, presumably with those reporting higher fatigue having smaller volumes. Healthy controls might have had similar right anterior superior volumes, but they didn't have fatigue.

Maybe that's an interesting finding. For sure, it should be looked for in other imaging of people with ME/CFS. But, the chances are that it is just a cherry, plucked from the orchard created by the facts that 1. there are a lot of ways to slice and dice a hypothalamus and 2. they probably collected a lot of data from the young people on all sorts of things to use in regressions. Even if the difference is real, the chances are high that it is related to changes in lifestyle related to fatigue severity.

I think it's worth thinking about how accurate any reporting of fatigue severity is likely to be in the relatively small ME/CFS group - I doubt that it is particularly accurate. For a start, the sample is unlikely to include young people with very severe fatigue. Most are probably clustered in the mild to moderate range. In that range, fluctuations in the illness will affect the levels of fatigue experienced and reported on any one day, fluctuations in a time scale that is unlikely to be relevant to brain volumes. Another thing is that a young person who has well-managed ME/CFS and is not pushing at the bounds of their PEM threshold may not actually have a lot of fatigue, but their PEM threshold may be a lot lower (and therefore their illness more severe) than someone who rates their fatigue level higher.

It would be interesting to see what tool they used to measure fatigue, to consider its likely accuracy in measuring fatigue (as opposed to some other symptoms and behaviours) and in measuring the severity of fatigue.

It would also be interesting to see how many regressions between participant characteristics and hypothalamus structure size they conducted, in order to find this one statistically significant one.

 

The study is written up clearly. They use a statistical analysis approach that I'm not familiar with, so that makes it a bit harder to know what is going on.

Brain measures
They had data for volumes of whole hypothalamus (left and right) and 5 regions (left and right). That's 12 possible brain volumes.
They compared each of those in ME/CFS vs controls. None of those 12 comparisons found any significant difference, at all.


This study is part of a bigger study which collected a lot of data from questionnaires and cognitive testing. But this study focussed on fatigue and illness duration.

Fatigue measure
Fatigue was measured by the Pediatric Quality of Life Multidimensional Fatigue Scale - Child report. This scale has 3 sub scales: General Fatigue; Sleep/rest fatigue; Cognitive fatigue. Each one has 6 questions, examples are given:
General fatigue e.g. I feel tired; I feel physically weak
Sleep/rest fatigue e.g. I feel tired when I wake up in the morning; I rest a lot
Cognitive fatigue e.g. it is hard for me to keep attention on things; it is hard for me to remember what people tell me

These are converted into a 0 to 100 scale. Importantly, high numbers mean less fatigue. the paper doesn't say if there is a Likert scale associated with each question, or if it is just a yes/no response. The scale definitely has a problem with the respondent's frame of reference. A person with mild ME/CFS would answer 'yes' to 'I rest a lot' comparing themselves to when they were healthy, not really realising that they would be resting a lot more (and in a typically much more boring way) if they had severe ME/CFS. So, questions there, and some major problems too, I think.

Illness duration measure
4 point scale: 3-6 months; 7-12 months; 13-24 months; >24 months. From parent interview.
About a quarter of each participant was in each group.


Here are the charts relating to the one statistically significant finding that is reported - Right anterior-superior volumes.
Remember that higher numbers on the Fatigue Scale mean less fatigue.



Looking at that data, I think it's pretty unlikely that there is a real difference here. There's a couple of outliers from the ME/CFS group that are having a big impact.

(crosspost with SNT Gatchaman)

 

I think the correction is for differences in volume due to age, sex and total brain size. Which is probably fair enough.

 

They were just looking at the volume of bits of brain, which is a reasonable thing to do. I wouldn't expect the bits of brain to change in size with PEM or for the size to change quickly, although I could be wrong about that.

 

Thread with some ideas about why I might be wrong about that here:
Errors and other considerations in brain imaging