Review Long COVID Is Not a Functional Neurologic Disorder 2024 Davenport, Tyson et al

Primary author Todd Davenport, senior author Sarah Tyson
(sorry for edits)

 

The fact that ME/CFS (and Long Covid) looks different to functional seizures and functional motor disorders doesn't prove that ME/CFS (and Long Covid) aren't also functional disorders. Functional seizures and functional motor disorders present very differently from each other too. The imagination of psychosomatic medicine proponents has no trouble scooping all of them up into the 'functional' bucket.

 

I have done a quick skim read and can see a lot of thought and work has gone into this article and some useful points are made.

I have some concerns. If I'm misinterpreting the article please say so, I admit I haven't read every word. The neuroimaging part is over my head, for example.

I would want to see the argument framed as:

Neuraesthenia/hysteria/conversion disorder historically has morphed into several 'functional' disorders in some medical quarters, and for those clinicians still comes under the banner of psychosomatic/non organic disorders.

That is problematic because it stifles biomedical research and testing and leads to ineffective psychobehavioural treatments, gaslighting, lack of appropriate care etc etc.

Under that umbrella have variously been placed symptom patterns that can't currently be explained by any known pathology.

They have been subdivided according to symptoms and to some extent precipitating factors into

FND if there are specific neurological signs,

fibromyalgia or complex regional pain syndrome if predominantly pain,

PVFS, ME, CFS if predominantly fatigue with or without infectious trigger,

and MUS, SSD, PPS etc by those who want to lump them all together.

Since the biological basis of all of these conditions is not established, and attempts to treat them with psychobehavioural therapies have been ineffective and in the case of ME/CFS actively harmful, they all need to be regarded as probably having a physical basls, not as psychosomatic.

The symptom patterns of the different subgroups may overlap in some individuals, but the diagnostic criteria are distinct, so there is no reason to use the wrong label for any of them. So ME/CFS and LC are not FND because the diagnostic criteria are different.

I think the framing in this article of FND as the true inheritor of the neuraesthenia/conversion disorder/psychosomatic labelling and LC/ME/CFS as biomedical is a big oversimplification on the history and is unhelpful to people who are diagnosed with FND on the basis of positive signs. It feels uncomfortably to me like in order to insist LC is not FND the authors are suggesting FND is psychosomatic. Have I understood that correctly?

I'm thinking particularly of the diagram in the middle of the paper that starts with neuraesthenia on the left and the upper branch as biomedical ME/CFS/LC and the lower branch as psychosomatic FND. I notice the diagram leaves out the Oxford definition from 1990 that deliberately widened ME/CFS to all unexplained fatigue, and the subsequent 30 years of psychosomatic/psychobehavioural interpretation and treatment that has been so harmful. Edit: This is partly explained away in the text, but still a problematic diagram I think.

To summarise, why not just say LC is not FND because the symptoms are different.

 

You make a lot of good points Trish, and yes, it looks as though the authors have put a lot of time into the paper, but I agree with Jonathan that it is unfortunately counter-productive. I don't think the paper should be used in advocacy.

But that is essentially what the authors say. And, as per my comment above, that doesn't work as an argument, because the psychosomatic medicine proponents are happy to give any health condition or symptom that they believe isn't adequately explained by a physical cause a 'functional' label. Those conditions can have a very different physical presentations (e.g. seizures, tics, post-exertional malaise, muscle weakness). Unless conditions and symptoms have a clear physical cause, they are vulnerable to being labelled as psychosomatic.

There's a lot of reliance on there being 'rule-in' signs for FND that are usually absent in Long Covid and ME/CFS (and it is suggested that if the rule-in signs are present in a person with Long Covid or ME/CFS, then the person has both the post-infection condition and an FND). But, we know that the rule-in signs are not reliable. We know that not finding typical epileptic brain wave patterns doesn't mean that such patterns aren't there, just deeper in the brain than the sensor detects. It looks as though the paper has been shaped by Sarah Tyson's previous commitment to the idea that FND exists and can be (and should be) confidently identified.

There is no consistency in how Long Covid is framed - at times it is recognised that Long Covid is a collection of health conditions with different causes (e.g. ME/CFS, observable lung damage caused by the infection). At other times, as in the quote above, it is treated as an ME/CFS equivalent.

There is a lack of polishing of the paper, which we have seen in other papers by Davenport. For example the first sentence in section 4.2 says

So, people with FND often have chronic dizziness.
But Table 1 reports that ME/CFS and Long Covid (here again treated as one condition) can involve dizziness, but can be distinguished from FND, because FND does not.



There are issues with references, for example, this statement talks about chronic conditions associated with infections

Reference 18 is 'Multiple Chemical Sensitivity Syndrome: First symptoms and evolution of the clinical picture: Case-control study/epidemiological case-control study'.
Reference 19 is 'Chemical intolerance in primary care settings: Prevalence, comorbidity, and outcomes.'
Neither paper is directly related to post-infection diseases or specifically mentions the condition being caused by an infection, and both leave the door wide open to chemical intolerance being functional disorders e.g.

It is difficult to understand why these two references were included.

 

I think I expressed that point too baldly. I was trying to make the point that the symptoms are different but none have evidence of being psychosomatic.

I got stuck on that diagram which seems to me to be wrong historically. Maybe in the USA it's different, so they view ME/CFS as having been decided to be biomedical decades ago, whereas FND has continued to be stuck in the psychosomatic field as the inheritor of the conversion disorder neurasthenia label. In the UK the problem is much more that both ME/CFS and FND have been stuck in the psych category and continue to be so as we see with Miller's awful article a few days ago. And it's clearly still a big problem in the USA, with the likes of Walitt and his effort preference nonsense.

I hope one of the authors will explain where I'm misunderstanding. I can see the intention is good, and some excellent people are involved in the article, so I'm a bit bemused.

 

Yes, but it works against us if the holes in the paper are big enough to drive a truck through. If a patient takes this paper to their GP and says 'look, I don't have FND! because I get dizziness and Table 1 says that people with FND don't', or 'look, I don't have FND! because I have Raynaud's', 'look, I don't have FND! because I don't walk funny and I don't have seizures' I don't think the GP will be convinced.

This is a question closer to the mark. The literature is not yet at a point where we can be certain about any physical marker of ME/CFS-like LC. But we could usefully answer the question 'does classifying LC as an FND help the patients at all?'.

That section on neuroimaging is a nice review of the LC brain imaging field, but the uncertainties noted as relating to FND apply equally to those in the assortment of conditions that constitute Long Covid. Specifically this sums up the situation for both labels beautifully:

For both, the studies are typically small, and all sorts of findings are reported. When I read that summary, I'm not seeing convincingly replicated findings for ME/CFS-like LC that would allow a patient to go for a scan and know with some level of confidence that they have ME/CFS-like LC. I'd be very happy to be convinced otherwise.

 

Perhaps it would be more helpful to frame the argument as Long Covid and ME/CFS are not psychosomatic disorders.

 

I think the paper usefully makes the point that FND equals hysteria and conversion disorder and mental illness, a

That may have some people who have gladly accepted their FND label thinking again.

But, it notes that it has been said that a label of FND can be helpful, and does not really critically evaluate that:

There is a lot that could have been said about the negative impacts of labelling Long Covid and ME/CFS as FNDs (suicide, despair, withdrawal of state and family support, ineligibility for income insurance payouts, redirection of research effort to unhelpful areas, treatments that don't help and may harm....), but I didn't see that in the paper.

That doesn't really work, because, for example psychogenic non-epileptic seizures has a more specific case definition than FNDs in general, but it is still seen by psychogenic paradigm proponents as an FND.

This perpetuates the stereotype of the militant ME/CFS advocates rabidly opposed to a mental health diagnosis that have forced NICE into submission, while neurology and psychiatry clinicians and prominent national medical organisations favour an FND label. It might have been okay if the paper then went on to offer the promised 'compelling evidence', but it doesn't.

 

And boatloads of superficial, ideologically driven, subjective interpretations.

I remain unconvinced there is even such a thing as FND, certainly not in any sense that makes it a reliable and useful entity.

The concept, definition, and diagnostic process are a hot subjective-soaked methodological and ethical mess, and has been since it was first conjured up by Freud (as hysterical conversion), and there is currently no means to treat it. So what is even the point of it, certainly at the clinical, policy, and medico-legal levels?

It remains an empirical, intellectual, and ethical desert, and I predict it always will.

Yes, what about the very real costs of the FND labelling exercise?