Are current chronic fatigue syndrome criteria diagnosing different disease phenotypes?

[Murph]

Are current chronic fatigue syndrome criteria diagnosing different disease phenotypes?

• Laura Maclachlan,
• Stuart Watson,
• Peter Gallagher,
• Andreas Finkelmeyer,
• Leonard A. Jason,
• Madison Sunnquist,
• Julia L. Newton

• Published: October 20, 2017
• https://doi.org/10.1371/journal.pone.0186885

Abstract
Importance
Chronic fatigue syndrome (CFS) is characterised by a constellation of symptoms diagnosed with a number of different polythetic criteria. Heterogeneity across these diagnostic criteria is likely to be confounding research into the as-yet-unknown pathophysiology underlying this stigmatised and debilitating condition and may diagnose a disease spectrum with significant implications for clinical management. No studies to date have objectively investigated this possibility using a validated measure of CFS symptoms–the DePaul Symptom Questionnaire (DSQ).

Objective
To examine whether current CFS diagnostic criteria are identifying different disease phenotypes using the DSQ.

Design
Case control study.

Setting
Clinical Research Facility of the Royal Victoria Infirmary, Newcastle upon Tyne, UK.

Participants
49 CFS subjects and ten matched, sedentary community controls, excluded for co-morbid depression.

Main outcomes and measures
Self-reported autonomic and cognitive features were assessed with the Composite Autonomic Symptom Score (COMPASS) and Cognitive Failures Questionnaire (COGFAIL) respectively. Objective autonomic cardiovascular parameters were examined using the Task Force® Monitor and a battery of neuropsychological tests administered for objective cognitive assessment.

Results
Self-reported autonomic and cognitive symptoms were significantly greater in CFS subjects compared to controls. There were no statistically significant differences in objective autonomic measures between CFS and controls. There were clinically significant differences between DSQ subgroups on objective autonomic testing. Visuospatial memory, verbal memory and psychomotor speed were significantly different between DSQ subgroups.

Conclusions and relevance
The finding of no significant differences in objective autonomic testing between CFS and control subjects may reflect the inclusion of sedentary controls or exclusion for co-morbid depression. Consistent exclusion criteria would enable better delineation of these two conditions and their presenting symptoms. Findings across CFS subgroups suggest subjects have a different disease burden on subjective and objective measures of function, autonomic parameters and cognitive impairment when categorised using the DSQ. Different CFS criteria may at best be diagnosing a spectrum of disease severities and at worst different CFS phenotypes or even different diseases. This complicates research and disease management and may contribute to the significant stigma associated with the condition.

Fulltext free on Plos One: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0186885

 

[Murph]

The tl;dr seems to be that yes, there are differences in cfs patients but no, we don't know if that's just different severities or different diseases.

 

[alkt]

another waste of space study to find out what patients have been telling them for years. have they no concept of the word fluctuating since all results would change from time to time with a large number of patients.

 

[Jesse2233]

another waste of space study to find out what patients have been telling them for years. have they no concept of the word fluctuating since all results would change from time to time with a large number of patients.

True though I suppose it's good to have it codified as a study in writing

 

[alex3619]

True though I suppose it's good to have it codified as a study in writing

In the modern trend toward tickbox Evidence Based Medicine if its not in a study it will be ignored. Some patients said? Irrelevant. Twenty doctors said? Irrelevant. A large patient survey showed an issue? Irrelevant.

Now the issues within EBM also mean that even a study like this can be ignored, though that would depend on how the review were being done.

 

[boolybooly]

So this shows DSQ (DePaul Symptom Questionnaire) can distinguish subgroups with statistically significant differences in autonomic function. Sounds useful.

So this I guess is a recommendation to use DSQ as subgrouping criteria.

It may be stating the obvious, but it really needs to be stated since this is the elephant in the room. Characterising subtypes reliably is an essential step, without which all the other studies come adrift.

I get the distinct impression that experimentalists have long been working with insufficiently specific criteria. It stands to reason there needs to be a standard and one which can discriminate subtypes.

It may be DSQ needs developing further but this sounds like they feel they are making progress. In the interests of impartiality though it ought to be replicated by people who are independant of the institutions supporting these investigators.

IMHO if it works use it but I hope it will be a stepping stone on the road to molecular subtyping which I feel would be intrinsically more objective.

 

[A.B.]

So this shows DSQ (DePaul Symptom Questionnaire) can distinguish subgroups with statistically significant differences in autonomic function. Sounds useful.

That's questionable. Let's look at the sample sizes of the different groups:

Controls 10
Fukuda only 6
Fukuda + 2003 clinical 9
Fukuda + 2003 research 8
Fukuda + 2003 + 2011 26

I did not count the variables measured but they look like several dozens. All the between-group comparisons provide plenty of opportunities for false positives. This study is woefully underpowered. Even if there were genuine differences you'd need some way to control for severity to make sure that it's the different symptom profile, and not severity (or something else), that makes the difference.

 

[Mithriel]

I was really disappointed in this paper considering who wrote it. Unless I missed something, they tested patients lying then standing and found that they were not significantly different from sedentary controls. They speculate this could be because it is depression that causes the difference and these patients were not depressed!

BUT we have known since the 80s that patients should be tested after exertion, that tests are normal as long as they are rested. Just standing is not enough stress to give a definitive result that reflects real life. Our responses are highly dependent upon exertion.

Decades ago doctors were advised to take readings from patients after having them climb a few sets of stairs.

 

[RogerBlack]

BUT we have known since the 80s that patients should be tested after exertion, that tests are normal as long as they are rested. Just standing is not enough stress to give a definitive result that reflects real life. Our responses are highly dependent upon exertion.

For very severe ME/CFS patients, standing, or 'standard tests' may be enough to provoke PEM - but they can push through it for long enough to complete tests - but the PEM is not of course then captured.

 

[RogerBlack]

It is possible to register at PLOS and make comments.
As always, evidence based, ideally citing papers showing methodological problems with the study - for example results pre and post PEM.
Or of course which bits of the study you found valuable and want to draw readers attention to.