Perception of induced dyspnea in fibromyalgia and chronic fatigue syndrome (2018) Van Den Houte et al

[hixxy]

J Psychosom Res. 2018 Mar;106:49-55. doi: 10.1016/j.jpsychores.2018.01.007. Epub 2018 Jan 11.

Perception of induced dyspnea in fibromyalgia and chronic fatigue syndrome.

Van Den Houte M, Bogaerts K, Van Diest I, De Bie J, Persoons P, Van Oudenhove L, Van den Bergh O.

Abstract
OBJECTIVE:
Dyspnea perception is distorted in patients with medically unexplained dyspnea. The goals of this study were 1) to replicate these results in patients with fibromyalgia and/or chronic fatigue syndrome (CFS), and 2) to investigate predictors of distorted symptom perception within the patient group, with a focus on negative affectivity (NA), psychiatric comorbidity and somatic symptom severity.

METHODS:
Seventy-three patients diagnosed with fibromyalgia and/or CFS and 38 healthy controls (HC) completed a rebreathing paradigm, consisting of a baseline (60s of room air), a rebreathing phase (150s, gradually increasing ventilation, partial pressure of CO2 in the blood, and self-reported dyspnea), and a recovery phase (150s of room air). Dyspnea, respiratory flow and FetCO2 levels were measured continuously.

RESULTS:
Patients reported more dyspnea than HC in the recovery phase (p=0.039), but no differences between patients and HC were found in the baseline (p=0.07) or rebreathing phase (p=0.17). No significant differences between patients and HC were found in physiological reactivity. Within the patient group, the effect in the recovery phase was predicted by somatic symptom severity (p=0.046), but not by negative affectivity or by the number of psychiatric comorbidities.

CONCLUSION:
This study extended earlier findings in patients with medically unexplained dyspnea to patients with fibromyalgia and CFS. This suggests that altered symptom perception is a non-symptom-specific mechanism underlying functional somatic syndromes in general, particularly in patients with high levels of somatic symptom severity. The results are discussed in a predictive coding framework of symptom perception.

Copyright © 2018 Elsevier Inc. All rights reserved.

KEYWORDS:
Chronic fatigue syndrome; Fibromyalgia; Functional somatic syndromes; Predictive coding; Symptom perception

https://www.ncbi.nlm.nih.gov/pubmed/29455899
http://www.jpsychores.com/article/S0022-3999(17)31032-2/fulltext

 

[Hoopoe]

This suggests that altered symptom perception is a non-symptom-specific mechanism underlying functional somatic syndromes in general

No it doesn't, because there is no data showing that the observed results are specific to "functional somatic syndromes".Where is the comparison to similar chronic disabling illnesses such as heart disease or multiple sclerosis?

Even if it were a specific finding, it would still be unclear what this means, even though the authors would probably jump to the conclusion that patients attention to their body is causing symptoms.

 

[Valentijn]

Full text is at https://sci-hub.tw/https://doi.org/10.1016/j.jpsychores.2018.01.007

These are Belgian quacks, primarily from Leuven.


So the real test is how long it takes patients to recover from breathing in an elevated amount of carbon dioxide, after they return to breathing normal air without being aware of the switch:

In these experiments, participants breathed through a circuitry either connected to room air or to a bag initially filled with 5% CO2 and 95% oxygen, causing gradual increase in ventilation, partial (arterial) CO2 pressure and self-reported feelings of dyspnea (air hunger). Concealed from the participant, participants switch to room air breathing after 150 s of rebreathing allowing recovery.

They're explicitly starting with the assumption that the patients are completely physically healthy:

These studies have shown that healthy high habitual symptom reporters and patients with medically unexplained dyspnea show a reduced within-subject correspondence between induced physiological changes and perception thereof, compared to healthy controls.

Their hypothesis is that FM/CFS-crazy will look like medically unexplained dypsnea-crazy on this test:

Given the debate on the specificity of different FSS and the hypothesis that the deficit in symptom perception underlies FSS in general, the first goal of this study was to investigate whether results found in patients with medically unexplained dyspnea extend to patients with fibromyalgia and CFS.

Fukuda was used to recruit CFS patients, with exclusions based on "chronic cardiovascular, respiratory or neurological disorders" among other things. I'm not quite sure how you exclude patients with neurological disorders in a study involving a disease which is officially classified as neurological


This doesn't sound like standard practice, and it's not clear if patients and controls are being tested with the same equipment in a reasonable ratio with each other:

To correct for equipment differences, FetCO2 was defined as the relative change in FetCO2 compared to right before the rebreathing test.

It sounds like they made a lot of comparisons without making statistical corrections for the increased likelihood of a false positive:

Average dyspnea ratings, FetCO2 levels and minute ventilation were calculated for every 30 s. Multiple mixed model analyses were performed on the phases separately, with dyspnea ratings, FetCO2 levels and minute ventilation as dependent variables in separate analyses. Group (patient versus controls) and time were used as independent variables in all analyses. To control for potential baseline differences in any of the outcome variables, dyspnea rating, FetCO2 level or minute ventilation (respectively) in the last 30 s of the baseline phase was added as a covariate in the analyses for the rebreathing and recovery phase.

They lost a lot of data, primarily due to technical problems. The distribution of those losses suggests that patients and controls were not being tested with the same equipment:

Eighty-one patients and 41 HC participated in the study. Six patients and one HC stopped the rebreathing test. Data of two patients and two HC could not be analyzed due to technical problems. The final sample consisted of 73 patients (mean age: 42.37 (SD: 12.90), 10 men) and 38 HC (mean age: 40.79 (SD: 13.74), 5 men). Detailed information on marital and working status and educational level of the participants is displayed in Appendix A (Table A.1). Respiratory flow could not be calculated for three HC and eight patients, and FetCO2 could not be calculated for one HC and ten patients because of technical problems.

What a mess. I guess this answers the question of what happens when you exclude neurological disorders from a study of a neurological disease. And what was the point of the diagnostic questionnaires if they were going to be ignored? Presumably they needed the results from those "patients" to get their desired conclusion:

According to the fibromyalgia and CFS criteria questionnaire, 36/73 patients fulfilled both the CFS and fibromyalgia criteria, 2/73 fulfilled the CFS criteria alone, and 31/73 patients fulfilled the fibromyalgia criteria alone. Four patients did not meet the criteria for CFS or fibromyalgia according to these questionnaires, but had received a doctor-based diagnosis, so were not excluded from analysis. The diagnosis the patients received (or did not receive) from their doctors did not always concur with the diagnosis they fulfilled according to our checklist.

This is a pretty shitty p-value, especially since this group didn't correct for making multiple comparisons and has done these sorts of studies before with other groups, and should be able to focus on the essential outcomes by now. It's likely a false positive, and even if it weren't it would only indicate that FM/CFS patients take longer to recover - it would not rule out a physical cause:

Patients reported more dyspnea than HC in the recovery phase (p=0.039), but no differences between patients and HC were found in the baseline (p=0.07) or rebreathing phase (p =0.17).

Again, a shitty p-value. And even that is only correlated with physical symptoms, and not with mood or psychiatric disorder. Even most psychosomatic questionnaires would only consider symptoms to be psychosomatic if there is also a mood disorder, hence the lack of association between the results and the mood/psychiatric symptoms seems to disprove their hypothesis:

Within the patient group, the effect in the recovery phase was predicted by somatic symptom severity (p=0.046), but not by negative affectivity or by the number of psychiatric comorbidities.

 

[Trish]

Thanks @Valentijn for saving me the effort of trying to understand it.

Have I got this right? On the basis of a very brief test on a small sample of patients that showed a barely significant difference in perception of breathlessness between FM/CFS patients and healthy controls, they extrapolate to say this is evidence that patients with all so called functional syndromes have altered symptom perception for all symptoms? That is the most outrageous extrapolation.

 

[Valentijn]

Have I got this right? On the basis of a very brief test on a small sample of patients that showed a barely significant difference in perception of breathlessness between FM/CFS patients and healthy controls, they extrapolate to say this is evidence that patients with all so called functional syndromes have altered symptom perception for all symptoms? That is the most outrageous extrapolation.

Yes, exactly. It's also combined with some severe methodological problems which make it very unlikely that there was any statistical significance at all.