Two-Day Cardiopulmonary Exercise Testing in Long COVID Post-Exertional Malaise Diagnosis, 2024, Gattoni

Chiara Gattoni, Asghar Abbasi, Carrie Ferguson, Charles W. Lanks, Thomas W. Decato, Harry B. Rossiter, Richard Casaburi, William W. Stringer,
Two-Day Cardiopulmonary Exercise Testing in Long COVID Post-Exertional Malaise Diagnosis,
Respiratory Physiology & Neurobiology,
2024,
104362,
ISSN 1569-9048,

https://www.sciencedirect.com/science/article/pii/S1569904824001551?via=ihub

Respiratory Physiology & Neurobiology
Available online 28 October 2024, 104362

Two-Day Cardiopulmonary Exercise Testing in Long COVID Post-Exertional Malaise Diagnosis
https://doi.org/10.1016/j.resp.2024.104362
open access
Highlights

• •
  No physiologic differences between day one and two CPET in Long COVID.
• •
  Questionnaire identification of PEM did not associate with 2-day CPET.
• •
  PEM was not identified despite repeated exercise testing.

Abstract

BACKGROUND

Long COVID patients present with a myriad of symptoms that can include fatigue, exercise intolerance and post exertional malaise (PEM). Long COVID has been compared to other post viral syndromes, including myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), where a reduction in day 2 cardiopulmonary exercise test (CPET) performance of a two-day CPET protocol is suggested to be a result of PEM. We investigated cardiopulmonary and perceptual responses to a two-day CPET protocol in Long COVID patients.

METHODS

15 Long COVID patients [n=7 females; mean (SD) age: 53(11) yr; BMI = 32.2(8.5) kg/m2] performed a pulmonary function test and two ramp-incremental CPETs separated by 24 hr. CPET variables included gas exchange threshold (GET), V̇O2peak and WRpeak. Ratings of perceived dyspnoea and leg effort were recorded at peak exercise using the modified 0-10 Borg Scale. PEM (past six months) was assessed using the modified DePaul Symptom Questionnaire (mDSQ). One-sample t-tests were used to test significance of mean difference between days (p<0.05).

RESULTS
mDSQ revealed PEM in 80% of patients. Lung function was normal. Responses to day 1 CPET were consistent with the presence of aerobic deconditioning in 40% of patients (V̇O2peak <80% predicted, in the absence of evidence of cardiovascular and pulmonary limitations). There were no differences between day-1 and day-2 CPET responses (all p>0.05).

CONCLUSION
Post exertional malaise symptoms in Long COVID patients, in the absence of differences in two-day CPET responses separated by 24 hours, suggests that post-exertional malaise is not due to impaired recovery of exercise capacity between days.

Key words
cardiopulmonary exercise testing
PEM
exercise tolerance
long COVID
fatigue
deconditioning

 

The sample size (n = 15) was probably too low.

 

Female 47%
BMI 32.2 ± 8.5 kg/m2

VE/VCO2 32.1 ± 3.7 -> 31.3 ± 3.1 (p = 0.008)

 

VO2_peak (19.43 on day 1 versus 19.67 on day 2) and workload_peak (36 on day 1 versus 36 on day 2) did not change at all.

At the gas exchange threshold (GET), VO2 declined from 1.09 to 1.06 (in absolute numbers) corresponding to a cohen's d of 0.318. Workload at GET does not seem to be reported.

 

I wonder why. They tested just about everything else, including work at another timepoint. And decrease in work at AT is one of the most consistent findings in ME/CFS, so it seems strange to not include that.

 

I think 15 is enough to show a difference of clinical importance.
At least it looks as if it is perfectly possible to have PEM as a symptom without finding a major dip at day 2. That doesn't worry me particularly. I think the link between the two would have to be indirect anyway.

 

It would've been nice if they used ME/CFS patients as the control. It says ".. a reduction in day 2 was suggested to be the result of PEM", so we don't know if the result is a verdict on LC PEM or 2-day CPET.

 

Or perhaps this shows that the mDSQ is poor at defining PEM.

1. Dead, heavy feeling after starting to exercise.
2. Next day soreness or fatigue after non-strenuous, everyday activities.
3. Mentally tired after the slightest effort.
4. Minimum exercise makes you physically tired.
5. Physically drained or sick after mild activity.
6. If you were to become exhausted after actively participating in extracurricular activities, sports, or outings with friends, would you recover within an hour or two after the activity ended?
   
7. Do you experience a worsening of your fatigue/energy related illness after engaging in minimal physical effort?
   
8. Do you experience a worsening of your fatigue/energy related illness after engaging in mental effort?
   
9. If you feel worse after activities, how long does this last?
   
10. If you do not exercise, is it because exercise makes your symptoms worse?

Take a look at table 2 at https://www.sciencedirect.com/science/article/pii/S1569904824001551?via=ihub#tbl0010

I don't see 80% of this sample experiencing PEM in anyway that I understand the term.

 

The mDSQ seems to be a tool to detect exercise intolerance, not PEM.

The delayed worsening, aggravation/appearance of symptoms not typically aggravated by activity, and impairment of ability to do things are the key feature of PEM.

For example, being active over the day or exercising doesn't normally disrupt your sleep.

Sore muscles and low energy on the next day can be caused by exercise, but if you feel sick (or like shit) and can't get anything done due to cognitive impairment the day after then it's PEM.

 

Seems like you're right.

When looking into this, I tried to calculate the minimum effect size d (mean_difference/sd_baseline) that could be statistically significant (p = 0.05) given the sample size n = 15. To get an easy approximation, I assumed sd_baseline = sd_post = 1 and the sd of the difference = sqrt(sd1^2 + sd2^2 - 2 * sd1 * sd2 * r).

Here are the results for various values of the correlation r between baseline and post-test values. If the correlation is large (r > 0.5) a sample size of 15 can indeed pick up on moderate effects. In the Keller 2024 study, the VO2_max values before and after exercise test had a very high correlation of 0.87, which means it would be able to identify small effects as well.

 

I contacted one of the authors, William Stringer, to ask about workload at AT, and he very kindly provided the data for that timepoint and allowed me to share it here.



The mean did decrease, but just by a little bit. First column is day 1, second column is day 2.



Here's a plot I made of the workloads for the two days for each participant:



It does not appear to be a significant difference:



----

I think these are the same participants as this paper:

A Pilot Study on the Effects of Exercise Training on Cardiorespiratory Performance, Quality of Life, and Immunologic Variables in [LC], 2024, Abbasi+

So they're doing lots of CPETs. Two in that study, and two in this one.