2 Day CPET discussion - is it evidence that GET is harmful, and is it a biomarker?

But isn't that a bit problematic?

From recent https://www.frontiersin.org/articles/10.3389/fped.2018.00242/full

That's why I'm a bit hesitant to undergo this test, because if I get a 10% reduction on day 2 someone can always dismiss it as non significant..

 

Try not to confuse increases in healthy controls due to increased familiarity with the test, versus a decrease in patients. I agree that statement is a little misleading and may be taken out of context by insurance lawyers and doctors who don't actually understand the science and don't read the references.

Keller 2014 stated this:

I was referring to the work rate at the ventilatory threshold, not the VO2@VAT, which is often not decreased in patients on the second day.

Anyway, lets actually look at the cited studies:
Studies [21] and [22] (from Stevens 2018) which recruited heart disease patients, the first with long term followups (3 to 22 months) and the second which repeated after somewhere between 24-48 hours. The later study found the mean VO2@VAT to be slightly higher (~2.5%) on the second test and did not report the wattage.

[5] and [6] were studies comparing CFS patients to healthy controls.
In healthy controls, [5] found a ~2.5% increase of the mean VO2@VAT, wattage not reported on the second test. The 36% VO2 drop at the VT in patients seems much larger than other studies since.
In healthy controls, [6] found a ~8% increase of the mean VO2@VAT and a 12% increase in workrate (wattage) on the second test. Patients had a 7% drop in workrate and a 7% drop of VO2@VAT.

[18] recruited male runners with at least 1 week separating the tests and found a ~3% drop in VO2 at the lactate threshold on the second test, but this was not 'statistically significant' and there were increases of oxygen consumption at blood lactate concentrations of 2.5 nm. No workrate reported. Note that the ventilatory threshold and the lactate threshold are related, but not exactly the same thing - the ventilatory threshold tends to lag a little bit.

[28] recruited healthy males, many of which were athletic or in the armed forces. The test was repeated after 1 week. They found a 1.5% increase in work rate and no change in VO2@VAT.

In Hodges 2018 (not cited by Stevens 2018) which compared ME CFS patients to MS and healthy controls:

In healthy controls, there was a 20% increase in VO2@VAT and an 11% increase in workrate on the second day. In MS patients there was a 7% decrease in VO2@VAT, but a ~4% increase in workrate on the second day. CFS patients had a ~6% increase in VO2@VAT, but an 11% drop in workrate on the second day (I made an error in previous post stating 7%).

So in conclusion, a drop in performance (workrate) around the ventilatory threshold is the central finding in patients, compared to either similar or increased numbers in controls. The increases over 1 week or less between tests are basically a familiarity effect in controls which leads to slightly more efficient performance. The athletic cyclists that I know that have done multiple CPETs over time tend to have very similar performance at the ventilatory threshold (if their level of fitness is similar), due to familiarity with the test and lots of experience riding bikes efficiently.

The reduction in efficiency in patients is notable and based on my own personal experience, the legs feel physically weaker and cannot put out the same level of power as the day before. This contrasts to my pre-CFS (I was an adolescent then) experience where even when I went on long (50km) bike rides or 10 km runs and although my legs felt very stiff and sore on the next day, I could still put out the same bursts of power on the bike.

 

I think you have to exceed the anerobic threshold for some time on the first day though, I agree you can stop a bit past the threshold on the second day though.

As far as a biomarker goes, it is uncertain whether it is a nonspecific marker for severe fatigue or chronic fatigue syndromes in general (including fatigue associated with other conditions - MS might be a special case where there are minimal peripheral fatigue issues).

 

There was a New Zealand study a year or so back that showed MS do not have our CPET marker, and anecdotally its unknown in other diseases, but there is no formal proof of that.

The finding is a biomarker in my view, but we need to keep in mind that a biomarker is not necessarily diagnostic or useful, and additional research is needed to demonstrate those.

 

Here is the study.

 

I mentioned that study three posts before yours, and it was the reason why I put a special note about MS.

 

In a way this is like Russian roulette. There may only be one cartridge in the gun (there may be none!), but the consequences are terrible.

There are enough concerns expressed about GET, and enough anecdotal evidence, to say that the potential is there for GET to create significant, long-term harm. That puts the onus on people supporting the use of GET to show that such fears are unfounded.

We don't have to prove GET is unsafe: we are able to show that such fears exist, that such claims are made by a substantial number of people, and that existing tests show the potential for it to be true. That shifts the responsibility.