New Dr Snell paper on exercise and CFS

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
After watching Dr Snell's presentation at the FDA stakeholder conference, my guess is that this is that paper people have been discussing.

It seems like a very helpful result, with 95% accuracy of predicting CFS, although this was a follow-up analysis, and so perhaps it wasn't put to the test, in terms of accuracy of predicting CFS...

Results Multivariate analysis showed no significant differences between controls and CFS for Test 1. However, for Test 2 the individuals with CFS achieved significantly lower values for oxygen consumption and workload at peak exercise and at the ventilatory/anaerobic threshold. Follow-up classification analysis differentiated between groups with an overall accuracy of 95.1%.

I hope that Beth Unger will see these results.

It involved 51 CFS patients. With such good results, I hope it will lead to a larger study.
It'd be interesting to read the full paper.
 

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
Note that, in his FDA stakeholder presentation (watch at about 81.45), Dr Snell said that his follow-up study of two-day VO2 max testing (which involved larger numbers than his initial VO2 max study: 51 CFS patients vs 6 CFS patients) did not give meaningful results for VO2 max (it wasn't able to discriminate CFS patients from healthy controls very effectively). It seems that Dr Snell is moving away from VO2 max testing. I think perhaps he was referring to this new research paper, with 51 CFS patients, when discussing the failure of his VO2 max tests to discriminate CFS patients.

Instead, he stated that two-day 'anaerobic threshold work-load' (ATWL) testing was giving helpful results, and was better able to discriminate for CFS patients. This new paper also seems to be his report on ATWL testing.

I think the ATWL test looks at the efficiency of oxygen utilisation, in relation to work done. So, in other words, he looks at how much effort is put into an activity before the anaerobic threshold is reached. (I think he said he uses a cycling machine for this test.) (The anaerobic threshold is the point at which the body starts producing lactic acid as a by-product of producing energy. For athletes, this threshold is very high, as they are very efficient at using oxygen to produce energy.)

Don't quote me on any of this. I might be wrong on the details, and I need to read the full paper before I understand it properly.
 

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
Slide199.jpg


This is Dr Snell's presentation slide from the FDA conference.
There are 51 CFS patients, and 10 controls, so it looks like he's presenting the results of this new paper.
The day-one test results (test 1) are in the left hand column, and the day-two test results (test 2) are in the right hand column.
He stated that the most significant and most helpful differences between patients and controls are in the VTWL* group, and this seems to be what he is reporting on, in the abstract of the new paper.
He indicated that the VO2 max results were not helpful, in this larger study.

* VTWL = ventilatory threshold work-load, which seems to be the same as anaerobic threshold work-load (ATWL).
 

biophile

Places I'd rather be.
Messages
8,977
Likes
16,579
Here is the full abstract, a part of which I have a (non-significant?) bone to pick with:

http://ptjournal.apta.org/content/early/2013/06/26/ptj.20110368.abstract

Discriminative Validity of Metabolic and Workload Measurements to Identify Individuals With Chronic Fatigue Syndrome. Christopher R. Snell, Staci R. Stevens, Todd E. Davenport and J. Mark Van Ness.

Objectives: Reduced functional capacity and post-exertional fatigue following physical activity are hallmark symptoms of chronic fatigue syndrome (CFS) and may even qualify for biomarker status. That these symptoms are often delayed may explain the equivocal results for clinical cardiopulmonary exercise testing among individuals with CFS. Test reproducibility in healthy subjects is well documented. This may not be the case with CFS due to delayed recovery symptoms. The objectives for this study was to determine the discriminative validity of objective measurements obtained during CPET to distinguish individuals with CFS from non-disabled sedentary individuals.

Methods: Gas exchange data, workloads and related physiological parameters were compared between 51 individuals with CFS and 10 control subjects, all females, for two maximal exercise tests separated by 24 hours.

Results: Multivariate analysis showed no significant differences between controls and CFS for Test 1. However, for Test 2 the individuals with CFS achieved significantly lower values for oxygen consumption and workload at peak exercise and at the ventilatory/anaerobic threshold. Follow-up classification analysis differentiated between groups with an overall accuracy of 95.1%.

Conclusions: The lack of any significant differences between groups for the first exercise test would appear to support a deconditioning hypothesis for CFS symptoms. However, results from the second test indicate the presence of a CFS related post-exertional fatigue. It might be concluded that a single exercise test is insufficient to reliably demonstrate functional impairment in individuals with CFS. A second test may be necessary to document the atypical recovery response and protracted fatigue possibly unique to CFS, which can severely limit productivity in the home and workplace.
Concerning this statement: "The lack of any significant differences between groups for the first exercise test would appear to support a deconditioning hypothesis for CFS symptoms." I have not read the full-text yet, and I suspect this was just said in passing as a prelude to then contradicting the deconditioning hypothesis, but I do not understand how the lack of any significant differences between the two groups, on objective measurements (1st test), allegedly supports the deconditioning hypothesis? If anything it refutes it, how can CFS symptoms be attributed to deconditioning if non-disabled sedentary individuals with similar levels of fitness do not report such symptoms either before or after exercise?
 

Simon

Senior Member
Messages
3,788
Likes
14,822
Location
Monmouth, UK
Agreed biophile, don't see how lack of difference on test 1 supports decon hypothesis - looks like it refutes it to me.
However, for Test 2 the individuals with CFS achieved significantly lower values for oxygen consumption and workload at peak exercise and at the ventilatory/anaerobic threshold. Follow-up classification analysis differentiated between groups with an overall accuracy of 95.1%.
At th FDA workshop, Chris Snell said VO2 max on test 2 was NOT significantly lower, which also fits with the slide Bob posted above. It was just the Workload at Anaerobic Threshold that was significant. So I'm a bit confused by the abstract (best practice is that these report actual data for primary outcomes).

Will try to access the full text - it certainly looks interesting.
 

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
Concerning this statement: "The lack of any significant differences between groups for the first exercise test would appear to support a deconditioning hypothesis for CFS symptoms." I have not read the full-text yet, and I suspect this was just said in passing as a prelude to then contradicting the deconditioning hypothesis, but I do not understand how the lack of any significant differences between the two groups, on objective measurements (1st test), allegedly supports the deconditioning hypothesis? If anything it refutes it, how can CFS symptoms be attributed to deconditioning if non-disabled sedentary individuals with similar levels of fitness do not report such symptoms either before or after exercise?
Agreed biophile, don't see how lack of difference on test 1 supports decon hypothesis - looks like it refutes it to me.
I noticed that as well.
I agree with you both that it appears to refute the deconditioning theory.

I'm certain, as biophile suggests, that Dr Snell uses this as a prelude to contradicting the deconditioning hypothesis.
He absolutely only had negative things to say about the PACE Trial in his FDA presentation.
 

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
At the FDA workshop, Chris Snell said VO2 max on test 2 was NOT significantly lower, which also fits with the slide Bob posted above. It was just the Workload at Anaerobic Threshold that was significant. So I'm a bit confused by the abstract (best practice is that these report actual data for primary outcomes).
If you are referring to the word 'peak', in the abstract, perhaps it refers to "work-load peak", rather than VO2 peak? (see the graph that I posted above for details.)
(I don't know how 'work-load peak' differs from VO2 max and ATWL/VTWL.)
 

MeSci

ME/CFS since 1995; activity level 6?
Messages
8,215
Likes
13,134
Location
Cornwall, UK
Here is the full abstract, a part of which I have a (non-significant?) bone to pick with:

Concerning this statement: "The lack of any significant differences between groups for the first exercise test would appear to support a deconditioning hypothesis for CFS symptoms." I have not read the full-text yet, and I suspect this was just said in passing as a prelude to then contradicting the deconditioning hypothesis, but I do not understand how the lack of any significant differences between the two groups, on objective measurements (1st test), allegedly supports the deconditioning hypothesis? If anything it refutes it, how can CFS symptoms be attributed to deconditioning if non-disabled sedentary individuals with similar levels of fitness do not report such symptoms either before or after exercise?
Maybe it means that the 'non-disabled sedentary individuals' (couch potatoes?) were (also) deconditioned through inactivity, and that deconditioning is not viewed as disability?
 

biophile

Places I'd rather be.
Messages
8,977
Likes
16,579
Maybe it means that the 'non-disabled sedentary individuals' (couch potatoes?) were (also) deconditioned through inactivity, and that deconditioning is not viewed as disability?
Perhaps, but the couch-potatoes apparently aren't experiencing debilitating CFS symptoms at rest or the equivalent prolonged post-exertional symptoms for several days after exercise. I see no mention of subjective measurements in the abstract, but it would be safe to say there would be significant differences in those too between groups.
 

Simon

Senior Member
Messages
3,788
Likes
14,822
Location
Monmouth, UK
If you are referring to the word 'peak', in the abstract, perhaps it refers to "work-load peak", rather than VO2 peak? (see the graph that I posted above for details.)
(I don't know how 'work-load peak' differs from VO2 max and ATWL/VTWL.)
"Oxygen consumption AND workload at PEAK exercise"
Oxygen consumption at peak exercise is VO2max. Sorry about the CAPS, not shouting, but no editor controls when using a tablet :)
 

Bob

Senior Member
Messages
16,455
Likes
34,051
Location
England (south coast)
"Oxygen consumption AND workload at PEAK exercise"
Oxygen consumption at peak exercise is VO2max. Sorry about the CAPS, not shouting, but no editor controls when using a tablet :)
But, like you, I'd be surprised if he is referring to VO2 max, if his FDA presentation is correct.
Do you happen to know what 'work-load peak' is, and how it differs to ATWL?
 

Simon

Senior Member
Messages
3,788
Likes
14,822
Location
Monmouth, UK
But, like you, I wouldn't have thought that he can be referring to VO2 max, if his FDA presentation is correct.
Do you know what 'work-load peak' is?
That's why it doesn't appear to make sense

Peak workload is the maximum amount of work you apply during the session, usually shortly before collapsing... Usually it's measured in watts. On a bike its a combination of how fast you are cycling, and how step the slope (resistance) and on a treadmill it's how fast and at what slope.

For comparison, a Tour de France cyclist sprinting to win can peak at about 750w, compared with 140W for the sedentary controls above - and that's at the end of a 100 mile race.
 

SOC

Senior Member
Messages
7,849
Likes
16,372
Perhaps, but the couch-potatoes apparently aren't experiencing debilitating CFS symptoms at rest or the equivalent prolonged post-exertional symptoms for several days after exercise. I see no mention of subjective measurements in the abstract, but it would be safe to say there would be significant differences in those too between groups.
I suspect we're just picking at a less than ideally worded sentence in the conclusion of an abstract. Admittedly, that's a really bad place to have a poorly worded sentence, but there you have it. Sometimes the writing isn't as clean and sharp as we'd like.

It's also possible that a more strongly worded conclusion got bashed by a (biased? timid?) reviewer who wanted the conclusion statement softened. I see hints to that also in these phrases (my bolding):
It might be concluded that a single exercise test is insufficient to reliably demonstrate functional impairment in individuals with CFS. A second test may be necessary to document the atypical recovery response and protracted fatigue possibly unique to CFS, which can severely limit productivity in the home and workplace.
Notice how the that first sentence has a more definite phrase "would appear to support" where we are thinking a less definitive phrase like "would initially suggest" fits the data better. At the same time, conclusions that seem more supported by the data as we currently understand it ("a single exercise test is insufficient" and "a second test is necessary to document") are softened with "might" and "may be".

Sometimes you have to give into a jerk in a small thing in order to get the paper published. :rolleyes:

As long as the data in the paper shows a clear difference between sedentary controls and patients and has sufficient explanation of the procedure for replication/confirmation studies to be done, I'm happy.
 

alex3619

Senior Member
Messages
13,810
Likes
37,668
Location
Logan, Queensland, Australia
Its also worth noting the slide does show a lower VO2max for CFS than controls on day one. Its just that its not a lot lower on day one. In other words, there is a problem but its less than obvious. The ventilatory workload on day two however shows a major crash. As a marker that would be much superior. That crash is so big its actually scary, but do spare a thought these results are for mild or moderate patients, probably not severe and definitely not very severe.

My guess about VO2max, without seeing a data plot, is that patients and controls would overlap, which is why its not that usefull. Its an indicator but not strong enough to be diagnostic.

I was unable to get this paper yesterday due to the paywall.
 

Firestormm

Senior Member
Messages
5,055
Likes
4,817
Location
Cornwall England
What's with only having the 10 controls and with them all being female? I'd also like to hear (as ever) what happened to all the subjects prior to the test: how activities prior to might have contributed to any results. It is important to compare like with like as much as possible. Ideally, I would prefer all subject to be on-site for a week of 'conditioning' prior to the actual testing - or testing across with week - but under observation at all times. Just a little 'bugbear' of mine :)
 

alex3619

Senior Member
Messages
13,810
Likes
37,668
Location
Logan, Queensland, Australia
I would also prefer they were booked into a controlled facility a day or two earlier - and kept there for a bit if they need time to recover. However we always have to keep in mind how tight funding is.

The all female issue is again driven by cost. Add in males and issues can arise with statistical significance, leading to a bigger study at higher cost. I hope future studies will do that though.

The other thing is they don't need controls, not really - thats more for scientific completeness. This is a standard test used by sports medicine for a very long time, well understood, with well known results. They could probably just look up the hundreds they and others have already tested. What the controls would do though is show if there is a problem should their results be very different from what was expected from controls.
 

Firestormm

Senior Member
Messages
5,055
Likes
4,817
Location
Cornwall England
Agreed. If PEM is for some the defining symptom for 'ME' then I think more resource should be put into discovering what it is and quantifying it with the goal of producing a cost-effective test that can be used fairly. If PEM 'comes and goes' then this needs to be accounted for - and if it is 'ever-present' this too needs to be confirmed.

I want matched-controls as a standard. I want anything to do with PEM especially, to be 'cast-iron' before I personally hang my hat on it. I think we are heading in this direction.

Funnily enough, I had an email alert earlier, pro-health featuring a letter from Mia Farrow about Post polio Syndrome: talked about PEM too.

Maybe PEM exists - in some way - in many diseases. Maybe it can be clinically defined more appropriately for each of those diseases; or maybe it is a result of the same physical process.

Questions, questions, questions... Someone could write an article ;) Whatever the case though it clearly needs to be better defined and quantified for ME.

Interesting paper :) So what test is now 'in' and what is 'out'?
 

Firestormm

Senior Member
Messages
5,055
Likes
4,817
Location
Cornwall England
alex3619 Dear chap, I was pondering another aspect that sometimes arises when answering or basing results upon questionnaires but could perhaps be applied to exercise tests: motivation.

I wonder to what extent - could be negligible - motivation is a factor in exercise testing of people with any chronic condition affecting functionality; but perhaps more so in connection to ME?

You or anyone happen to have any thoughts? I mean as a patient, if I suspected the 'pay-back' for me was to be dire - I might not try as hard as I could on the bike; then again as it was research I might 'push through' on the earlier testing.

How might they account for this? Of course a week-long study involving observation, acclimatisation, and matching of similar patient cohorts to controls might help. I mean such a study of 'moderate' patients could affect outcome in general compared to 'severe' (should that even be possible); might also be a greater factor if using 'mild' and 'moderate'...

But overcoming 'motivation' critiques is something I would like to understand better - or rather how in science they try to account for this factor.

Adding MS to the control group would be a good move generally I think. If you get the full paper could you let me know more about the controls and the patients? Thanks.
 

Valentijn

Senior Member
Messages
15,786
Likes
45,658
I was pondering another aspect that sometimes arises when answering or basing results upon questionnaires but could perhaps be applied to exercise tests: motivation..
I think that some of the objective results (oxygen uptake?) are completely biological results, and lower levels can't be reflective of a lack of effort. And there's also blood tests of proteins and such, showing objective abnormal results for ME/CFS patients, presumably compared to controls doing a similar amount of exertion.