Discussion in 'Phoenix Rising Articles' started by Mark, Jun 28, 2013.
View the Post on the Blog
View the Post on the Blog
Nice review Simon.
Methodological issues aside (and thanks for keeping us all grounded in this respect) I do wonder how many significant findings may be obscured by simple group comparisons where the patient cohort may be heterogenous.
"However, the STOPP group pain increase doesn’t appear to be significantly different from START, and I suspect the label was chosen mainly because it makes a snappy acronym"
A particular pet hate of mine (PACE, FINE, NICE, SMILE, BACME etc) and particularly naff in this context I felt.
I'm sure heterogeneity is a big reason for the lack of progress. Looking for subgroups is a good way of addressing this in smaller studies, so long as there is proper follow-up on any findings. Ultimately the answer is much bigger studies, which we are beginning to see, such as Mady Hornig's huge pathogen study and the CDC multi-clinic study which specifically aims to uncover heterogeneity.
Thanks Simon. Struggled with this one (well nothing new there!). So they did the original experiment, found nothing 'odd', and THEN drilled down into the data to find these 'sub-groups'.
Hmm... I see what you meant the other day. Still, if the next experiment is better designed with these 'sub-groups' in mind then replication might have a better chance of succeeding.
FWIW that cognition test? Forget it. I couldn't comprehend the rules and the test was a nightmare
That's exactly how it appears, and hopefully there will be replication that builds on this study.
Sorry the test was so hard - I only included it because I thought it migh be fun...
Another intriguing article - thanks, Simon. I won't even attempt the test at the moment - brain too fuzzy due to congestion. I'm sure that results will vary a lot depending on whether or not we have brain fog at the time!
All a question of timing. Am sure it's doable. Its got me hooked now - so I shall be back to try it again
I'm ignorant of science methodology unfortunately, so i'm thankful for others more informed and balanced judgement.
I would have thought the American government would have a huge desire to follow up and replicate such experiments for the benefit of their military...and that me/cfs patients would benefit downstream.
But upon further reading it seems that this is not necessarily the case.
Cort, over on HR writes,
'It’s not clear how many of this studies findings will apply to ME/CFS but one thing’s for sure – GWI patients have received the same, if not worse kind of neglect from the federal establishment that the ME/CFS community has endured. Six months ago Dr. Stephen Coughlin, an immunologist at the Veteran Affairs Office, resigned while accusing the VA of manipulating and hiding data and basically doing everything it could to ensure Gulf War Illness was not taken seriously. Coughlin asserted the VA did not want anything published that might prove GWI is a neurological condition.'
It's 23 years since the start of the Gulf war! Progress is certainly slow.
Similar kind of neglect, fair enough. But worse?? I've heard several researchers - including Baraniuk, Klimas and Mikovits I think, from memory - say that they find it much easier to get funding for their GWI research than for their ME/CFS research. Rayhan said at Invest in ME that they would love to repeat their GWI study for ME/CFS patients, but expected to find it much harder to get funding for that. I even get the impression that some researchers interested in ME/CFS are doing GWI studies instead in the hope that what they find will be applicable to ME/CFS as well. So my impression would be that GWI has been neglected 'almost as badly' as ME/CFS rather than 'the same, if not worse'. Comparison of historical research funding stats would be interesting.
I've updated the main blog to clarify some of the differences in brain activation during that cognitive test. For the STOPP (Phantom Pain Perception) subgroup, the areas activated in addition to the ones expected for woking memory both have a role in pain, which is another reason the authors chose the pain label. However, pain was not significantly different between START and STOPP groups, and those brain regions are involved in other activity apart from pain perception.
Here's an update too on the future from Rakib Rayhan:
I pulled this from the net- Sorry if it doesn't link. I just googled 'cost of gulf war'.
Costs of Major U.S. Wars
Congressional Research Service
A Trillion Dollars for Wars Since 9/11
Since the terrorist attacks of September 11, 2001, Congress has appropriated more than a trillion
dollars for military operations in Afghanistan, Iraq, and elsewhere around the world. The House
and Senate are now considering an additional request for $33 billion in supplemental funding for
the remainder of FY2010, and the Administration has also requested $159 billion to cover costs of
overseas operations in FY2011.
Holy crap! A trillion dollars! That's $1,000,000,000,000.... Or, a thousand billion dollars. Or a million million if you like.
And that's only up to 2010.
I don't know what the situation in USA is, either with GWI research or me/cfs research.
It seems like there's nothing the USA military can't do when undertaking war. I would be interested to know if they back a real desire to solve GWI with sufficient funding?
I know that funding money doesn't always equate to results, and that the researchers themselves are doing their utmost to solve the problem.
Governments always seem to be more willing to invest in taking lives than in saving them.
Whoops - controversial...
Thanks for this, Simon!
"But they didn’t find the expected drop off in performance on the exercise bike at the second test (this fits with the latest results for CFS patients presented by Christopher Snell at the FDA workshop)."
...made me a bit worried. Did Chris Snell report at the FDA workshop that his team does NOT find the drop on Day 2 of the Stevens Protocol Exercise Test in ME/CFS patients??
I have so much been hoping that the second day drop in oxygen use and anaerobic threshold would hold up as biomarkers to distinguish ME/CFS patients and to possibly measure outcome in future treatment trials. I just saw the newly published paper by Snell/Stevens/et al:
And I read this at the Prohealth report on the FDA workshop:
"Test results for CFS/ME patients
Dr. Snell’s group was not able to distinguish CFS patients from controls during an exercise test, until they looked at the aftermath. People with CFS were sick after the test. Strikingly, their ability to utilize oxygen decreased on the second day of testing. This was in direct contrast to healthy controls, who were able to utilize oxygen better on the second day. Later, this study was replicated using 56 patients. While the results were not as dramatic, the second larger study showed that there was a big drop-off in metabolic efficiency. CFS patients produced less work for more effort and had a greater buildup in lactate.
Post exercise morbidity distinguishes deconditioning from CFS. CPET is a valid and accepted form of testing - there are just over 400 clinical trials around the world that currently include CPET (clinicaltrials.gov). CPET should be used as an endpoint to evaluate the efficacy of any drug treatment in clinical trials."
Was there something Snell said at the FDA workshop that indicated that the Stevens Protocol does NOT work in distinguishing ME/CFS patients (or hopefully measuring treatment efficacy)?
Thanks for keeping us up to date on everything!
It's a bit ambiguous at the moment. The data presented at the FDA workshop, and what Chris Snell said too, indicated that the big drop-off seen in max oxygen consumption in the very small pilot wasn't replicated. But they did find a substantial drop in efficiency on day 2, so the CFS patients were producing less 'power' on the bicycle on day 2, despite using similar amounts of oxyge However, the newly published study, which appears to use the same data, says peak oxygen consumption also fell on Day 2 (ie the same finding as the small pilot).
So the good news is that the new paper definitely distinguishes between CFS and controls. However, it seems to do this in a different way to the original pilot study that got so much attention.
Hopefully things will become clearer when we get to see the full text of the new paper.
Separately, fMRI studies by Julia Newton's groups indicates muscle metabolism abnormality on muscle flexing alone. My view is that whatever the results of the latest Snell paper, we don'thave a definitive exercise test until we have consistent replication of any finding.
There are multiple physiological measures in exercise testing to be considered. What the people doing exercise testing are currently sorting out is which of those measures are clearly and predictably different in ME/CFS patients -- which would be beneficial for diagnostic purposes -- as well as which others might be useful for other purposes such as investigating causality, or evaluating treatment.
For example, a diagnostic marker has to be distinct enough that there isn't an overlap in the measure between healthy people and ill ones. For a diagnostic marker to be useful, every ill individual needs to fall outside the healthy range on that measure.
OTOH, discovering that the average (on some measure) of the patient group is significantly different from the average of healthy controls tells us that something is indeed different about the patient group, and what areas (relating to that measure) should be investigated for treatments and/or causality.
A current example of a useful measure that is not diagnostic is NK cell function. Every ME/CFS patient (as far as we currently understand it) does not have NK cell function outside the normal range, so it is not currently considered diagnostic. However, many of us do have low NK cell function, which indicates there's something wrong in that arena. Also, if a patient does have low NK cell function, treatments for poor NK cell function can be given and their effectiveness tracked by measuring change in NK cell function.
One thing we're hoping for, but which no one has published about afaik, is what measure will be useful in evaluating the effectiveness of any fundamental treatment for ME/CFS. For example, my AT is low which is common in ME/CFS, but not universal enough to be a diagnostic. Still, if a treatment increased my AT, that would be really useful for me. However, symptomatic treatments that have moved me from bedbound to 60-70% functional and feeling hugely better (no pain, no flu-like symptoms) have not budged my AT one iota. So if we used AT as a measure of the effectiveness of a treatment, the treatments that have helped me a lot would be considered ineffective because they didn't change my AT.
Lots more work to be done. Thanks to Dr Snell and any others who are trying to tease apart this knotted ball of confusion that is ME/CFS.
I think that there are few diagnostic markers for any condition that are 100% specific and 100% sensitive, so something in the 90s would be great. But of course there are probably subgroups, and co-morbidities (which will be more numerous the later the diagnosis is attempted), and there is always the inescapable circularity/tautology of having to rely on incomplete knowledge and having to make assumptions, which themselves may not be correct!
This is not unique to ME, and we will never have a perfect solution.
It keeps scientists in work at least.
Agree with that,esp the circularity of not having a validated gold standard.
90%+ is pretty good, but with one caveat:
Doctors don't usually have problems distinguishing between ME/CFS patients and healthy sedentary controls so the issue is specificity. Would be great to have a test that reliably distinguishes between ME/CFS and and other chronically fatiguing illnesses.
It is important to note that these studies don't imply any directionality. The brain changes may simply reflect the changes in the body, since we are talking about two parts of a single system.
Similarly exciting changes in brain activity can be measured before and after a thirsty person has a drink.
The problem I have with these brain imaging studies is they tell us very little about potential treatments.
I'm a bit puzzled by this. Measuring differences in brain activity during a congitive test, compared with baseline, seems a reasonable way to identify brain areas that might be involved in that test. Or have I missed something?
My bigger concern is whether the findings are real and replicable.
Isn't that what would be expected? Given the fundamental importance of water to a thirsty person, I would have predicted exciting changes in brain activity in response to a drink.
The authors are arguing that their fMRI test separates the patients into meaningful subgroups, which will aid research:
I'm disappointed to see here what seems like Snell waffling-- or, even more disappointing, what seems like Snell trying to confuse the issue.
I think the two-day exercise test is so important because it demonstrates that THERE EXIST people whose VO2 max drops on the second day. Because of this, you could consider PEM to be a clinical entity in itself. Everything I read up until this (Snell waffling) seems consistent with that.
--PEM defined that way might be unique to people who seem to have something wrong with them but do not have another disease label other than Fukuda etc. Thus: a new clinical entity.
--PEM defined that way might turn out to be a symptom shared by some/all CFS people with (say) MS and/or cardiomyopathy and/or ALS or who knows. This would be evidence for something severely wrong with CFS people and would suggest what it might be.
--PEM defined that way is evidence that people with PEM will be made worse or killed by exercise. That is most definitely a treatment suggestion.
It would work best for our purposes if someone simply divided people into two groups-- those whose VO2 max falls on second day and those who don't, and then study the first group.
Of course this happens in those lab(s) that test a large number of private individuals for disability. What is happening to that evidence?
Simon, would it be possible for you to write a summary from this perspective? (I am completely confused by the different labs, different studies, different buzz words, etc. even though there seems to be a very limited number of labs/studies.) Thanks for your analysis so far. I'm terrified that something apparently so clear-cut is going to get obfuscated out of existence, and that it is already too complex for the casual reader.
You can also try a Google Site Search
Separate names with a comma.