Two Is Better Than One

My confidence in two-day cardiopulmonary exercise testing (CPET) is pretty obvious on this blog. A new study from ME/CFS experts Dr. Chris Snell, Staci Stevens, Dr. Todd Davenport, and Dr. Mark VanNess supplies hard data that shows how important a two-day maximal CPET is for diagnosis and documenting ME/CFS. The Physical Therapy Journal has made the author manuscript available (behind a paywall), and so that version I review here is not necessarily the final version that will be published by the journal.

The purpose of this study was to establish whether the objective measurements in CPET could distinguish patients with CFS from healthy controls. The study enrolled 51 female CFS patients and 10 female sedentary controls. The CFS patients were diagnosed using the Fukuda criteria, and they also reported symptom exacerbation after activity. All subjects completed two maximal effort CPETs conducted 24 hours apart. Unfortunately, the study did not include testing for post-exercise gene expression (like the Light study). No evaluation or follow up is reported, so we do not know how long it took the subjects to recover from the testing.

In Test 1, the CFS patients did not perform as well as controls. Multiple measurements were lower in the CFS group, including VO2max, peak workload, and workload at the anaerobic threshold. However, only the peak workload difference was statistically significant. In Test 2, the differences were quite dramatic. The controls performed the same or even better on the second test. But the CFS patients demonstrated a drop in VO2max, oxygen consumption at the anaerobic threshold, peak workload, and workload at the anaerobic threshold. The mean for the last value – workload at the anaerobic threshold – dropped by more than 50%. Respiratory measurements prove that all subjects gave a maximal effort in both tests, so the reduction is not due to lack of effort.

So what does this mean? The inability of people with CFS to reproduce their CPET results on day two is extraordinary. The authors state that it “could be utilized diagnostically as an objective indicator of abnormal post-exertional response, and possibly even a biomarker for this condition.” In fact, statistical analysis of the results correctly classified CFS patients and controls with 95.1% accuracy.

Healthy individuals stay below the anaerobic threshold most of the time. But this study showed that for many CFS patients, even activities of daily living require them to push past their anaerobic thresholds. My own test results demonstrated the same impairment. This could explain not only our limitations on good days, but why those limitations shrink during post-exertional malaise.

This study demonstrates the importance of using a two-day test protocol. While there were differences between patients and controls on the first day, only the CFS patients demonstrated a dramatic drop in performance on the second day. This significant reduction in performance appears to be unique to CFS. A recent study in patients with sarcoidosis (an inflammatory condition) failed to find any difference in CPET results between patients and controls, despite using a two day protocol.

One of the questions this paper cannot answer is: why? What do these results tell us about the underlying cause? The authors say, “It is very possible that a synergy of small effects across multiple systems is responsible for the poor exercise performance of the individuals with CFS in this study.” The results are consistent with reduced oxygen carrying capacity, possibly due to low blood volume, cardiac abnormalities, or autonomic dysfunction.

The conclusion of the paper is worth quoting at length:

In conclusion, a serial CPET protocol with measurement of expired gases demonstrates efficacy in distinguishing between individuals with CFS and sedentary, but otherwise healthy controls. As in the only other studies identified employing a dual CPET paradigm with measurement of expired gases, individuals with CFS showed a decrease in performance on the second test that was not seen in controls. This functional deficit may provide an objective indication of PEM. Despite considerable patient heterogeneity with respect to illness duration and type of onset, analysis of data from the second test was able to correctly classify 49 out of 51 individuals with CFS and 9 out of 10 controls. Non-invasive biomarkers for CFS do not currently exist. Physical therapists may consider the use of CPET performance measures to differentiate between individuals with CFS and otherwise non-disabled sedentary persons. Work efficiency (i.e. oxygen consumption and work output) at the ventilatory/anaerobic threshold appears to have diagnostic potential for CFS. (emphasis added)

This paper had a long journey to publication. The manuscript was submitted in October 2011, but was not accepted until June 2013. I hope its publication will lead to wider use of the two day maximal CPET protocol, particularly in research like the CDC’s multisite study. The suffering and agony of a two day test can be severe, as my own experience shows. But an objective way to diagnose CFS has held this field back for thirty years. If the two day CPET can provide objective diagnosis, then I say let’s go full speed ahead.


This entry was posted in Research and tagged , , , , , , , . Bookmark the permalink.

10 Responses to Two Is Better Than One

  1. Joe Landson says:

    Would really like to know the backstory of why this is taking so long to publish. Also it is astonishing to me that exercise physiology practitioners have not embraced these findings, given the importance it confers upon their field. Are they uncomfortable with being diagnostically relevant? Or are they just uncomfortable with ME/CFS?

  2. Sasha says:

    I welcome these findings but I worry about ‘destructive testing’! Is this a test that could be harmful or dangerous to patients beyond the short term? I’m not aware of any other diagnostic test that requires a patient to put themselves at risk of a serious and/or long-term relapse in their condition. I’d be scared of doing such a test without data on that. Also, as a patient on the very borderline of being able to live independently, I would probably need a carer while I recovered from a test like this.

    Perhaps I’m confused and this is being proposed as a research tool rather than a general diagnostic tool to be rolled out to diagnose any patient who presents to their doctor with ME/CFS symptoms?

    I wonder if there’s a modified, less demanding version of this test that would still show clear separation from controls.

    But I really do welcome this study – it’s a good step forward.

    • Jennie Spotila says:

      Great questions, Sasha. There is no published data on long-term impact of this testing. However, Staci Stevens has performed this test hundreds of times. I have not heard of anyone who was hurt permanently (although that’s what I’ve heard, not definitive data). It is not a risk free test. Although, remember that cardiac patients are put through this test too, with a doctor in the room in case they have a cardiac event!

      An easier version of this test, called submaximal testing, has been used by researchers to provoke symptoms. However, there is a strong difference of scientific opinion over whether a submax test is sufficient. I’ll be posting about that in the future. There is also no published data showing that submax testing can be used diagnostically.

      It’s early days yet for these kinds of tests. The two day max test was brutal for me, but the results were worth a pot of gold.

  3. Sasha says:

    Thanks Jennie – and I’m glad it was worth it for you!

  4. Firestormm says:

    Thanks Jennie. Nicely explained as ever. I do think this raises some concerns and that the general support for these tests might need to be considered.

    What ‘type’ of CFS/ME patients would you say these tests are more suitable for, Jennie? Presumably not those in the ‘severe’ category. Therefore if only suitable, let us say, for people experiencing ‘mild’ symptoms at the times prior to and during that of testing: how would such a test be used for others? And how could it possibly be used to diagnose the disease as a whole?

    What of other diseases for whom patients report similar feelings of a general ‘worsening’ of symptoms post-exertion? Does the above serve to make PEM unique to ME? Does it even have to?

    Personally, I am amazed that people who generally object to prescribed exercise regimes consider this test as something to welcomed diagnostically. Sounds like torture to me and ethically debatable; and I can’t think of anything less suited for testing ME patients in a doctors surgery with.

    Exercise endurance and the ability to recover and do the same thing again, as successfully; are but elements – perhaps – of what patient’s consider PEM. So I agree the ‘why?’ needs some further thought and explanation. As does how these specific exercises might be better applied to general life experience.

    I’d really like to think that the above can be adapted and reapplied to the kind of real-life situations that many of us find ourselves unable to deal with. I am also concerned that seeing patients with ME being able to exercise at such a rate and in this way could bite us in the butt – to coin a phrase. Are you not concerned that those with ME in this test could effectively exercise to the same level on Day 1 as the controls? Or maybe if these people are to be considered ‘mild’ then it shows the longer-lasting effects of ME.

    If those with ME who can do this test, are subsequently observed on Day 2 to have this marked problem; and it can be applied to the ME population as a whole then sure it is a good thing. I’m just not sure we can reapply these findings to the whole population or the disease; as easily as we might like to think.

    If I recall – from your blog and from the various discussions on PR – the CDC will not be looking at the 2-day test because Unger has been advised by ME experts it would be too much for patients to consider; although some form of objective testing will occur in this regard. Is that about right? Do you yet know what that testing might be? Thanks.

    • Jennie Spotila says:

      All good questions, Firestormm. Regarding the “type” of patients who can complete the test, how do you define severe and mild? The most severely ill, the bedridden, cannot complete this test that’s true. I believe that a person must cycle 6 to 8 minutes minimum for the test results to be valid and interpretable. I am not usually bedridden (although I am today), but am housebound, unable to drive, etc etc. I consider myself severly ill, and certainly not mildly ill! I was able to complete both days, at great physical cost.

      Regarding whether the abnormal test results are unique to ME, the data seems to indicate they may be. The sarcoidosis study I linked to did not find the same drop in function on day 2, despite the severe fatigue reported by the patients. The severe drop in function is not present in cardiac patients, even those who are disabled. I think more work needs to be done to establish if those with MS or rheumatoid arthritis show the same results on both days.

      The test can feel like torture, and there are ethical concerns about using it in studies (as the authors of this paper acknowledge). One reason I support use of this test is that it shows how wrong prescribed exercise regimens can be if they do not take the patients’ reports of symptoms seriously. True GET advances the patient’s exercise on a schedule, regardless of the patient’s report of PEM. The two day test proves how wrong that can be. The two day test also gives us hard data on our personal limits by defining the anaerobic threshold and the workload at the AT. I’ve been using that data for over a year to guide my activity levels.

      I’m not concerned that we look like deconditioned people on day 1 as long as the second day testing is completed. The standard physical therapy prescription for a deconditioned person is exercise. That’s why most of us are prescribed some form of exercise. I think it was very powerful for my (already excellent) physical therapist to see what she expected on day 1 and then get the shocking report for day 2. It fit her worldview of the normal prescription, and the dire need to change it in my case. It explains, in hard numbers, why I did not respond normally to her program.

      You are correct about the CDC study. Unger said the participating clinicians advised against the two day test, although it is not clear why. Their current plan is to do a one day exercise test to provoke PEM, and then follow with a cognitive test (although I don’t know what will be used), and patient reported symptoms. There is no question in my mind that CDC has a perfect opportunity to use a two day test, and I know there are people working on trying to convince them of that.

  5. floydguy says:


    “CFS” is currently wastebasket diagnosis. If exercise testing helps drain the swamp a little bit, I am all for it. I welcome the opportunity to take the test to see what the results indicate. I am currently in “no man’s land” with one top MD saying I have “CFS” and another who has recently retracted it saying I am a lot closer to Gulf War Illness than “CFS” – despite not being a vet.

    Personally, I am shocked that there are those who are trying to sabotage a test that has potential as a diagnostic test. Some people in this “community” like throwing out boat anchors as much if not more than the CDC.

  6. Jane L. says:

    Thanks for this synopsis, it’s great to know the research was published. As a patient who is not bedridden, but 90% housebound and unable to participate in the outside world (I get the sense there are a lot of us in this range – I would not say this is “mild” disability in any absolute sense), I was able to handle the test, even with a pretty big crash, and it was both sobering and validating to see the second day decline results. It was incredibly powerful to be able to put numbers and data to symptoms, both as a reference point to understand my objective functional capacity and a clean way to explain to doctors and disability handlers that exercise is most certainly NOT the answer to improvement.

  7. Annie says:

    I do welcome this testing but as someone who went from mainly housebound to bedbound three years ago after a ten minute walk (2 mins was my limit) and where I remain I do wonder if it could cause long term further deterioration in some. I have no answers

Comments are closed.