Cognitive assessment in ME/CFS: a cognitive substudy of the multi-site clinical assessment of ME/CFS (MCAM), 2024, Lange, Unger +

[Dolphin]

https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2024.1460157/full

ORIGINAL RESEARCH article
Front. Neurosci., 01 November 2024
Sec. Translational Neuroscience
Volume 18 - 2024 | https://doi.org/10.3389/fnins.2024.1460157

Cognitive assessment in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS): a cognitive substudy of the multi-site clinical assessment of ME/CFS (MCAM)

Gudrun Lange1*Jin-Mann S. Lin2Yang Chen2Elizabeth A. Fall2

Daniel L. Peterson3Lucinda Bateman4Charles Lapp5Richard N. Podell6

Benjamin H. Natelson1

Andreas M. Kogelnik7Nancy G. Klimas8,9Elizabeth R. Unger2 on behalf of the MCAM Study Group

• 1Pain and Fatigue Study Center, Department of Neurology, Icahn School of Medicine at Mount Sinai, New York, NY, United States
• 2Division of High-Consequence Pathogens and Pathology, National Center for Emerging and Zoonotic Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA, United States
• 3Sierra Internal Medicine, Incline Village, NV, United States
• 4Bateman Horne Center, Salt Lake City, UT, United States
• 5Hunter-Hopkins Center, Charlotte, NC, United States
• 6Richard N. Podell Medical, Summit, NJ, United States
• 7Basis Diagnostics, Inc., Newark, CA, United States
• 8Institute for Neuro Immune Medicine, Nova Southeastern University, Fort Lauderdale, FL, United States
• 9VA Medical Center, Geriatric Research and Education Clinical Center, Miami, FL, United States

Introduction: Patients with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) experience cognitive problems with attention, information processing speed, working memory, learning efficiency, and executive function. Commonly, patients report worsening of cognitive symptoms over time after physical and/or cognitive challenges. To determine, monitor, and manage longitudinal decrements in cognitive function after such exposures, it is important to be able to screen for cognitive dysfunction and changes over time in clinic and also remotely at home. The primary objectives of this paper were: (1) to determine whether a brief computerized cognitive screening battery will detect differences in cognitive function between ME/CFS and Healthy Controls (HC), (2) to monitor the impact of a full-day study visit on cognitive function over time, and (3) to evaluate the impact of exercise testing on cognitive dysfunction.

Methods: This cognitive sub-study was conducted between 2013 and 2019 across seven U.S. ME/CFS clinics as part of the Multi-Site Clinical Assessment of ME/CFS (MCAM) study. The analysis included 426 participants (261 ME/CFS and 165 HC), who completed cognitive assessments including a computerized CogState Brief Screening Battery (CBSB) administered across five timepoints (T0-T4) at the start of and following a full day in-clinic visit that included exercise testing for a subset of participants (182 ME/CFS and 160 HC). Exercise testing consisted of ramped cycle ergometry to volitional exhaustion. The primary outcomes are performance accuracy and latency (performance speed) on the computerized CBSB administered online in clinic (T0 and T1) and at home (T2-T4).

Results: No difference was found in performance accuracy between ME/CFS and HCs whereas information processing speed was significantly slower for ME/CFS at most timepoints with Cohen’s d effect sizes ranging from 0.3–0.5 (p < 0.01). The cognitive decline over time on all CBSB tasks was similar for patients with ME/CFS independent of whether exercise testing was included in the clinic visit.

Conclusion: The challenges of a clinic visit (including cognitive testing) can lead to further cognitive deficits. A single short session of intense exercise does not further reduce speed of performance on any CBSB tasks.

https://twitter.com/user/status/1852372138558169427

 

[ME/CFS Skeptic]

So accuracy showed no difference but performance speed does. Exercise had no further impact on cognitive functioning.

Here's quote from the discussion section that summarizes the findings:

This MCAM cognitive sub-study showed that persons with ME/CFS are able to react and attend to simple cognitive tasks as well as HCs at the outset of the study visit (T0), but are not able to maintain these challenges over time. In fact, while their simple reaction time remains unchanged simple attention becomes more variable over 48 h leading to poorer task performance compared to HCs. In contrast, on tasks requiring increased cognitive efficiency and “multitasking” involving learning, memory, working memory, and executive function, significant differences in response latency or information processing speed (Sweet, 2011), are already evident at T0, in clinic, and increase over time to clinically meaningful degrees. The use of accuracy as an outcome measure did not prove to be sensitive enough to determine ME/CFS cognitive dysfunction. Performance accuracy of ME/CFS participants was similar to that of HCs across timepoints and tasks. In contrast, performance latency/speed of mental information processing objectively determined the presence and nature of cognitive dysfunction experienced by persons with ME/CFS in clinic and up to 48-h later at home

 

[ME/CFS Skeptic]

A description of the tasks is available in the supplementary material. Except for the maze (GML) they seem rather easy tasks focusing on detection and memory.

CogState Brief Screening Battery (CBSB)

Detection Task (DET): Task stimuli are images of playing cards showing either red or black jokers. The DET measures simple reaction time (RT) defined as the time it takes to press a “Yes” button as soon as a playing card in the center of the screen flips over.

Identification Task (IDN): Task stimuli are again images of playing cards showing either red or black jokers. The IDN task measures attention by employing a choice reaction time paradigm. Participants have to indicate whether or not a playing card is red as quickly as possible by pressing “Yes” if the playing card is red and “No” if it is black.

One Card Learning Task (OCL): Task stimuli are again playing cards, but this time without jokers. The OCL task measures recognition memory by employing a pattern separation paradigm. Participants press the “Yes” or “No” button to indicate whether or not they recognize having seen the currently displayed card at any time previously in the deck.

1-Back Working Memory Task (ONB): Task stimuli again are playing cards without jokers. The ONB measure working memory by employing the n-back paradigm. Again, participants press a “Yes” or “No” button as quickly as possible to indicate whether or not the current card presented is the same as the card just previously presented (one back) or not.

2-Back Working Memory Task (TWOB): Task stimuli again are playing cards without jokers. The TWOB measures working memory also by employing the n-back paradigm. In the TWOB task, participants press the “Yes” or “No” button as quickly as possible to indicate whether or not the current card presented is the same as the card shown two cards previously (two back). The first response is always “No” because no comparison card has been presented yet.

Groton Maze Learning Task (GML): The GML assesses problem solving, reasoning, and efficient learning under time pressure using a maze learning paradigm. A 28-step pathway is hidden among 100 possible locations in a 10 × 10 grid of tiles on the screen. Each box represents move locations, and the grid refers to the box array (i.e., 10 × 10). Participants are required to find a hidden pathway guided by four search rules. These rules are: 1) do not move diagonally, 2) do not move more than one box (i.e., do not jump), 3) do not move back on the pathway, and 4) return to the last correct location after an error. Feedback is given with visual and auditory cues (green check marks and red crosses as well as beeps) to indicate whether the selected box is correct or incorrect. The last correct location, flashes with a green check when two errors are made in succession (failing to return errors). At each step only the most recently selected box is shown. There are 20 well-matched alternate pathways available. The software records each move as either an error or as a correct move. The primary outcomes are the total number of errors made in attempting to learn the same hidden pathway on five consecutive trials at a single session (GML-TER) and the total number of correct moves made per second (GML-MPS).​

 

[ME/CFS Skeptic]

Traditional Neuropsychological Tests
The following two traditional neuropsychological tests were used and took about 10 minutes to be completed.

Test of Premorbid Functioning (TOPF (Pearson, 2009): The TOPF requires participants to read a list of 70 phonetically irregular words. The ability to successfully pronounce irregularly spelled words is relatively resistant to neurological injury and a sensitive marker of intellectual attainment (Holdnack, et al., 2013). Participants must read and pronounce a list of words printed in two columns on the front and back of a card. The TOPF was scored according to procedures outlined in the manual (Pearson, 2009). The best possible raw score for the TOPF is 70 and a derived age-corrected standard score (SS) can be used to predict the expected premorbid WAIS-IV Full Scale IQ score. The TOPF was only administered at T0.

The Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) Digit Span Forward and Backward tasks (DSF & DSB) (Wechsler, 2008): These two tasks were verbally administered. The DSF is often described as a test of simple auditory attention while the DSB is a test of simple auditory working memory. In the DSF test, participants are asked to repeat digits, the length of the digit sequence is increased across trials until there is a failure across two consecutive trials of a particular length. In the DSB test, participants are asked to repeat digits backwards across trials until there is a failure across two consecutive trials of a particular length. Raw scores range from 0 to 16 on both tasks and can be converted to age-corrected Standard Scores (SS). DSF and DSB were only administered at T0 and T1.​

 

[bobbler]

So accuracy showed no difference but performance speed does. Exercise had no further impact on cognitive functioning.

Here's quote from the discussion section that summarizes the findings:

This is more important than we might think.

it underlines the issue is exhaustion and load (same category - more load = more exhaustion) and not ‘ability’ or importantly ‘capacity’ in the sense of DOLS because if not already antagonised health-wise or wsiting fir right point we can both think very clearly having taken in info to make decisions and communicate them with adjustments appropriate to our disability. Yet if someone rushes us having decided to shine lights to desensitise us to photophobia or noise or whatever then that’s like the old cliche of interrupting a stroke patient fir speaking slowly or not pronouncing something right and assuming they also can no longer think.

I can’t help but being very aware and pointing out to non-Uk people that in the UK it has always been a strange historical point that dementia patients including Alzheimer’s got screwed on care and had to sell their home ending up with care for many years where other health illnesses it was different

alzheimer I mention specifically because the rest are ‘old age’ and that was the justification (pay for your own old age care if you’ve savings) but Alzheimer’s has early-onset variations.

 

[mango]

"A new scientific paper shows impaired cognitive functioning in ME/CFS. While this will not be a surprise to anyone with the condition, the study helps us understand the nature and extent of the problem. You can find the open access full paper here: https://pmc.ncbi.nlm.nih.gov/articles/PMC11565701/

Jarred Younger"

 

[Hutan]

Introduction

Over the years, neuropsychological studies have shown that cognitive dysfunction in ME/CFS is independent of mood disorders (Cockshell and Mathias, 2013; Majer et al., 2008; Robinson et al., 2019) and is not reflective of poor effort or motivation (Cvejic et al., 2016; Cockshell and Mathias, 2012; Lange et al., 2005).

Look at the years of the cited studies - most from over a decade ago. No mention of Walitt's effort preference mess.

The Multi-site Clinical Assessment of ME/CFS (MCAM) study (Unger et al., 2017) provided this opportunity. This study enrolled a large sample of diverse and well-characterized patients with ME/CFS cared for in seven U.S. ME/CFS specialty clinics.

The cognitive sub-study was designed with three primary objectives in mind. The first objective was to determine whether a brief cognitive battery, tailored to the ME/CFS cognitive deficiency profile as established in the peer-reviewed literature, (Sebaiti et al., 2022) reliably distinguishes between ME/CFS and HC over time. The second objective was to assess whether an all-day in-clinic study visit had a decremental effect on cognitive function. The third objective sought to determine whether a single short session of strenuous physical exercise would impact on cognitive function over and above that of the baseline clinical visit over 48 h comparing participants who did and did not undergo exercise testing (Cook et al., 2022).

 

[Hutan]

Methods

This MCAM cognitive sub-study was conducted between November 2013 and February 2019.

I wonder why this study took so long to be published?

Thus, the sample included in the current analysis are 426 participants: 261 ME/CFS and 165 HC. Of 261 participants with ME/CFS, 182 also completed exercise testing.

Computerized cognitive assessment was administered across five timepoints: two in-clinic assessments (T0, near the beginning of the clinic visit, before the exercise testing (if done) and T1 immediately after exercise or at end of clinic visit if no exercise testing) and up to three assessments administered at home (approximately 6–12 (T2), 24 (T3) and 48 h (T4) after participants’ clinic visit).

Socio-demographics such as age and sex, and pre-illness estimate of overall intellectual function (measured by TOPF) were used to adjust for the associations of the cognitive outcomes with study groups.

 

[Hutan]

Results

ME/CFS and HC were significantly different in age, race/ethnicity, and employment.

The lack of good matching on age is a shame. 9.6% of the ME/CFS sample were aged 18 to 29. 24.9% of the healthy controls were aged 18 to 29. That's a really big difference in the group of people who could be expected to have the quickest processing speed. I know that they made adjustments made for age, but that's a lot of scope for investigator bias. Age is such an important thing to match on, so it's disappointing that this study missed the mark by so much. The study ran over years, so I'm not sure what reason there could have been for such a poorly matched control group.

 

[Hutan]

ME/CFS participants had higher TOPF standard scores than HC (116.5 vs. 111.8, p < 0.01, Cohen’s d = 0.4). No significant difference was found in age-corrected standard scores of DSF and DSB between groups at T0 and T1.

The p value was actually 0.0002, so a very significant difference. This supposedly is the measure of pre-existing intellectual attainment, with the ME/CFS participants scoring better. This again suggests that the healthy controls were not matched well.


CogState brief screening battery DET (Detection Task) is claimed to be a measure of psychomotor speed/simple reaction time.


Red top line is the controls, blue bottom line is ME/CFS. X axis is T0-T4 - Times are as described above in Methods. I don't think this data has been adjusted for the age differences of the cohorts. What is clear is that the mean ME/CFS performance declines over time, while the mean control performance does not.

 

[Hutan]

Impact of exercise

The “no-exercise” group did not differ from the “exercise” group in response latency on any CBSB tasks except for GML (GML-MPS: (T0: 56% versus 66%, d = 0.5; T1: 59% versus 70%, d = 0.6; all p < 0.001). Significant differences were found in race and education between “exercise” and “no-exercise” groups but not in any measures for overall functioning and illness symptoms, prompting us to also adjust the comparison for race and education in addition to age, TOPF performance, and psychomotor speed. With the adjustment of age, race, education, TOPF, and psychomotor speed, the GML-MPS differences remained significant.

It's a bit hard to know what to make of all that.
"GML assesses problem solving, reasoning, and efficient learning under time pressure using a maze learning paradigm." It's challenging, and it was the one test where ME/CFS performance significantly differed with and without exercise.

Perhaps, for many participants, the amount of exercise was just a drop in the bucket compared to the effort of getting ready and getting to the clinic, doing what was required and getting home again? It also seems that the two sub cohorts (exercise and no exercise) were not very well matched. Perhaps what is needed is a study where individuals act as their own control, one time doing just the cognitive tasks and another time doing both the cognitive and exercise tasks.

Second, TWOB and GML tasks were not administered remotely. Thus, we were not able to determine whether exercise would impact performance speed on complex working memory and executive function/learning efficiency tasks after the clinic visit (approximately 6–12, 24, and 48 h after the exercise testing).

So, they didn't measure the GML task at home, only at the clinic (T0 and T1). It certainly would have been good to know what would have happened.

 

[Hutan]

Discussion

In fact, while their simple reaction time remains unchanged simple attention becomes more variable over 48 h leading to poorer task performance compared to HCs.

It seems that a similar but much larger study was done by others, published in 2022. The tests were only done once, and the ME/CFS participants were probably poorly characterised.

Joustra et al. (2022) administered the online CogState battery comprised of four tasks (DET-LMN, IDN-LMN, OCL-ACC, OBK-LMN) during an onsite study visit to a large population-based sample that included 70,951 healthy controls and 2461 participants with ME/CFS fulfilling the 1994 research case definition as assessed with symptom questionnaires and physical examination. They found that participants with ME/CFS performed significantly poorer on all tasks compared to healthy controls though with a small effect size. They did not include a correction for motor reaction time (DET-LMN) and education. The length of time of the study visit and timing of the cognitive test relative to other steps of the study visit are not specified and may explain the small effect size. Our findings of larger effect sizes after a full day study visit (T0 at start and T1 at end) with further differences in follow-up indicate the importance of repeat and at home testing.

Our findings add to the growing realization that persons with ME/CFS are compromised cognitively, they are reacting slower, attention is variable over time, and cognitive efficiency supporting executive function, learning and memory is significantly decreased to a clinically meaningful degree. The adoption of remote cognitive screening over time shows that cognitive dysfunction is and remains present after a clinical visit with or without stressful physical activity.

Protocolized repeatable cognitive screenings have long been devised and implemented for disorders with anatomical brain illness markers such as Multiple Sclerosis i.e., (Benedict et al., 2012) and Epilepsy i.e., (Kurzbuch et al., 2013). However, this development has been lacking for disorders without focal, observable brain involvement including ME/CFS. In our opinion, this is a significant disservice to these patient groups. It is essential to compare results of cognitive screenings across studies in a more reliable, standardized, and valid way than is currently done. Efforts to do so are ongoing by the ME/CFS Common Data Elements (CDE) consortium spearheaded by NIH and CDC. While that is an important step forward for research, brief, repeatable, computerized, and cost-effective screening tools need to be available for clinicians, not necessarily trained in neuropsychological assessment, to quickly make an initial diagnosis about whether or not cognitive decrements in their ME/CFS patients are present and changing over time to optimally manage illness symptoms and quality of life in ME/CFS patients.

 

[Hutan]

I thought this was a decent and interesting study. It's good to see such a study being funded by the CDC. Of course, there are things that could be improved or that warrant more investigation.

I'm thinking that there must be potential prospective studies of reaction time that could be done. There must be existing records of reaction time from the military and from other groups, maybe sports development squads, that could be accessed to see if/how it changes after developing ME/CFS or LC ME/CFS. Have we seen any studies like that?

There's no discussion about a possible mechanism.

 

[duncan]

This is a step in the right direction, but I worry about the abstract/summary; many clinicians will jump to results and conclusions and only see

Results: No difference was found in performance accuracy between ME/CFS and HCs whereas information processing speed was significantly slower for ME/CFS at most timepoints with Cohen’s d effect sizes ranging from 0.3–0.5 (p < 0.01). The cognitive decline over time on all CBSB tasks was similar for patients with ME/CFS independent of whether exercise testing was included in the clinic visit.

Conclusion: The challenges of a clinic visit (including cognitive testing) can lead to further cognitive deficits. A single short session of intense exercise does not further reduce speed of performance on any CBSB tasks.

I've taken five tests of this type over the last several years, and I've found them generally off-mark.

They need to compare premorbid scores of pwME to current scores.

They need better tests. Cognitive decline is substantive in many pwME, and those declines can be global. They also seem to dovetail with other red flags for brain involvement like OI and POTS and balance issues. These tests should also assess the sifting and sorting and executing of multiple cognitive demands simultaneously like the real world requires. And what about domains such as planning and creativity and inference abilities?

Moreover, these tests don't seem to consider the pedestrian capability to focus and perform better than usual for relatively short bursts, but a capability that cannot be sustained. Give five different but similar tests over five days in a row and see where the values fall.

There are some good names on this list of authors, but some other ones that cause me concern.

ETA: Geez, do even I have any idea how many times I had to edit this short narrative that years ago would have just effortlessly flowed?? And it's still muddled! Test that figgen deficit and assess it accurately. Please.

 

[Turtle]

Prior to my appointment with researchers in The Netherlands I was asked to do a N 2(pictures) back N 3 back test over a 3 week period, on-line, at home, on good moments and bad moments.
Reaction time and memory accuracy were tested. The differences in good or bad moments were pretty obvious. In 3 back misstakes doubled and reaction time was up 50% on bad moments.
Had I only done a few tests elsewhere, with travel time, the difference would not have been so clear. (effect of adrenaline?)

Had I even tried after my appointment, the outcome would have been disastrous. I had to travel during covid-lockdown, 6 hours (twice) of travel time on public transport. In the middle of the day a tilt-table-test.
Up at 4.00 and in bed at 22.00, I lost my marbles for 48 hours, (working memory?) (@MelbME that doesn't mean I'm bonkers), PEM lasted a month. I was given a standing test, 60+ and just 40% VO2max on CPET. Stepping away on 90 degrees standing still afterwards (-25%CBF). A disaster in the making.

To all people doing tilt-table testing, professionals and patients; Please end the test in supine position. The test takes away your coping mechanisms.
Ending supine means you get your coping mechanisms back and restores the CBF-drop at least a little.
@duncan this post took me over 2 hours, on and off.

edited for clarity

 

[Medfeb]

I wonder why this study took so long to be published?

Typical of CDC's MCAM studies

 

[Yann04]

Does it say how long after exercise the cognitive testing was done?

Because if I overexert myself, I get a bout of what feels like adrenaline and my cognitive function doesn’t decline until the adrenaline (or whatever) wears off and PEM hits the next day.

Edit: Hmm it seems they had a test after 48hrs

Computerized cognitive assessment was administered across five timepoints: two in-clinic assessments (T0, near the beginning of the clinic visit, before the exercise testing (if done) and T1 immediately after exercise or at end of clinic visit if no exercise testing) and up to three assessments administered at home (approximately 6–12 (T2), 24 (T3) and 48 h (T4) after participants’ clinic visit)

Well a lack of difference there is certainly intriguing, but also worth keeping in mind these people had the functional capacity to spend an entire day at a clinic, which leaves out a specific portion of ME pateints.

 

[Dolphin]

ME Research UK

Cognitive difficulties, often referred to as “brain fog”, are one of the four key symptoms required for a diagnosis of ME/CFS under the 2021 NICE guidelines.

Despite this, existing methods to assess cognitive function in people with ME/CFS are limited and may not fully capture the complex nature of the symptom.

Therefore, a study, published in the journal “Frontiers of Neuroscience”, investigated whether a short series of questions relating to cognitive function, reliably distinguishes between people with the disease and healthy controls over time.

Find out more about the study, and what the researchers found here: https://bit.ly/4h41m7H

https://twitter.com/user/status/1877656930333643004

 

[dave30th]

I wonder why this study took so long to be published?

I assume the pandemic put it on the back burner.