Neurodivergent-Specific Psychological Consequences of ME/CFS
Research stream: neurodivergent-psychological-consequences-mecfs
1 Raymaker et al. 2020 — Defining Autistic Burnout (Foundational)
Population: n=196 autistic adults (qualitative + online survey; community-engaged CBPR). Not an ME/CFS cohort. Key Findings:
- Defines autistic burnout as chronic, extensive exhaustion + reduced stimulus tolerance + reduced ability to perform everyday tasks.
- Drivers: cumulative sensory/social/cognitive overload + masking + unmet support needs.
- The exhaustion phenotype is phenomenologically continuous with ME/CFS energy loss — an overlap/differential-diagnosis claim, not comorbidity equivalence. Conclusion: Foundational definition of the ND exhaustion construct relevant to ME/CFS energy-loss overlap. Limitations: Autistic-only sample; defines burnout, does not diagnose ME/CFS. Certainty Assessment: 0.78 raw × 0.90 ND-exhaustion overlap weight = discounted ~0.70
2 Higgins et al. 2021 — Defining Autistic Burnout by Lived Experience (Delphi)
Population: n=69 autistic lived-experts + 14 professionals/researchers (Grounded Delphi). Key Findings:
- Consensus features: exhaustion, withdrawal, reduced tolerance, reduced functional capacity, skill loss / cognitive regression.
- Autistic burnout explicitly distinguished from clinical depression. Conclusion: Lived-experience consensus definition; separates ND burnout from depression. Limitations: Expert consensus, not empirical outcome data. Certainty Assessment: 0.68 raw × 0.90 = discounted ~0.61
3 Clarey et al. 2026 — Shame, Identity Disruption, and Functional Collapse in Autistic Burnout
Population: n=11 autistic adults (qualitative interviews). Key Findings:
- Shame shapes interpretation of functional limits and help-seeking.
- “Dual masking” (concealing autistic traits AND burnout signs) intensifies strain and delays recovery.
- Identity disruption is central to the burnout experience. Conclusion: ND-specific shame + identity disruption + dual masking — psychological-consequence texture absent from general ME/CFS psych chapters. Limitations: Small qualitative n=11. Certainty Assessment: 0.48 raw × 0.90 = discounted ~0.43
4 Ali et al. 2026 — How Autistic Burnout Feels (Earlier vs Later Diagnosis)
Population: n=20 autistic adults (8 childhood-diagnosed); KCL / UCL qualitative thematic analysis. Key Findings:
- Burnout experienced as “powering down” and/or “overactivation” of mind and body, producing craving for sensory and social rest.
- Later-diagnosed adults: burnout more chronic and confusing; substance coping; “not knowing why” took a (sometimes dangerous) toll — a few contemplated suicide. Conclusion: Energy-regulatory reading of ND exhaustion; later diagnosis = double burden (undiagnosed ND + unexplained chronic exhaustion). Limitations: Qualitative; suicidality reported anecdotally within themes, not quantified. Certainty Assessment: 0.52 raw × 0.90 = discounted ~0.47
5 Pagán et al. 2025 — Autistic Burnout and Posttraumatic Stress
Population: n=91 trauma-exposed autistic adults without intellectual disability. Key Findings:
- Autistic burnout strongly correlates with posttraumatic stress symptoms (r=0.70, p<0.001); exploratory factor analysis suggested shared factors (self-image, memory).
- Burnout related to depression (PHQ-9) and anxiety (GAD-7). Conclusion: Quantifies the psychiatric-correlate burden of the ND exhaustion phenotype. Limitations: Cross-sectional; novel burnout instrument. Certainty Assessment: 0.55 raw × 0.85 = discounted ~0.47
6 Kentrou et al. 2024 — Perceived Misdiagnosis of Psychiatric Conditions in Autistic Adults
Population: n=1,211 autistic adults (52.6% women, mean 42.3 y), Netherlands register. Key Findings:
- 24.6% (n=298) report ≥1 previous psychiatric diagnosis perceived as a misdiagnosis.
- Most frequent perceived misdiagnoses: personality, anxiety, and mood disorders.
- Chronic fatigue syndrome / burnout-related disorders perceived as misdiagnoses by 4% (n=49) of the total sample; ADHD by 2.8% (n=34).
- Autistic women (31.7%) report perceived misdiagnoses more often than men (16.7%). Conclusion: Evidence of diagnostic confusion / overshadowing around fatigue and energy-related conditions in autistic adults; CFS/burnout is one of the perceived prior misdiagnoses. Does not demonstrate that clinicians dismiss ME/CFS as “just autism.” Limitations: Perceived (self-report) misdiagnosis; register sample skewed educated. Certainty Assessment: 0.72 raw × 0.90 = discounted 0.65; further discounted to 0.55 for the corrected, more indirect ME/CFS framing.
7 Grant, Norton & Hoekstra 2026 — Camouflaging and Illness Perceptions in Central Sensitivity Syndromes
(Grant, Norton, and Hoekstra 2026)
Population: n=504 adults (autism, CSS, both, neither); CSS = fibromyalgia, IBS, migraine. Key Findings:
- Autistic camouflaging significantly predicts CSS symptoms.
- Autistic people with a CSS report significantly more negative impact of chronic illness on life despite no more symptoms than non-autistic people with CSS.
- Authors raise diagnostic overshadowing of autism by a CSS label. Conclusion: Direct masking + central-sensitivity link; autistic people with chronic-fatigue-adjacent syndromes carry disproportionate mental-health/identity burden. Limitations: CSS (not pure ME/CFS); online self-report. Certainty Assessment: 0.60 raw × 0.85 = discounted ~0.51
8 Rogers et al. 2017 — Fatigue in Adult ADHD: A Trans-Diagnostic Approach
Population: n=243 adult ADHD, n=86 CFS, n=211 healthy controls (Maudsley services). Key Findings:
- 62% of adults with ADHD meet fatigue caseness; ADHD group more fatigued than healthy controls.
- ADHD and CFS groups did not differ on functional impairment, mood, or self-efficacy — shared transdiagnostic psychological characteristics. Conclusion: Nearest ADHD-specific overlap anchor: adult ADHD and CFS share a mood/self-efficacy/functional psychological profile. Limitations: Group comparison, not a formal comorbidity cohort. Certainty Assessment: 0.60 raw × 0.85 = discounted ~0.51
9 Hodge & Meltzoff 2026 — Autistic Camouflaging and Mental Health: Scoping Review
Population: 48 studies (autistic participants). Key Findings:
- Positive relationship between camouflaging and poor mental health (depression, anxiety, stress, wellbeing, burnout), effect sizes small to large.
- Bidirectional in qualitative studies; association stronger for assimilation than masking subscales. Conclusion: Broad support that masking is a psychological-burden mechanism relevant to the ND double burden with energy limits. Limitations: No ME/CFS-specific content. Certainty Assessment: 0.62 raw × 0.80 = discounted ~0.50
10 Whaling et al. 2026 — Health-Condition Profiles Among Autistic Adults (Latent Class)
Population: n=640 autistic adults (autism-affirming telehealth; retrospective chart review). Key Findings:
- Three classes: Mental + Physical Health (17.7%), Mental Health Predominant (56.3%), Minimal (26.1%).
- Mental + Physical class shows mood/anxiety alongside chronic pain, chronic fatigue, headaches, dizziness, IBS, hormone problems. Conclusion: Documents autism + chronic fatigue + psych comorbidity clustering as a distinct higher-need class. Limitations: Chronic fatigue is a reported indicator, not an ME/CFS diagnosis; telehealth sample. Certainty Assessment: 0.50 raw × 0.55 = discounted ~0.28 (context only)
11 Quadt et al. 2026 — Neurodivergence and Variant Connective Tissue in Chronic Pain/Chronic Fatigue
Population: n=83 chronic pain/fatigue (CP/CF) adults vs n=91 non-clinical. Key Findings:
- CP/CF patients far more likely to screen likely-autism (aOR 14.3) and likely-ADHD (aOR 12.9).
- Joint hypermobility significantly mediates neurodivergence → CP/CF.
- Authors urge routine ND screening in CP/CF to improve fair access and quality of life. Conclusion: Reinforces high ND prevalence in chronic fatigue; flags fair-access/QoL need. Limitations: Predominantly prevalence/mechanism (overlaps predisposition stream); QoL peripheral. Supporting context only for this psych-consequence stream. Certainty Assessment: 0.62 raw × 0.50 (psych-consequence angle) = discounted ~0.31 (supporting)
13 Wanigasekera et al. 2025 — First Impressions Towards Autistic People: Systematic Review and Meta-Analysis
Full Citation:: Wanigasekera LC, Maybery MT, Palermo R, Whitehouse AJO, Tan DW. First Impressions Towards Autistic People: A Systematic Review and Meta-Analysis. Autism Research. 2025;18(5):983–1010. (Wanigasekera et al. 2025) DOI:: 10.1002/aur.70019 Key Findings::
- Meta-analysis of 21 articles (221 effects) comparing first impressions of autistic vs non-autistic people.
- Less favorable impressions of autistic people across all presentation modalities EXCEPT speech transcript, with moderate-to-large effects — replicating Sasson's "style not substance" at the meta-analytic level.
- Differences were most pronounced for interpersonal attraction and social/communication presentation.
- Non-autistic observer characteristics (autism knowledge, quality of contact) moderated impressions.Conclusion: Independent (Western Australia) meta-analytic confirmation that negative thin-slice impressions of autistic people are robust and modality-specific (audio-visual “style” vs verbal transcript “substance”), strengthening the generalizability of the anchor mechanism. Limitations: Synthesis of autism-target studies only; no ME/CFS/illness application; observer-moderation findings are correlational. Certainty Assessment: 0.80 raw × 0.75 (general-population perceivers of another condition) = discounted ~0.60 (supporting mechanism for ME/CFS by analogy)
14 DeBrabander et al. 2019 — Do First Impressions Differ Between Autistic and Nonautistic Observers?
Full Citation:: DeBrabander KM, Morrison KE, Jones DR, Faso DJ, Chmielewski M, Sasson NJ. Do First Impressions of Autistic Adults Differ Between Autistic and Nonautistic Observers? Autism in Adulthood. 2019;1(4):250–257. (DeBrabander et al. 2019) DOI:: 10.1089/aut.2019.0018 Key Findings::
- 32 autistic and 32 typically-developing (TD) raters judged videos of 20 TD and 20 autistic adults, with/without diagnostic disclosure.
- Autistic raters SHARED the TD tendency to rate autistic adults less favorably on traits, but this did NOT reduce their social interest in interacting (as it did for TD raters).
- Diagnostic disclosure did not improve first impressions for autistic raters (as it did for TD raters).Conclusion: IN-GROUP / NULL-RELEVANT evidence: autistic raters separate trait judgments from willingness-to-interact — dissociation that suggests the reduced-willingness-to-interact effect (the stigmatizing component) is strongest in cross-neurotype (TD perceiver) judgment, informing where perceiver bias vs stylistic difference operates. Limitations: Same Sasson lab (cohort-overlap with anchor); moderate sample (n=64 raters); hypothetical interaction intentions, not real behavior. Certainty Assessment: 0.60 raw × 0.75 (general-population) = discounted ~0.45 (supporting mechanism)
15 Morrison et al. 2019 — Rater Characteristics Drive First-Impression Variability
Full Citation:: Morrison KE, DeBrabander KM, Faso DJ, Sasson NJ. Variability in first impressions of autistic adults made by neurotypical raters is driven more by characteristics of the rater than by characteristics of autistic adults. Autism. 2019;23(7):1817–1829. (Morrison et al. 2019) DOI:: 10.1177/1362361318824104 Key Findings::
- Individual differences among neurotypical RATERS (autism knowledge, contact, personality) explained more variability in first impressions of autistic adults than characteristics of the autistic targets.Conclusion: Supports the relational framing (impression is co-produced by the perceiver, not a deficit solely in the target): the modifiable locus of first-impression bias is the perceiver, not the patient’s presentation. Limitations: Sasson lab (cohort-overlap); general-population raters; autism targets. Certainty Assessment: 0.62 raw × 0.75 = discounted ~0.47 (supporting mechanism)
16 Morrison et al. 2020 — Real-World Interaction: Autistic Paired with Autistic vs TD Partners
Full Citation:: Morrison KE, DeBrabander KM, Jones DR, Faso DJ, Ackerman RA, Sasson NJ. Outcomes of real-world social interaction for autistic adults paired with autistic compared to typically developing partners. Autism. 2020;24(5):1067–1080. (Morrison et al. 2020) DOI:: 10.1177/1362361319892701 Key Findings::
- Autistic adults paired with autistic partners reported more favorable interaction outcomes and rapport than when paired with TD partners.
- Autistic–autistic dyads showed greater interaction success.Conclusion: Direct behavioral support for the relational/double-empathy model: social difficulty is co-produced by neurotype mismatch rather than an intrinsic deficit — the frame by which first-impression stigma toward presentational style is understood. Limitations: Sasson lab (cohort-overlap); self-report + observer outcomes; real-world context but lab-facilitated pairing. Certainty Assessment: 0.62 raw × 0.75 = discounted ~0.47 (supporting mechanism)
17 Foster et al. 2025 — Facial Affect Differences and First-Impression Formation
Full Citation:: Foster SJ, Jones DR, Pinkham AE, Sasson NJ. Facial Affect Differences in Autistic and Non-Autistic Adults Across Contexts and Their Relationship to First-Impression Formation. Autism in Adulthood. 2025;7(5):581–593. (Foster et al. 2025) DOI:: 10.1089/aut.2023.0199 Key Findings::
- Quantified reduced/atypical facial affect in autistic vs non-autistic adults across contexts and linked it to first-impression formation.Conclusion: Mechanism paper identifying the specific presentational signal (facial expressivity) that drives negative thin-slice impressions — the most direct analogue for how reduced facial affect in invisible-illness patients (flat affect of chronic fatigue) could drive first-impression dismissal. Limitations: Sasson lab + Pinkham (cohort-overlap); autism context; cross-sectional. Certainty Assessment: 0.60 raw × 0.75 = discounted ~0.45 (supporting mechanism)
18 Jones et al. 2024 — Non-autistic Observers Both Detect and Demonstrate the Double Empathy Problem
Full Citation:: Jones DR, Botha M, Ackerman RA, King K, Sasson NJ. Non-autistic observers both detect and demonstrate the double empathy problem when evaluating interactions between autistic and non-autistic adults. Autism. 2024;28(8):2053–2065. (Jones et al. 2024) DOI:: 10.1177/13623613231219743 Key Findings::
- Non-autistic observers both detect and themselves demonstrate reduced empathy toward autistic adults when evaluating cross-neurotype interactions.Conclusion: Empirical operationalization of Milton’s double-empathy problem — the relational framework underpinning the claim that first-impression bias is a two-way (perceiver + target) phenomenon, applicable to invisible-illness stigma. Limitations: Sasson lab + Botha (cohort-overlap); evaluation of interaction vignettes, not clinical/illness settings. Certainty Assessment: 0.60 raw × 0.75 = discounted ~0.45 (supporting mechanism)
19 Boucher et al. 2023 — Negative First Impressions of Autistic Children by Non-Autistic Adults
Full Citation:: Boucher TQ, Lukacs JN, Scheerer NE, Iarocci G. Negative first impression judgements of autistic children by non-autistic adults. Frontiers in Psychiatry. 2023;14:1241584. (Boucher et al. 2023) DOI:: 10.3389/fpsyt.2023.1241584 Key Findings::
- Independent lab (Simon Fraser University) confirmed non-autistic adults form negative first impressions of autistic children from brief video exposure.Conclusion: Independent replication of the thin-slice first-impression effect in the pediatric population, outside the Sasson lab — reduces cohort-overlap concern for the core effect. Limitations: Pediatric autism targets; general-population adult perceivers; independent but single study. Certainty Assessment: 0.55 raw × 0.75 = discounted ~0.41 (supporting mechanism)
20 Norris et al. 2024 — Disclosing an Autism Diagnosis Improves Employment Interview Ratings
Full Citation:: Norris JE, Prosser R, Remington A, Crane L, Maras K. Disclosing an autism diagnosis improves ratings of candidate performance in employment interviews. Autism. 2024;28(4):1045–1050. (Norris et al. 2024) DOI:: 10.1177/13623613231203739 Key Findings::
- In mock employment interviews, disclosing an autism diagnosis IMPROVED rater assessments of candidate performance vs non-disclosure.
- Providing additional information about autism alongside the label did not add further benefit.Conclusion: NULL-OF-BIAS-WHEN-INFORMED evidence: first-impression bias toward presentational style is partially reducible when perceivers know the diagnosis — the mechanism implication that informing clinicians/peers of a patient’s condition can attenuate dismissal. Limitations: Independent UK group; employment-interview context; ratings not diagnostic-dismissal context. Certainty Assessment: 0.58 raw × 0.75 = discounted ~0.44 (supporting mechanism)
21 Efthimiou et al. 2025 — Diagnostic Status Influences Rapport in Dyadic Interaction
Full Citation:: Efthimiou TN, Lewis S, Foster SJ, Wilks CEH, Dodd M, Jiménez-Sánchez L, Ropar D, Ackerman RA, Sasson NJ, Fletcher-Watson S, Crompton CJ. Diagnostic status influences rapport and communicative behaviours in dyadic interactions between autistic and non-autistic people. PLoS ONE. 2025;20(8):e0330222. (Efthimiou et al. 2025) DOI:: 10.1371/journal.pone.0330222 Key Findings::
- Pre-registered Actor-Partner analysis (57 autistic, 51 non-autistic) in matched/mixed dyads.
- Autistic participants reported lower rapport regardless of partner status, but awareness of partner diagnostic status moderated it.
- Behavioural/kinematic indices did NOT mediate the neurotype–rapport gap.Conclusion: Diagnostic awareness moderates cross-neurotype rapport; behavior alone does not explain the rapport gap — refining the relational account and reinforcing that information/knowledge (not behavior change alone) shifts perception. Limitations: Multi-lab (Edinburgh, Sasson UT Dallas, UCL — partial cohort-overlap); dyadic rapport, not clinical dismissal. Certainty Assessment: 0.62 raw × 0.75 = discounted ~0.47 (supporting mechanism)
22 Edey et al. 2016 — Interaction Takes Two: Mind-Blindness Toward Those with Autism
Full Citation:: Edey R, Cook J, Brewer R, Johnson MH, Bird G, Press C. Interaction takes two: Typical adults exhibit mind-blindness towards those with autism spectrum disorder. Journal of Abnormal Psychology. 2016;125(7):879–885. (Edey et al. 2016) DOI:: 10.1037/abn0000199 Key Findings::
- Typical adults were less accurate at reading the mental states of autistic compared to typical targets — the mind-reading difficulty in cross-neurotype interaction is BIDIRECTIONAL ("double mind-blindness").Conclusion: Foundational experimental support that the perceptual/cognitive gap in cross-neurotype interaction is not confined to the autistic individual — underpinning the relational framing and its generalization to how perceivers misread the mental state of invisible-illness patients. Limitations: UCL/Birkbeck group (independent of Sasson); mental-state reading accuracy, not social interaction intention. Certainty Assessment: 0.62 raw × 0.75 = discounted ~0.47 (supporting mechanism)
24 Hemmesch et al. 2014 — Atypical Nonverbal Behavior and First Impressions in Parkinson’s Disease (Comparator)
Full Citation:: Hemmesch AR. The detrimental effects of atypical nonverbal behavior on older adults’ first impressions of individuals with Parkinson’s disease. Psychology and Aging. 2014;29(3):521–527. (Hemmesch 2014) DOI:: 10.1037/a0036637 Key Findings::
- Older adults viewed short clips of individuals with Parkinson's disease varying in facial masking (reduced expressivity) and abnormal bodily movement.
- Higher masking or abnormal bodily movement yielded more negative first impressions of social positivity.
- Masking was more detrimental for impressions of women and emotional-relationship goals; abnormal movement more detrimental for instrumental goals.Conclusion: TRANSDIAGNOSTIC COMPARATOR: atypical nonverbal “style” (facial masking — directly analogous to flat affect) drives negative first impressions in a chronic neurological condition, supporting that the mechanism generalizes beyond autism to chronic conditions with altered movement/expressivity. Limitations: Single author; older-adult perceivers; Parkinson’s target (not ME/CFS); first impressions only, not clinical dismissal. Certainty Assessment: 0.60 raw × 0.75 = discounted ~0.45 (supporting mechanism, comparator)
25 Slepian et al. 2014 — Thin-Slice Judgments in the Clinical Context
Full Citation:: Slepian ML, Bogart KR, Ambady N. Thin-slice judgments in the clinical context. Annual Review of Clinical Psychology. 2014;10:131–153. (Slepian, Bogart, and Ambady 2014) DOI:: 10.1146/annurev-clinpsy-090413-123522 Key Findings::
- Review establishing that clinicians form rapid thin-slice judgments from patients' nonverbal behavior, and these can bias clinical decisions.Conclusion: MECHANISM FRAMING bridge: the general thin-slice-judgment phenomenon (from the Ambady lab) applies in clinical contexts, providing the mechanism connecting autism first-impression findings to clinician first-impression bias toward invisible-illness (e.g. ME/CFS) patients. Limitations: Review (no primary data); general clinical context, not ME/CFS or autism specifically; Ambady lab. Certainty Assessment: 0.75 raw × 0.75 = discounted ~0.56 (supporting mechanism framing)
26 Parisi et al. 2026 — Reduced Facial Expression and Stigma in First Impressions of Schizophrenia (Comparator)
Full Citation:: Parisi M, Marin L, Lozano-Goupil J, Fauviaux T, Vattier V, Capdevielle D, Raffard S. Exploring the Role of Stigma and Reduced Facial Expressions in Shaping First Impressions of Individuals with Schizophrenia. Schizophrenia Bulletin. 2026;52(4):sbaf104. (Parisi et al. 2026) DOI:: 10.1093/schbul/sbaf104 Key Findings::
- 20 patients with schizophrenia vs 20 matched controls filmed 30 s; 300 general raters judged first impressions.
- Reduced facial expression (OpenFace analysis) drove negative first impressions, particularly mouth movement.
- Disclosing a SCHIZOPHRENIA diagnosis WORSENED ratings (stigma); mislabeling the same targets as autistic IMPROVED ratings.Conclusion: TRANSDIAGNOSTIC COMPARATOR + disclosure nuance: reduced facial affect drives negative impressions in another condition, AND diagnostic disclosure effects are CONDITION-SPECIFIC (helps for autism, harms for schizophrenia) — critical for how the paper frames diagnosis-disclosure as a stigma-reduction intervention for invisible illness. Limitations: Schizophrenia context; disclosure manipulation in abstracted video, not real clinical encounter; independent (Montpellier) group. Certainty Assessment: 0.62 raw × 0.75 = discounted ~0.47 (supporting mechanism, comparator)
27 Tasoula et al. 2026 — Multi-omics of Long COVID Reveals Persistent Mitochondrial Dysfunction
Full Citation:: Tasoula A, Arif S, Waisberg E, Bauer L, Aslinger E, Guarnieri JW. Multi-omics analysis of long COVID (post-COVID-19 condition) reveals persistent mitochondrial dysfunction, suppressed oxidative phosphorylation, and immune dysregulation. Frontiers in Immunology. 2026;17:1776555. (Tasoula et al. 2026) DOI:: 10.3389/fimmu.2026.1776555 Key Findings::
- Sustained suppression of OXPHOS across species and tissues, with skeletal muscle showing the most pronounced and persistent mitochondrial repression.
- Mitochondrial stress-response activation and inflammatory enrichment persist to ≥12 months post-infection.
- PBMC transcriptomics and serum metabolomics indicate prolonged OXPHOS downregulation up to 12 months.
- Links bioenergetic suppression to chronic immune activation.Conclusion: Integrative multi-omics identify persistent mitochondrial repression + immune dysregulation as recurring features of post-COVID condition, reinforcing the OXPHOS-suppression / reduced-ATP-yield cascade. Limitations: Secondary re-analysis of public datasets; hamster (animal) + in silico components; authors employed by commercial Guarnieri Research Group LLC (COI disclosed). Certainty Assessment: 0.55 raw × 0.70 (mixed LC human + animal + in silico) = discounted ~0.39
28 Weber-Fahr et al. 2026 — Reduced Brain ATP/PCr Ratios in Neuropsychiatric Post-COVID (³¹P MRS)
Full Citation:: Weber-Fahr W, Dommke S, Sack M, Alzein N, Becker R, Demirakca T, Ende G, Schilling C. Reduced adenosine triphosphate-to-phosphocreatine ratios in neuropsychiatric post-COVID condition: evidence from (31)P magnetic resonance spectroscopy. Biological Psychiatry. 2026;100(5):546-555. (Weber-Fahr et al. 2026) DOI:: 10.1016/j.biopsych.2026.01.004 Key Findings::
- Whole-brain ³¹P-MRSI in 27 post-COVID patients vs 23 recovered controls showed a significant cluster of reduced ATP/PCr centered on the cingulate cortex.
- Lower anterior-cingulate ATP/PCr correlated with poorer cognitive performance.
- Intracellular-pH trend and pH–ATP/PCr correlation only in patients suggest disease-specific bioenergetic/pH dysregulation.
- A post-COVID subgroup meeting ME/CFS criteria showed similar metabolic alterations.Conclusion: First in-vivo human evidence of impaired brain energy metabolism in post-COVID condition, directly linking anterior-cingulate bioenergetic deficit to cognitive impairment — reinforcing the brain energy-crisis / reduced-ATP claim. Limitations: Moderate n; primarily Long-COVID population (ME/CFS only a subgroup); cross-sectional. Certainty Assessment: 0.72 raw × 0.85 (Long COVID; ME/CFS subgroup) = discounted ~0.61
29 Germain et al. 2025 — Plasma Proteomic Maladaptation in ME/CFS Following Exertion
Full Citation:: Germain A, Glass KA, Eckert MA, Giloteaux L, Hanson MR. Temporal dynamics of the plasma proteomic landscape reveals maladaptation in ME/CFS following exertion. Molecular & Cellular Proteomics. 2025;24(12):101467. (Germain et al. 2025) DOI:: 10.1016/j.mcpro.2025.101467 Key Findings::
- Longitudinal 7K-aptamer plasma proteomics across two maximal CPETs (79 ME/CFS vs 53 sedentary controls) shows widespread proteomic changes in the PEM-recovery phase.
- Upregulation of glycolysis/gluconeogenesis suggestive of mitochondrial stress; suppression of T/B-cell signaling and IL-17.
- Decoupling of protein abundance from VO₂max and anaerobic threshold in ME/CFS vs controls.
- Sex-stratified molecular responses differ between females and males.Conclusion: Dynamic proteomic profiling shows maladaptive immune-metabolic responses to exertion in ME/CFS, with a sedentary-control arm distinguishing disease effects from deconditioning — reinforcing the glycolytic-shift and exercise-maladaptation cascade claims. Limitations: Circulating-proteome proxy (not tissue); same Cornell (Hanson/Giloteaux) cohort as other Germain entries in corpus (site overlap). Certainty Assessment: 0.72 raw × 1.00 (ME/CFS) = discounted ~0.72
30 Keele et al. 2026 — Cross-Tissue Gene/Proteomic Evidence Supports Mitochondrial Dysregulation in ME/CFS
Full Citation:: Keele GR, Enger M, Barnette Q, Ruiz-Esparza R, Alvarado M, Mathur R, Stratford JK, Giamberardino SN, Brown LM, Webb BT, Carnes MU. Systematic examination of gene expression and proteomic evidence across tissues supports the role of mitochondrial dysregulation in ME/CFS. International Journal of Molecular Sciences. 2026;27(4):1997. (Keele et al. 2026) DOI:: 10.3390/ijms27041997 Key Findings::
- Lower expression of mitochondrial genes MT-RNR1 and MT-RNR2 in ME/CFS in two studies (mapMECFS + DecodeME).
- Increased mitochondrial-gene expression in platelets in a third study — supports cross-tissue mitochondrial dysregulation.
- ME/CFS-associated genes map to approved compounds targeting mitochondrial-support/energy-metabolism genes.Conclusion: Bioinformatic re-analysis supports a role for mitochondrial dysregulation in ME/CFS across independent datasets, reinforcing the general mitochondrial-defect cascade. Limitations: No new primary data; in silico; authors note little cross-study consistency in specific genes (heterogeneity caveat). Certainty Assessment: 0.45 raw × 0.30 (in silico) = discounted ~0.14 (supporting)
31 Magawa et al. 2026 — Deficient TRPM3-Linked Mitochondrial Ca²⁺ Influx in ME/CFS NK Cells
Full Citation:: Magawa CT, Eaton-Fitch N, Muraki K, Marshall-Gradisnik S. Deficient TRPM3-linked mitochondrial Ca(2+) influx in natural killer cells associated with myalgic encephalomyelitis/chronic fatigue syndrome. BMC Immunology. 2026;27(1):65. (Magawa et al. 2026) DOI:: 10.1186/s12865-026-00849-1 Key Findings::
- In NK cells from n=10 ME/CFS (CCC) vs 10 HC, cytosolic Ca²⁺ influx amplitude/slope reduced and mitochondrial Ca²⁺ influx altered in ME/CFS.
- TRPM3-dependent cytosolic and mitochondrial Ca²⁺ mobilization significantly reduced in ME/CFS.Conclusion: Altered TRPM3-mediated Ca²⁺ handling affects mitochondrial Ca²⁺ mobilization in ME/CFS immune cells — extending the mitochondrial-Ca²⁺ dysregulation cascade into the immune compartment. Limitations: Small n; single lab (Griffith University TRPM3-NK line already extensive in corpus); ex-vivo/in-vitro readout, not muscle. Long-COVID companion paper (PMID 42484498, same lab) not separately added to avoid redundancy. Certainty Assessment: 0.50 raw × 0.60 (ME/CFS population, in-vitro readout) = discounted ~0.30
32 Groysman et al. 2026 — Fragile Mitophagy as a Recovery-Failure Endotype for PEM (Hypothesis)
Full Citation:: Groysman R. Fragile mitophagy in long COVID: a proposed recovery-failure endotype for post-exertional malaise. Frontiers in Medicine. 2026;13:1905758. (Groysman 2026) DOI:: 10.3389/fmed.2026.1905758 Key Findings::
- Proposes "fragile mitophagy" (mitochondrial injury + mitophagy engagement but inadequate lysosomal completion) as a PEM recovery-failure endotype.
- Predicts delayed crashes, progressive baseline lowering with repeated exertion, and poor tolerance of mitophagy-inducing interventions.
- Predicts PEM severity tracks impaired lysosomal completion / danger signaling more than baseline ATP deficit.Conclusion: Hypothesis offering a novel recovery-failure framing that extends the mitochondrial-damage → mitophagy-impairment cascade into PEM endotyping. Limitations: Single-author opinion/hypothesis article with no primary data; Long-COVID framing; conceptual only. Certainty Assessment: 0.30 raw × 0.40 (conceptual Long-COVID) = discounted ~0.12 (concept only)
33 Ryback et al. 2026 — Indistinguishable Mitochondrial Phenotypes After ME/CFS vs Control Serum (Failed Replication)
Full Citation:: Ryback AA, Hillier CB, Loureiro CM, Ponting CP, Dalton CF. Indistinguishable mitochondrial phenotypes after exposure of healthy myoblasts to myalgic encephalomyelitis/chronic fatigue syndrome or control serum. PLoS One. 2026;21(2):e0341334. (Ryback et al. 2026) DOI:: 10.1371/journal.pone.0341334 Key Findings::
- Pre-registered replication of Fluge 2016 with sera from 67 ME/CFS and 53 controls and >1,700 Seahorse mitochondrial stress tests.
- NO significant difference in maximal respiratory capacity between ME/CFS and control serum treatment of healthy myoblasts.Conclusion: Strong evidence against the hypothesis that ME serum factors differentially alter healthy-myoblast mitochondrial phenotype in vitro — a null result limiting the serum-transferable layer of the mitochondrial-defect cascade. Limitations: In-vitro model only; does not refute in-vivo tissue-mitochondrial findings (muscle biopsy WASF3/supercomplex/PDH; bed-rest comparison; brain ³¹P-MRS) which rest on tissue rather than serum-cell transfer. Certainty Assessment: 0.75 raw × 0.40 (in vitro mechanistic test) = discounted ~0.30 (negative)