Exercise and Movement

1 The Exercise Paradox

1.1 Why Standard Exercise Programs Fail in ME/CFS

Exercise is beneficial for most chronic conditions and healthy populations, improving cardiovascular fitness, strength, mood, and metabolic health. However, ME/CFS represents a notable exception where standard exercise physiology does not apply.

Normal Exercise Adaptation Healthy individuals respond to exercise training with:

• Improved mitochondrial density and function
• Enhanced cardiovascular capacity
• Increased muscle strength and endurance
• Positive mood effects (endorphin release, reduced depression)
• Progressive tolerance of higher workloads

Pathological Exercise Response in ME/CFS ME/CFS patients instead experience:

• Worsening symptoms following exertion (PEM)
• No adaptive improvement with repeated exercise
• Measurable physiological deterioration (documented by two-day CPET—though group-average decline is contested; see Section Exercise Precautions)
• Cumulative functional decline with sustained exercise programs
• Prolonged recovery periods (days to weeks) after single exertional episodes

This fundamental difference reflects underlying metabolic dysfunction rather than deconditioning or psychological factors.

1.2 Graded Exercise Therapy (GET): Controversy and Evidence of Harm

Graded exercise therapy—progressive incremental increases in physical activity—was historically recommended for ME/CFS based on the assumption that symptoms reflected deconditioning, fear avoidance, or deconditioning-related fatigue. This assumption has been decisively refuted by objective evidence.

The PACE Trial and Subsequent Reanalysis

The 2011 PACE trial initially claimed benefits from GET and cognitive behavioral therapy (CBT). However, subsequent reanalysis using objective outcomes (rather than subjective questionnaires) found (Wilshire et al. 2018)

• No significant improvement in objective measures (6-minute walk distance, step counts, employment, benefits claims)
• High rates of patient-reported harm in long-term follow-up
• Methodological concerns including subjective outcomes, non-blinded assessments, and changing outcome definitions

Major health authorities have since revised guidelines to recommend against GET for ME/CFS, including NICE (UK, 2021), CDC (USA, 2022), and others (National Institute for Health and Care Excellence 2021).

The strongest defense of the PACE/GET evidence remains case-definition stratification: GET proponents argue that trials using Oxford criteria enrolled heterogeneous patients, many without PEM, and that modest subjective improvements in that broader population are not contradicted by the evidence of harm in ICC/CCC-defined patients with obligate PEM. This is a methodologically legitimate distinction — the evidence against GET applies specifically to PEM-positive patients, not necessarily to all chronic fatigue presentations.

Not all national bodies have followed suit. In Australia, the Royal Australian College of General Practitioners (RACGP) maintained a HANDI recommendation for “incremental physical activity” in ME/CFS until it was formally challenged in 2026. Stallard et al. demonstrated that the guideline scored only 2% on the AGREE II rigor domain, relied on the methodologically contested PACE trial and an outdated Cochrane review, and ignored patient surveys reporting 54–74% harm rates (Stallard et al. 2026). The authors called for immediate withdrawal and contraindication of GET pending the NHMRC’s review of ME/CFS guidelines, recommending that Australian general practitioners refer to NICE and BMJ Best Practice guidelines in the interim.

Two-Day CPET Evidence Against GET

Objective physiological evidence demonstrates why GET is contraindicated. Keller et al. (2024) showed that even a single maximal exertion produces (Keller et al. 2024).

• Day 2 performance decrements: 5–8% declines in VO₂peak, work output, ventilation
• Worsening impairment classification: Severe impairment cases nearly doubled (14% → 27%)
• Independence from fitness: Abnormal responses persisted when matched for baseline aerobic capacity
• Prolonged recovery: Full restoration requiring 13+ days versus \(\sim\) 2 days in controls

A 2026 null replication by Mancini, Natelson et al. (Fukuda criteria, n=58+25) found no significant group-average Day 1→Day 2 decline (Mancini et al. 2026), but both studies concur that ME/CFS patients experience elevated perceived exertion (Borg RPE) at all workloads and lower maximal HR (chronotropic incompetence). The contested CPET evidence does not weaken the clinical conclusion — GET produced 51% harm rates regardless of which CPET result is ultimately replicated, because the harms are driven by repeated exertion exceeding individual tolerance, not by the presence or absence of a group-average VO₂ decline signal.

Mechanistic Understanding

The two-day CPET results validate patient reports by demonstrating:

• Exercise triggers measurable metabolic failure beyond normal fatigue or deconditioning
• Recovery systems are impaired, requiring prolonged restoration periods
• Repeated exertion before recovery worsens baseline function
• The phenomenon is reproducible and objectively quantifiable

Charlton et al. (2026) reinforced this conclusion in the British Journal of Sports Medicine, arguing that post-exertional malaise in long COVID cannot be attributed to cardiac deconditioning and that the pathophysiology of exercise intolerance must be fundamentally reconsidered (Charlton et al. 2026). GET’s failure in ME/CFS reflects accurate biology, not patient non-compliance or psychological factors.

Patient-Reported Harms

CautionWarning: Patient-Reported Harms from GET

Large patient surveys consistently report high rates of harm from GET. A systematic review of 10 patient surveys across four countries found that 51% of respondents reported GET worsened their health, compared to 20% for CBT (Kindlon 2011). These figures underline why GET is no longer recommended by NICE and other major guidelines (National Institute for Health and Care Excellence 2021).

• 51% of ME/CFS patients across surveys reported GET worsened their condition (Kindlon 2011)
• Many report GET as triggering transition to more severe disease states
• Some report permanent functional decline attributable to GET programs
• Very few (\(<\) 10%) report sustained benefit

WarningLimitation: Fairness Note: The BPS/GET Position Is More Nuanced Than Presented Here

The biopsychosocial model’s strongest form does not claim ME/CFS is purely deconditioning or psychological — it proposes that perpetuating factors including deconditioning and activity avoidance interact with biological abnormalities. The original PACE protocol included therapist-guided, patient-negotiated activity targets with explicit instructions not to push beyond tolerated limits, which is more nuanced than the “progressive incremental increases” summary above. Pre-PACE trials (Fulcher and White 1997; Powell et al. 2001) showed benefit on subjective outcomes and some objective measures, and these findings deserve engagement rather than omission. The evidentiary standard applied here to PACE — demanding objective primary outcomes from a trial powered and designed for subjective ones — is stricter than the standard applied to many speculative frameworks elsewhere in this document, including Architecture C (certainty 0.50, with a key component resting on n=5 case data). This asymmetry should be acknowledged: describing PACE’s findings as “decisively refuted” while presenting Architecture C with careful hedging reflects the author’s biomedical perspective rather than a balanced evidence assessment, and readers should weigh both sides accordingly.

Recent Meta-Analyses Do Not Rehabilitate GET

A cluster of systematic reviews and meta-analyses published in 2025–2026 claim that exercise improves fatigue in ME/CFS. However, all three share the same structural limitations: they pool unblinded trials using subjective outcomes, include studies with broad diagnostic criteria that do not require PEM, and mix mechanistically different interventions (aerobic exercise, yoga, qigong, tai chi). Zhao et al. (2026, 17 RCTs, n=1,944) is representative: exercise improved subjective fatigue (SMD = 0.85) but produced no improvement in objective cardiopulmonary function or functional capacity—the predicted signature of blinding bias, not genuine therapeutic effect (see Blinding Failures in ME/CFS Treatment Research for detailed analysis). The BRANDO meta-epidemiological synthesis demonstrates that unblinding alone inflates patient-reported outcomes by SMD 0.56 on average, sufficient to explain most of the effect Zhao et al. report. These meta-analyses do not provide grounds for reconsidering the consensus against GET in PEM-positive ME/CFS patients.

1.3 The Comorbidity Exercise Conflict

Many ME/CFS patients have concurrent conditions for which exercise is a standard, evidence-based treatment: type 2 diabetes, cardiovascular disease, obesity, osteoporosis, hypermobility spectrum disorders (hEDS/HSD), and depression. This creates a genuine therapeutic conflict: guidelines for the comorbidity recommend progressive exercise, while ME/CFS pathophysiology makes such exercise harmful. Because most comorbidity-specific guidelines do not include ME/CFS exceptions, clinicians are left to reconcile contradictory recommendations without formal guidance.

The conflict is not hypothetical. In countries where ME/CFS recognition lags behind international consensus, patients continue to receive exercise-based rehabilitation prescriptions from physicians treating their comorbidities—unaware that the standard protocol may trigger post-exertional malaise, functional decline, and disease progression (Kindlon 2011)

Scope of the Conflict

The following comorbidities illustrate the tension:

• Type 2 diabetes and metabolic syndrome. Exercise is first-line therapy for insulin resistance, yet ME/CFS patients cannot tolerate the recommended 150min/week of moderate-intensity aerobic activity. Obesity compounds functional impairment in ME/CFS, and standard exercise-based weight management is contraindicated by PEM (Flores et al. 2013). Metabolomic evidence shows ME/CFS involves a distinct hypometabolic state with impaired pyruvate dehydrogenase function (Naviaux et al. 2016), fundamentally different from the metabolic profile of sedentary obesity.

• Cardiovascular disease. Standard cardiac rehabilitation relies on progressive aerobic exercise. However, ME/CFS patients frequently have reduced stroke volume, reduced end-diastolic volumes, and reduced cardiac mass on MRI, with a significant inverse relationship between cardiac output and PEM severity (Natelson, Brunjes, and Mancini 2021). Applying standard cardiac rehabilitation without accounting for post-exertional physiology risks worsening both the cardiac condition and ME/CFS.

• Hypermobility spectrum disorders (hEDS/HSD). Muscle strengthening is essential to stabilize hypermobile joints and reduce dislocations. Resistance training with body weight and resistance bands is feasible and effective in hEDS/HSD (Buryk-Iggers et al. 2022), and isometric exercises are particularly valuable because they build strength without joint movement (Zabriskie 2022). However, no published study addresses the dual hEDS+ME/CFS population, and even low-load resistance protocols must be titrated against PEM risk.

• Osteoporosis. Weight-bearing exercise is recommended to maintain bone density, but many ME/CFS patients—especially those who are housebound or bedbound—cannot perform such exercise. Prolonged immobility itself accelerates bone loss, creating a vicious cycle. Whole-body vibration (WBV) has been proposed as a potential alternative in the general osteoporosis literature because it can be applied without ambulation, but no ME/CFS-specific trials exist; a small pilot in mechanically ventilated ICU patients (n=19) demonstrated safety in a supine application but found significant increases in energy expenditure during WBV, which is concerning for ME/CFS patients with metabolic dysfunction (Wollersheim et al. 2017); this remains an unaddressed evidence gap.

• Depression. Exercise is a first-line non-pharmacological treatment for depression. However, in ME/CFS, exercise-based interventions for depression may exceed the patient’s energy limit and trigger characteristic PEM, undermining any antidepressant effect (Frontiers in Psychiatry Collaborative 2025). Pacing combined with psychotherapy (CBT adapted for ME/CFS, acceptance-based therapies) should be the default approach when depression is comorbid.

The CPET-Based Activity Prescription Framework

The resolution of the comorbidity exercise conflict lies in individualized, physiology-guided activity prescription rather than population-based exercise guidelines. The Workwell Foundation’s two-day cardiopulmonary exercise test (CPET) protocol provides the scientific basis for this approach (Stevens et al. 2018)

Davenport et al. (2010) proposed that all physical therapy for ME/CFS must center on the ventilatory anaerobic threshold (VAT) determined by individual CPET (Davenport et al. 2010). Activity must stay below the patient’s VAT heart rate at all times. This model applies equally when the goal is comorbidity management: the type of exercise may be dictated by the comorbidity (e.g., resistance training for hEDS, aerobic activity for diabetes), but the intensity ceiling is always set by the ME/CFS pathophysiology.

Critically, formula-derived heart rate targets (e.g., 220\(-\)age \(\\times\) 0.6) do not reliably predict the individually measured lactic acidosis threshold in ME/CFS patients. Van Campen et al. (2020) demonstrated bias ranging from \(-28\) to \(+23\)bpm between formula estimates and CPET-measured thresholds in 90 ME/CFS adults Campen, Rowe, and Visser (2020). Population formulas are therefore inadequate for safe activity prescription; individual CPET-derived thresholds are necessary.

Heart rate monitor (HRM)-guided pacing operationalises this approach in daily life. Clague-Baker et al. (2025) demonstrated that HRM-guided pacing is feasible and acceptable in both ME/CFS and long COVID: 89% of participants continued using their HRM at 8 weeks, and 66% at 6 months (Clague-Baker et al. 2025).

Severity-Stratified Approach

Gloeckl et al. (2024) proposed a three-tier framework for exercise in post-viral conditions with PEM that is directly applicable to ME/CFS comorbidity management (Gloeckl et al. 2024).

• No PEM (rare in ME/CFS): Standard comorbidity-specific exercise guidelines may apply, with heart rate monitoring as a safety net.

• Mild-to-moderate PEM: Modified exercise tailored to individual tolerance. The comorbidity dictates exercise type; the VAT heart rate sets the intensity ceiling. Short sessions (5–15min), sub-anaerobic intensity, with 48–72h monitoring for delayed PEM.

• Severe PEM: Pacing strategies only. No structured exercise program. Comorbidity management must rely on pharmacological and non-exercise interventions (e.g., insulin sensitizers for diabetes, bisphosphonates for osteoporosis, psychotherapy for depression).

A systematic review of physiotherapy evidence in ME/CFS reinforces this stratification: Wormgoor and Rodenburg (2021) found that any apparent treatment effect of exercise disappeared when stricter PEM-inclusive diagnostic criteria and objective outcome measures were applied (Wormgoor and Rodenburg 2021). This confirms that exercise interventions validated in broader “chronic fatigue” populations cannot be extrapolated to patients meeting full ME/CFS criteria with documented PEM.

Practical Adaptations by Comorbidity

When a comorbidity requires some form of physical activity and the patient falls into the mild-to-moderate PEM tier, the following adaptations apply:

NoteProposal: N5: Postural Optimization During Cognitive Work (0.50)

Section label: @sec-postural-splanchnic-optimization

Mechanism and Rationale.

Recumbent positioning during and after cognitive work reduces splanchnic blood pooling by eliminating orthostatic gravitational forces on abdominal and pelvic venous reservoirs. In severe ME/CFS patients with POTS or severe dysautonomia, this reduces competing demands on venous return during periods when cognitive exertion already increases cardiac output requirements.

Practical Implementation.

Protocol: During any cognitively demanding work period (reading, writing, studying, professional work), maintain recumbent positioning (lying flat with knees elevated) for at least 10–15 minutes afterward. For severe patients, maintain recumbent position during the entire cognitive work session if possible.

Expected Benefits.

• Reduced splanchnic blood pooling, freeing venous return for cerebral perfusion during cognitive work
• Minimized post-cognitive-exertion splanchnic hypoperfusion, reducing GI symptoms (bloating, cramping) after mental tasks
• Improved cerebral perfusion during extended cognitive sessions (reduced brain fog, improved focus)
• Potentially lower PEM risk from cognitive overexertion by optimizing hemodynamic allocation

Evidence and Certainty.

Certainty: 0.50. Mechanistically grounded in orthostatic physiology and venous return regulation (well-established). Direct clinical evidence in ME/CFS populations is lacking; the proposed mechanism extends from normal POTS physiology to the specific context of ME/CFS-associated splanchnic hypoperfusion during cognitive exertion. The intervention is low-risk, low-cost, and easily implementable, making it an attractive adjunct for severe patients with dysautonomia.

Testable Predictions.

1. ME/CFS patients practicing post-cognitive-work recumbent positioning will show reduced post-exertional GI symptoms (bloating, cramping, diarrhea) compared to patients maintaining upright posture.
2. Recumbent positioning during cognitive work will improve objective measures of cerebral perfusion (e.g., transcranial Doppler velocity, cerebral oxygenation via NIRS) compared to upright positioning.
3. Combined postural optimization + butyrate supplementation will show additive improvement in gut barrier function markers (zonulin, LPS, I-FABP) compared to butyrate alone.

Limitations.

• No randomized controlled trials in ME/CFS populations exist.
• Requires ergonomic adaptation for severe patients (workspace modifications, ergonomically designed beds with accessible workspace).
• May not be feasible for ambulatory patients in professional settings requiring upright posture.
• Individual variability in splanchnic blood pooling magnitude is likely (patients with more severe POTS comorbidity will derive greater benefit).

Clinical Recommendation.

Consider adding postural optimization (recumbent positioning during and after cognitive work) to activity modification strategies, particularly for severe patients with POTS comorbidity or documented post-cognitive-exertion GI symptom exacerbation. The intervention is low-risk, low-cost, and may synergize with butyrate-based gut optimization strategies.

Integrative Note.

Postural optimization addresses the hemodynamic component of gut barrier dysfunction (splanchnic blood pooling during exertion), while butyrate addresses the mucosal integrity component (tight junction support, epithelial repair). Combined, they provide complementary mechanisms for gut barrier optimization.

(Section label: @sec-postural-splanchnic-optimization)

Certainty: 0.50. Mechanistically grounded; direct ME/CFS evidence lacking; intervention is low-risk and potentially synergistic with butyrate-based gut optimization for severe patients with dysautonomia.

• For diabetes/metabolic syndrome: Substitute prolonged aerobic sessions with multiple daily micro-sessions (3–5min walking or recumbent cycling) below VAT heart rate. Prioritize dietary management and pharmacotherapy (metformin, GLP-1 agonists) as primary interventions, using activity as an adjunct only when tolerated.

• For cardiovascular rehabilitation: Replace standard progressive protocols with CPET-guided sub-anaerobic activity. Recumbent positions reduce orthostatic demand. Monitor both cardiac parameters and PEM onset; cardiac benefit should never be pursued at the cost of post-exertional deterioration.

• For hEDS/HSD: Focus on isometric contractions (5–10s holds at 30–50% maximal effort) and low-load resistance band work, which have lower cardiovascular demand than dynamic exercise (Zabriskie 2022). Perform in supine or seated position. Target specific joint stabilization rather than general fitness.

• For osteoporosis: When weight-bearing exercise is impossible, consider pharmacological bone protection (bisphosphonates, vitamin D, calcium). For patients who tolerate brief standing, supported standing exercises (wall-assisted) for 1–2min may provide some mechanical stimulus without exceeding energy limits.

• For depression: Replace exercise-based antidepressant strategies with adapted psychological interventions (acceptance and commitment therapy, mindfulness-based approaches) and pharmacotherapy. Gentle movement within the energy envelope and socially meaningful activity within pacing constraints may support mood without triggering PEM (Frontiers in Psychiatry Collaborative 2025), though evidence specific to this approach in ME/CFS depression remains limited.

NoteOpen Question: Exercise Adaptation for ME/CFS Comorbidities

No clinical trial of any design has evaluated exercise protocols specifically adapted for ME/CFS patients with concurrent conditions requiring physical activity (type 2 diabetes, cardiovascular disease, hEDS, osteoporosis, depression). The current evidence base rests on extrapolation from separate ME/CFS pacing research and comorbidity-specific exercise literature, applied by analogy rather than direct evidence. Prospective studies testing CPET-guided, PEM-monitored activity protocols in these specific dual-diagnosis populations are urgently needed.

1.4 Risk of Post-Exertional Malaise

Any movement carries PEM risk in ME/CFS, necessitating careful calibration:

• Dose-response relationship: Greater exertion produces worse PEM
• Individual variability: Thresholds vary widely (severe patients may crash from showering; mild patients tolerate gentle walks)
• Cumulative effects: Multiple small exertions may sum to trigger PEM
• Unpredictable triggers: Same activity may be tolerated one day but trigger PEM another day
• Delayed onset: 12–72 hour lag makes cause-effect connections difficult

2 Safe Movement Approaches

Despite exercise intolerance, complete immobility causes problems (muscle atrophy, joint stiffness, orthostatic intolerance worsening). The goal is movement within the energy envelope—enough to prevent deconditioning complications without triggering PEM.

2.1 Principles of Safe Movement

• Stay below anaerobic threshold: Use heart rate monitoring (AT \(-\) 10-15 bpm)
• Horizontal postures: Recumbent or supine exercise reduces orthostatic demand
• Constant body position: Avoid activities requiring positional transitions (standing → supine, turning, starting/stopping) — these transitions exhaust autonomic reserve in POTS+ME/CFS patients
• Short duration: 5–10 minute sessions may be tolerable where 20–30 minutes would crash
• Consistency over intensity: Very gentle daily movement better than intermittent harder sessions
• Immediate cessation: Stop at first signs of excessive exertion (heart rate elevation, breathlessness, fatigue)
• Monitor delayed effects: Track PEM onset 12–72 hours post-activity to calibrate appropriately

2.1.1 Constant Body Position as Primary PEM-Safety Criterion

ImportantHypothesis: Constant Body Position as Primary PEM-Safety Criterion for POTS+ME/CFS

Certainty: 0.50. Mechanistically grounded in orthostatic intolerance pathophysiology; not ME/CFS-validated by controlled trial.

Mechanistic Rationale: Constant body position eliminates the orthostatic demand fluctuations that drive autonomic compensation in ME/CFS/POTS. Upright-to-supine transitions require rapid blood redistribution (venous return adjustment, baroreflex-mediated HR changes), which exhausts autonomic reserve and depletes metabolic resources. Sports where body position remains constant throughout — swimming (horizontal), rowing (seated), recumbent cycling (reclined), and resistance band work (seated/supine) — reduce autonomic demand by maintaining stable hemodynamic parameters, thereby raising PEM threshold.

Cross-disease resonance: The Leeds POTS exercise protocol (Fu et al. 2010 (Fu et al. 2010), reviewed in (Fu and Levine 2018)) begins patients in recumbent positions (rowing, cycling) precisely because constant position reduces orthostatic stress, only progressing to upright activity after 12–16 weeks of recumbent training. Rowing is the protocol’s starting modality alongside recumbent cycling and swimming. However, the protocol was designed for POTS patients without ME/CFS, and community completion drops from 76 percent (research) to 41 percent (community) — meaning 59 percent of real-world POTS patients could not complete it; ME/CFS patients with additional post-exertional pathology may fare worse. This is not merely “resting” but actively avoiding a class of autonomic challenge that POTS patients cannot compensate for. See also Section Water Rowing as a POTS/ME/CFS-Tolerable Modality — Constant Position + Hydrostatic Pressure + Compound Distribution for land-based rowing as a home-accessible Tier 2 modality.

Practical application: The constant-position principle provides a simple filtering rule for patients and clinicians: If an activity requires getting up from the floor, transitioning from lying to standing or vice versa, or involves repeated start-stop positional changes, it carries elevated PEM risk for POTS+ME/CFS patients. Activities to prefer include:

• Stable supine: Isometric yoga, passive stretching, bed-based micro-movements
• Stable seated: Recumbent bike, seated rowing ergometer (land or water), resistance band exercises
• Stable horizontal in water: Swimming (backstroke), water rowing, water walking at constant depth

Activities where body position varies frequently — walking with stops/starts, circuit training, standard yoga with standing-to-floor transitions — impose orthostatic demand spikes that should be avoided in moderate-to-severe patients.

Evidence Link:

• Schmid et al. (2007): Water immersion decreased HR (vagal activation) while maintaining cardiac output (Schmid et al. 2007) — demonstrating hydrostatic preload benefit without positional change stress
• Oka et al. (2014, 2017): Seated and recumbent isometric formats were PEM-safe in CFS and feasible in severe ME/CFS (Oka, Wakita, and Kimura 2017) — no positional transitions during sessions
• Leeds POTS protocol: Recumbent-to-upright progression strategy validates positioning as exercise safety variable (Fu et al. 2010)
• Physiological rationale: Autonomic instability during orthostatic transitions is documented in 60–80% of ME/CFS patients (Section Neurally Mediated Hypotension)

Testable Prediction: In a crossover study comparing constant-position vs. varying-position activities at matched metabolic intensity (VO2, HR), ME/CFS patients will show: (a) lower PEM incidence (OR below 0.5) after constant-position activities; (b) reduced autonomic stress markers (catecholamines, HRV turbulence) during constant-position activities; (c) shorter recovery time to baseline biomarkers after constant-position activities.

Limitations:

• No randomized controlled trial directly testing constant-vs-varying position exercise in ME/CFS
• Individual variability in orthostatic tolerance means the constant-position benefit varies by patient severity
• Some aquatic activities (swimming turns, changes in stroke position) violate constant-position principle even in water
• Recommended activities are user-intuited from physiological rationale, not empirically validated in ME/CFS

2.1.2 Aquatic Exercise: Hydrostatic Pressure and Autonomic Modulation

Aquatic exercise occupies a unique position in ME/CFS movement planning because it combines several physiologically protective mechanisms in a single modality. The buoyancy of water reduces gravitational load to a fraction of body weight, seated or horizontal positions minimize orthostatic demand, and — most distinctively — hydrostatic pressure from water immersion produces autonomic shifts that may directly address the sympathovagal imbalance characterizing ME/CFS/POTS.

CautionSpeculation: Hydrostatic Pressure Augments Orthostatic Reserve via Parasympathetic Shift

Certainty: 0.68 — mechanism validated in cardiac populations (Schmid 2007, certainty 0.68); not yet tested as ME/CFS therapeutic mechanism.

Water immersion increases central venous pressure via hydrostatic compression of peripheral veins, shifting autonomic balance toward parasympathetic dominance (vagal activation overriding the Bainbridge reflex) (Schmid et al. 2007). In healthy and cardiac-impaired subjects, chest-deep thermoneutral water immersion produces:

• Increased central venous pressure → increased preload → increased stroke volume
• Decreased heart rate (vagal activation) of approximately 10 bpm
• Decreased peripheral vascular resistance (21–30%) — a sympatholytic effect
• Maintained or increased cardiac output despite bradycardia

For ME/CFS patients with documented sympathetic dominance, reduced vagal tone, and orthostatic intolerance (Section Neurally Mediated Hypotension), this parasympathetic shift during water immersion represents a potential therapeutic mechanism beyond simple buoyancy: the water itself may temporarily “recharge” autonomic capacity by augmenting preload and reducing compensatory sympathetic drive.

However, Hanna et al. (1993) demonstrated that populations with impaired cardiac/autonomic function may fail to fully utilize hydrostatic preload during upright exercise — in post-MI patients, cardiac output and stroke volume during upright cycling were no different in water vs. land despite enhanced resting preload (Hanna, Sheldahl, and Tristani 1993). This suggests that the hydrostatic benefit may be most reliable in horizontal positions (swimming supine, backstroke, supine floating) rather than upright exercise in water, consistent with the constant-body-position principle (Hypothesis Constant Body Position as Primary PEM-Safety Criterion for POTS+ME/CFS). Simply being in water is insufficient; the combination of water immersion + horizontal posture maximizes the benefit.

Falsifiable prediction: ME/CFS patients performing chest-deep aquatic exercise (thermoneutral 32°C, 10 minutes, gentle movement) should show: (a) immediate reduction in standing HR by ≥5 bpm post-immersion vs. no change after land exercise; (b) increased HF-HRV during and for ≥30 minutes after immersion; (c) reduced orthostatic symptoms on stand-test 1 hour post-aquatic exercise vs. post-land exercise.

Limitations: Testing is in cardiac populations, not ME/CFS. The autonomic physiology of ME/CFS differs from heart failure/MI. Some patients report subjective worsening during pool exercise — possibly due to temperature sensitivity, chlorine exposure, or the metabolic demand of moving against water resistance exceeding the hydrostatic benefit. The mechanism rests on a single direct study (Schmid 2007, n=30, cardiac populations) plus Hanna 1993 (n=15, post-MI). Not yet replicated in ME/CFS.

2.1.3 Thermoneutral Water Temperature Is Critical

CautionSpeculation: Thermoneutral Water Temperature (31–32°C) as Prerequisite for Safe Aquatic Exercise in ME/CFS

Certainty: 0.55. Mechanistic inference from Schmid et al. (2007) and thermoregulatory dysfunction literature; no ME/CFS temperature-comparison trial.

ME/CFS patients exhibit impaired thermoregulation and heightened sympathetic response to temperature stress (Section Neurally Mediated Hypotension). Water immersion below thermoneutral (28°C) triggers vasoconstriction and sympathetic activation to conserve heat, counteracting the beneficial parasympathetic shift from hydrostatic pressure. Water above thermoneutral (35°C+) triggers vasodilation that can produce hypotension in dysautonomic patients, especially those with POTS.

Schmid et al. (2007) conducted all aquatic testing at thermoneutral 32°C, explicitly citing the importance of avoiding thermal autonomic stress in compromised patients (Schmid et al. 2007). This is consistent with cardiac rehabilitation guidelines specifying 31–33°C water for patients with cardiovascular impairment. The same principle applies to ME/CFS patients whose autonomic systems are already operating at their compensation limit.

Practical guidance:

• Target pool temperature: 31–32°C (thermoneutral). Verify with personal pool thermometer before entry
• Cold pools (below 29°C): Likely to induce vasoconstriction, tachycardia, and sympathetic activation — counterproductive
• Warm pools (>34°C): May induce vasodilation, hypotension, lightheadedness, and post-exercise orthostatic intolerance
• If only non-thermoneutral pools are available: Restrict sessions to 5–8 minutes and monitor orthostatic symptoms vigilantly; brief exposure may be tolerable where extended sessions are not
• Post-pool warming: Have towels, warm clothing, and a dry robe ready to prevent post-swim chill, which triggers vasoconstriction and may prolong post-exercise sympathetic activation

Falsifiable prediction: In ME/CFS/POTS patients performing aquatic exercise at matched sub-AT intensity: cold water (28°C) will show increased HR and catecholamines; warm water (35°C) will show increased lightheadedness and orthostatic symptoms; thermoneutral (32°C) will show optimal parasympathetic shift and lowest PEM incidence (crossover design, n≥20).

Limitations: No ME/CFS-specific temperature-comparison data. Individual temperature sensitivity varies widely. The optimal temperature window may be narrower for some patients. Some patients with predominant heat intolerance may prefer water slightly below thermoneutral, while cold-intolerant patients may prefer slightly above — the 31–32°C range is a starting recommendation, not a rigid absolute.

2.1.4 Water Rowing: Constant-Position Compound Movement with Hydrostatic Benefit

CautionSpeculation: Water Rowing as a POTS/ME/CFS-Tolerable Modality — Constant Position + Hydrostatic Pressure + Compound Distribution

Certainty: 0.30. Mechanistic inference from component physiology; zero ME/CFS trial data. Extrapolation from constant-position principle + Schmid 2007 hydrostatic evidence + Oka 2014 isometric safety data. Not yet replicated.

Water rowing on a seated ergometer (Concept2-style rowing machine adapted for pool use, or a rowing shell in water) combines three protective mechanisms in a single activity: (1) seated constant body position — no positional transitions, no orthostatic demand spikes; (2) hydrostatic pressure — augmented venous return and parasympathetic tone from chest-deep immersion; (3) compound upper+lower body movement — the metabolic load is distributed across large muscle groups, potentially reducing peak local demand and regional ischemia risk compared to isolated movements (see Hypothesis Compound Movements May Produce Lower HR Response Than Isolated Movements at Equivalent Perceived Exertion in ME/CFS). Rowing’s rhythmic, reciprocal pattern (pull–recovery cycle) may additionally entrain breathing and HRV.

Unlike swimming, water rowing keeps the face above water, avoiding the breath-holding demands and cervical stress that some ME/CFS patients cannot tolerate. The seated position also accommodates patients with hypermobility whose shoulders may be unstable in overhead swimming strokes.

Practical protocol outline:

• Intensity: 5–10 strokes/minute (ultra-low), RPE ≤10, HR ≤85% estimated AT
• Duration start: 2 minutes/session, 2 sessions/week
• Progression: Increase by 1 stroke/minute or 1 minute every 2–3 weeks, only if zero PEM for 2 consecutive sessions
• Form: Gentle pull with focus on smooth technique, not power. Recover slowly (3–4 second recovery). Avoid explosive drive phase
• Water temperature: Thermoneutral (31–32°C); see Section Hydrostatic Pressure Augments Orthostatic Reserve via Parasympathetic Shift
• Monitoring: Continuous HR, symptom tracking at 6h, 24h, 48h post-session
• Post-rowing rest: 5 minutes supine rest for blood redistribution (Section Orthostatic-Demand-Based Exercise Modality Classification for ME/CFS/POTS)
• Rollback rule: If PEM occurs at any stage, return to previous level for 2 weeks before re-attempting progression
• Equipment: Standard water rowing ergometer or pool rowing machine; resistance provided by water drag rather than air flywheel, creating smooth accommodating resistance

Falsifiable prediction: In ME/CFS patients, low-intensity water rowing (≤10 strokes/min, RPE ≤10, HR ≤85% AT) will show: lower HR response per watt of power output vs. land rowing (hydrostatic preload effect); lower PEM incidence (OR below 0.7) vs. matched-intensity land cycling over 4 weeks; improved orthostatic tolerance on stand-test 24h post-rowing vs. baseline.

Limitations: Zero ME/CFS water rowing data — this protocol is constructed entirely from physiological first principles. Water rowing ergometers are rare outside specialized rehabilitation facilities. Transport to pool facilities may itself trigger PEM for moderate-to-severe patients. The compound metabolic demand of rowing may exceed some patients’ capacity even at minimal intensity. Rowing is a learned motor skill; technique errors could increase workload beyond intended intensity. Completely untested in ME/CFS.

2.1.5 Land-Based Rowing: Seated Constant-Position Compound Movement Without Water Access

CautionSpeculation: Land-Based Rowing as a Tier 2 Safe Modality for POTS+ME/CFS — Leeds Protocol Precedent, No Hydrostatic Component

Certainty: 0.40. (0.45→0.40: downgraded per adversarial review — 59 percent community dropout rate for the Dallas protocol in POTS patients without ME/CFS constrains extrapolation; no ME/CFS-specific rowing data. The protocol provides precedent from an adjacent condition, not validation for ME/CFS.)

Land-based rowing on an indoor water-resistance rowing machine retains the seated constant body position and compound movement distribution of water rowing while losing the hydrostatic pressure benefit. It gains home accessibility — eliminating the pool transport barrier that consumes energy envelope. The Dallas POTS protocol uses seated rowing alongside recumbent cycling and swimming as a starting modality, with documented outcomes of +8 percent peak VO2, +12 percent cardiac size, and +7 percent blood volume (76 percent research completion, 41 percent community completion (Fu and Levine 2018)). In the ME/CFS adaptation, the protocol’s structure validates rowing as the starting tier; the ME/CFS adaptation retains rowing as the maintenance tier permanently.

Protocol: 5–10 strokes/min, minimum water resistance, 2 min/session start, HR below 85 percent estimated AT, mandatory 5 min supine rest post-session, rollback on PEM. Water-resistance machines (WaterRower) may offer advantages over air flywheels — smoother force curve, reduced explosive motor unit recruitment, less sensory noise — but no comparative trial exists, making these theoretical inferences only.

Falsifiable prediction: Land rowing (home-based, water-resistance, sub-AT) will show PEM incidence below 30 percent in mild-to-moderate POTS+ME/CFS over 8 weeks, non-inferior to recumbent cycling and substantially better than upright walking. PEM incidence above 30 percent would refute the hypothesis at any severity level. The 30 percent threshold is a pre-specified refutation criterion — not a contraindication-defining boundary. If PEM incidence cannot be distinguished from recumbent cycling (non-inferiority margin ≤5 percentage points), the modality adds no benefit over existing alternatives.

Limitations: Zero ME/CFS rowing data. Economic access barrier (800–1500 euros). Rowing is a learned motor skill; technique errors can increase workload without patient awareness. Intensity floor may exceed capacity for severe patients. Water-vs-air resistance advantages are theoretical only.

2.1.6 Land Rowing: Stroke Rate as Continuous PEM-Threshold Calibration Dial

CautionSpeculation: Rowing Stroke Rate as Precision Calibration Dial for Individual PEM Threshold

Certainty: 0.25. (Reduced from 0.40 after literature review.) The power-stroke rate relationship is characterized only at ≥20 spm: Held et al. (2020) measured 245–369±23 W across 20–45 spm on a RowPerfect3 ergometer (Held, Siebert, and Donath 2020); Treff et al. (2022) validated Concept2 PM5 accuracy at 120–400 W but found unvalidated accuracy below ~30 W with start-stroke errors of 10–70% (Treff et al. 2022). Critically, zero published data exist at 5–10 spm. The entire ultra-low-stroke-rate protocol is an extrapolation across a data gap spanning 15 spm (from 20 spm, the lowest measured rate, down to the recommended 5 spm). Martindale & Robertson (1984) established that flywheel momentum resets between strokes at rates below approximately 12–14 spm (Martindale and Robertson 1984) — below this rate, each stroke is a “start from standstill” requiring substantially higher per-stroke force than continuous rowing (Sanderson and Martindale 1986). Hofmijster et al. (2009) found gross mechanical efficiency is not systematically affected by stroke rate at 20–32 spm (Hofmijster, Van Soest, and De Koning 2009) — but this provides zero information about efficiency at 5 spm where the stretch-shortening cycle is eliminated. Not yet replicated.

Rowing power output scales with stroke rate at ≥20 spm: Held et al. (2020) measured 245±13 W at 20 spm to 369±23 W at 45 spm on a RowPerfect3 (RP3) ergometer (Held, Siebert, and Donath 2020). At the ultra-low stroke rates (5–10 spm) used in the ME/CFS protocol, no data exist — this entire range is extrapolation. The lowest validated power output on a Concept2 PM5 monitor is approximately 30 W; below this, the PM5 exhibits start-stroke errors of 10–70% (Treff et al. 2022). Whether the minimum achievable power output of standard rowing ergometers (~30 W at 10 spm on minimum damper setting (Boyas et al. 2006)) is below the PEM threshold for any ME/CFS severity level is unknown.

Critical biomechanical caveat at ultra-low rates: Martindale & Robertson (1984) demonstrated that flywheel momentum resets between strokes at rates below ~12–14 spm (Martindale and Robertson 1984). Below this rate, each stroke accelerates the flywheel from near-standstill — per-stroke force is higher than at continuous rowing where the flywheel maintains momentum. Sanderson & Martindale (1986) confirmed that the stretch-shortening cycle (elastic energy recovery in the drive phase) is eliminated at full-stop rates (Sanderson and Martindale 1986). The total work per minute drops with lower rate (cardiac benefit), but per-stroke muscular effort increases (orthopaedic risk) — a trade-off that may be unfavourable for ME/CFS patients whose primary limitation is metabolic, not cardiovascular. Kane et al. (2013) showed damper setting interacts with stroke rate: at low rates on high damper, the increased resistance further amplifies per-stroke force (Kane et al. 2013). Patients should use the lowest possible damper setting (damper 1) and minimise the per-stroke force by focusing on stroke length (reach) rather than stroke power (force). Hofmijster et al. (2009) found that gross mechanical efficiency (~19%) is not systematically affected by stroke rate at 20–32 spm (Hofmijster, Van Soest, and De Koning 2009) — but this provides no information about efficiency at 5 spm where the biomechanics are qualitatively different (stop-start vs. continuous).

The concept extends generic sub-threshold pacing (start low, increase gradually, monitor PEM, roll back) using a rowing-specific metric rather than constituting an independent treatment principle. PEM threshold varies day-to-day with sleep quality, hormonal cycle, illness status, and prior activity — a calibration performed every 2–3 weeks may be invalid on days when threshold shifts significantly. On such days, the “calibrated” rate may become an overexertion rate.

Falsifiable prediction: Patients following a rowing-specific sub-threshold titration protocol will show no superiority over recumbent cycling with matched titration parameters (start low, increase gradually, PEM-monitored rollback), contradicting the claim that rowing provides unique precision for therapeutic-window calibration.

Limitations: Zero bioenergetic data at 5–10 spm — the entire protocol operates in an unmeasured range. The lowest published data are at 20 spm (245 W, Held 2020); the gap to the recommended 5 spm starting point is 15 spm of biomechanical and metabolic extrapolation. Concept2 PM5 power accuracy unvalidated below ~30 W (Treff et al. 2022) — the monitor may display a number at 5 spm but it has no validated relationship to the patient’s actual metabolic work. Threshold day-to-day variability is clinically documented but not quantitatively characterized in ME/CFS. Flywheel reset dynamics (Martindale 1984) are known to create higher per-stroke force at stop-start rates — the net metabolic cost of high-force, low-frequency rowing vs. lower-force, higher-frequency rowing at equal power output is not characterized at these rates. Technique errors introduce further nonlinearity. WaterRower and Concept2 behaviour diverge at ultra-low rates (Kane 2013) — a protocol calibrated on one ergometer type cannot be directly transferred to the other. The stability assumption underlying any calibration approach is clinically dubious. Completely untested in any clinical population.

2.1.7 Rowing + Isometric Yoga Hybrid Protocol: Alternating Active and Recovery Modalities

CautionSpeculation: Hybrid Rowing-Isometric Protocol Permits Longer Session Duration Without PEM Than Either Modality Alone

Certainty: 0.40. Individual modality safety established (isometric yoga (Oka et al. 2014) cert 0.50, land rowing Land-Based Rowing as a Tier 2 Safe Modality for POTS+ME/CFS — Leeds Protocol Precedent, No Hydrostatic Component cert 0.40); zero hybrid protocol data in any clinical population. Concept from athletic active-recovery training: low-intensity activity between work bouts clears metabolites faster than passive rest.

Rationale. A rowing-only session concentrates metabolic demand in one movement pattern for the entire duration, potentially accumulating fatigue metabolites in specific muscle groups. Supine rest between bouts is effective but passive — it restores blood redistribution but does not actively clear local metabolites. Isometric yoga (Oka et al. 2014) engages different muscle groups at substantially lower intensity (20–30 percent perceived effort, no HR elevation >10 bpm above baseline), providing a different metabolic signal — one that has been shown to produce a parasympathetic-shift and anti-inflammatory profile (decreased cortisol, decreased TNF-alpha, increased DHEA-S, enhanced vagal tone, reduced HR) — without adding cardiovascular demand.

A hybrid protocol that alternates rowing (compound, cardiovascular, continuous) with isometric yoga (localized, isometric, parasympathetic) may distribute metabolic stress across different tissue compartments, prevent local metabolite accumulation in rowing-specific muscles, and use the isometric phases as active recovery that maintains muscle engagement and vagal tone. The total session can extend without PEM because no single metabolic pathway is continuously stressed.

Protocol outline:

• Session structure: 15 min total, three 5-min segments alternating: rowing → isometric yoga → rowing
• Rowing segments: 5–10 strokes/min, HR ≤85 percent estimated AT, minimum water resistance, 1 min supine rest between segments
• Isometric yoga segment (5 min): Seated or supine poses only. Gentle contraction at 20–30 percent perceived effort. No HR elevation >10 bpm above resting baseline. Example sequence: seated spinal twist (gentle, 30s hold alternating sides), supine bridge (modified, knees bent, 20s hold ×2), seated wall push-press (isometric, 20s hold), supine diaphragmatic breathing (remainder of segment)
• Transition: Rowing → isometric yoga within 60 seconds (rower to adjacent mat or chair). Isometric yoga → rowing within 60 seconds
• Progression: Increase rowing stroke rate OR isometric hold duration, never both simultaneously. Increase by +1 stroke/min or +5s hold every 2–3 weeks only if zero PEM for 2 consecutive sessions
• Post-session: Mandatory 5-min supine rest (Section Orthostatic-Demand-Based Exercise Modality Classification for ME/CFS/POTS)

Falsifiable prediction: In a three-arm trial (n=60 ME/CFS, 8 weeks, 3 sessions/week, all 15 min total), hybrid protocol vs. rowing-only vs. isometric-only will show: (a) hybrid produces lowest PEM incidence (primary endpoint: below 15 percent vs. 25–30 percent for single-modality arms), (b) hybrid permits highest total session duration before PEM-triggered termination, (c) hybrid shows superior 8-week FUNCAP-27 stability. Refuted if hybrid shows no PEM advantage over rowing-only (isometric yoga adds no synergistic benefit, hypothesis rejected regardless of isometric-only arm outcome). Also refuted if hybrid shows worse PEM than rowing-only (the isometric segment adds net metabolic cost without providing recovery benefit — active recovery hypothesis is contradicted by data).

Clinical significance. If validated, this protocol would be immediately actionable for patients who already have access to a rowing machine. No additional equipment is needed — a mat or adjacent chair for the isometric segment is sufficient. The cognitive load is manageable: the session has a fixed structure (row → isometric → row → rest) that does not require real-time decisions during exertion.

Limitations: Zero data on hybrid protocols in any clinical population. Active recovery is superior to passive recovery in trained athletes — the adaptation to ME/CFS, where the metabolic defect may prevent effective clearance regardless of recovery type, is untested. The protocol adds a modality transition (stand from rower, move to mat, return to rower) that introduces brief orthostatic demand — this may offset the isometric recovery benefit for POTS patients. The 5-min isometric segment may be insufficient duration for metabolite clearance or vagal recovery (mechanism unknown). Cognitive load of sequencing may exceed capacity for patients with cognitive PEM. Completely untested in ME/CFS.

2.1.8 Severity-Stratified Rowing Protocol: Matching Modality Intensity to Functional Capacity

CautionSpeculation: Severity-Stratified Starting Parameters Improve Tolerability Over Uniform Rowing Protocol

Certainty: 0.50. Severity variability is well-documented; the Leeds POTS protocol was not severity-stratified and produced 59 percent community attrition, suggesting uniform protocols fail across the functional spectrum. Zero ME/CFS severity-stratified exercise trial data.

Rationale. ME/CFS severity spans from mild (can work part-time) to very severe (bedbound, tube-fed). A uniform rowing protocol — the same starting stroke rate, session duration, and progression speed for all patients — will simultaneously under-stimulate mild patients (who could tolerate faster progression and would benefit from more conditioning) and over-stress severe patients (whose metabolic floor may be above even 5 strokes/min at minimum resistance). The same approach that the paper criticizes in GET — applying one protocol to a heterogeneous population — is replicated in rowing guidance.

A severity-stratified protocol matches four parameters to baseline functional capacity: (a) whether rowing is the primary modality or contraindicated, (b) starting stroke rate and session duration, (c) progression speed, and (d) session duration ceiling. The stratification uses consensus severity categories adapted from the paper’s existing severity framework.

Stratification table:

Severity	Modality	Start	Progression	Ceiling
Mild (part-time work)	Rowing primary	8–10 spm, 3 min	Weekly (+1 spm OR +1 min)	30 min/session
Moderate (homebound)	Rowing only	5–8 spm, 2 min	Biweekly (+1 every 2 wks)	15 min/session
Severe (room-bound)	Isometric primary; rowing trial	5 spm, 1 min	Monthly (+1 every 4 wks)	5 min/session
Very severe (bedbound)	Passive ROM only; rowing contraindicated	N/A	N/A	N/A

Application rules:

• Severity classification should use existing functional scales (FUNCAP-27, Bell Disability Scale) — not self-reported category alone
• The first rowing session at any severity level is a trial session: if PEM occurs at the starting intensity, the patient is reclassified one severity level down for rowing protocol purposes (regardless of global severity classification)
• Progression rules are identical to the general protocol: increase only one parameter per cycle, only if zero PEM for 2 consecutive sessions, roll back on any PEM
• A patient who stabilizes at a progression level for ≥6 months without PEM and with stable FUNCAP-27 scores is succeeding at fitness maintenance; further progression should not be attempted without objective evidence of envelope expansion

Falsifiable prediction: In a stratified implementation study (n=120 ME/CFS, 12 weeks), severity-stratified starting parameters vs. uniform protocol (all start at 5 spm, 2 min, biweekly progression) will show: (a) lower PEM incidence in moderate and severe arms (stratified advantage — severe/moderate patients start at appropriate intensity rather than uniform protocol’s over-stress), (b) higher session duration ceiling achieved in mild arm (stratified advantage — mild patients are not under-stimulated to the point of dropping out). The hypothesis is refuted if the stratified protocol shows no PEM advantage in any severity tier over uniform, or if the mild arm shows no higher ceiling (the mild-arm prediction is secondary; the moderate/severe PEM advantage is the primary refutation criterion).

Clinical significance. If validated, this protocol converts rowing from a single recommendation (“try 5 spm, 2 min”) into a decision support tool that patients and clinicians can use to match the intervention to the patient. The stratification table is designed to be printable in a form suitable for both home logs and clinic charts.

Limitations: The severity bins (mild, moderate, severe, very severe) are consensus categories, not empirically derived thresholds for rowing tolerance. The specific starting parameters (3 min vs. 2 min vs. 1 min) are reasoned estimates, not calibrated from patient data. The severity-to-modality mapping may misclassify patients whose rowing tolerance differs from their global severity (e.g., a moderate-severity patient with excellent seated orthostatic tolerance but severe cognitive PEM). The progression speeds (weekly/biweekly/monthly) assume linear tolerance accumulation, which is unvalidated. Zero ME/CFS severity-stratified exercise trial data.

2.1.9 Adapted Swimming for ME/CFS

CautionSpeculation: Adapted Swimming Protocol for ME/CFS — Backstroke, Thermoneutral, No Turns

Certainty: 0.35. Clinical trial registered (NCT07454395) — no results yet. Extrapolation from Schmid 2007 hydrostatic evidence + adapted protocol rationale. Not yet replicated.

Swimming provides best-in-class constant body position (horizontal, supine in backstroke) combined with hydrostatic pressure benefit. However, standard lap swimming introduces PEM risks: breath-holding during face-down strokes (Valsalva, CO2 swings), cervical stress from head-turning to breathe, and positional transitions at wall turns. An adapted protocol eliminates these triggers:

• Stroke selection: Backstroke only — face always out of water, eliminates breath-holding, cervical spine supported by water, minimal head movement
• No turns: Use pool width (not length) or walk-turns at wall; avoid flip turns and push-offs that involve rapid positional change
• Flotation support: Pool noodle or kickboard under the head/neck for patients with cervical instability or hypermobility (see Fairness Note: The BPS/GET Position Is More Nuanced Than Presented Here for hEDS considerations)
• Intensity: RPE 9–11, continuous HR monitoring (target below 85% estimated AT)
• Duration: 5–10 minutes maximum initially; 30-second supine rest at wall every 2–3 minutes
• Water temperature: 31–32°C thermoneutral (see Section Hydrostatic Pressure Augments Orthostatic Reserve via Parasympathetic Shift)
• Environment: Quiet pool, minimal crowding, consistent depth. No lap swimming crowds or noisy environments
• Post-swim: Mandatory 5 minutes supine rest for blood redistribution (Section Orthostatic-Demand-Based Exercise Modality Classification for ME/CFS/POTS)

Progressive protocol outline:

• Baseline (Weeks 1–2): Backstroke, 5 min/session, 2 sessions/week, pool noodle support, technique focus
• Build (Weeks 3–6): Backstroke, 8–10 min/session, 2–3 sessions/week, introduce 30-second rest intervals every 2 min
• Maintenance (Weeks 7–12): Backstroke, 10–15 min/session, 3 sessions/week maximum; do not increase beyond this without CPET confirmation

Falsifiable prediction: Backstroke at matched RPE will show lower HR response than freestyle swimming due to reduced Valsalva and cervical stress; adapted protocol will show PEM incidence below 20% in mild-to-moderate POTS+ME/CFS patients over 8 weeks.

Limitations: Zero published data on this adapted swimming protocol in ME/CFS. NCT07454395 pending results. Backstroke may be technically challenging for non-swimmers. Pool access is a significant socioeconomic and geographical barrier. Chlorine exposure may trigger mast cell-mediated symptoms in some patients (Connections to Allergies and Mast Cell Activation). Swimming competence is a prerequisite; non-swimmers should not attempt even adapted swimming without qualified instruction and supervision.

2.1.10 Resistance Bands: Controlled-Load Strengthening with PEM-Safe Characteristics

Resistance band exercises performed in seated or supine positions share PEM-protective properties with isometric training (Oka et al. 2014): submaximal intensity is tightly controlled via band resistance selection; concentric-eccentric phases are slow and controlled, avoiding explosive power demands; and bands provide accommodating resistance (increasing force at end-range) that matches natural strength curves, reducing excessive demand at vulnerable joint angles. Unlike free weights, bands eliminate gravitational stabilization demands that require additional energy expenditure.

For patients with hypermobility spectrum disorders (hEDS/HSD) who require muscle strengthening for joint stabilization (see Section Fairness Note: The BPS/GET Position Is More Nuanced Than Presented Here), resistance bands are a particularly suitable modality because they allow targeted strength work in mid-range positions without end-range loading that risks subluxation.

• Position. Seated or supine only. No standing exercises. Chair with back support.
• Intensity. 30% estimated 1RM. Resistance estimated by band color/cross-section and stretch distance. Start with lightest band; increase only if zero PEM for 4 consecutive sessions.
• Tempo. Very slow: 5-second concentric, 2-second hold, 5-second eccentric (5-2-5). The slow tempo avoids explosive power demands and allows continuous form monitoring.
• Volume. 2 sets × 8 reps for compound movements (seated chest press, seated row, overhead press). 3 sessions/week maximum, 10–15 minutes/session.
• Rest between sets. Mandatory 30 seconds supine rest for blood redistribution.
• Monitored HR. Target below 85% estimated AT during contractions.
• Compound preference. Prefer compound movements over isolation to distribute metabolic load across larger muscle mass.

CautionSpeculation: Resistance Bands Produce Lower HR Response Than Equivalent-Load Free Weights in ME/CFS

Certainty: 0.40. Mechanistic inference from reduced stabilization demand + slow-tempo physiology; no ME/CFS comparative trial. Not yet replicated.

Because resistance bands eliminate the gravitational stabilization demands that free-weight exercises impose (stabilizing a dumbbell against gravity while moving it through space requires continuous muscle co-contraction), band exercises should produce lower HR response and lower PEM incidence at equivalent external load. The very slow tempo (5-2-5) further reduces HR response vs. standard tempo (1-0-1, 2-0-2) by preventing rapid force development that necessitates explosive motor unit recruitment. This makes resistance bands the preferred loading modality for ME/CFS patients who need strength maintenance but cannot tolerate free-weight protocols.

Falsifiable prediction: Seated resistance band exercises (2 × 8 reps, 30% 1RM, 5-2-5 tempo) will show ≥5 bpm lower peak HR vs. equivalent-load free-weight exercises (match movement pattern, reps, and external load) in ME/CFS patients.

Limitations: No ME/CFS band vs. free-weight trial. Maximum resistance is limited by band availability; patients needing higher loads for progression may eventually require free weights or cable machines. Band tension is non-linear, making precise progressive overload more difficult to quantify than plate-loaded equipment. Movement pattern limitations — compound rowing and pressing are feasible with bands but some movements (squats, deadlift patterns) are difficult to replicate safely. Not yet replicated.

2.1.11 Modality Safety Classification by Orthostatic Demand

The combination of body position, hydrostatic pressure exposure, and compound vs. isolated movement pattern determines a modality’s orthostatic demand profile. The following taxonomy integrates these factors into a practical decision tool for clinicians and patients:

CautionSpeculation: Orthostatic-Demand-Based Exercise Modality Classification for ME/CFS/POTS

Certainty: 0.55. Mechanistic taxonomy anchored in POTS/orthostatic intolerance physiology; no validation study. Only the “recommended” tier has any ME/CFS evidence (isometric yoga, recumbent bike from Oka et al.).

Tier 1 — Lowest orthostatic demand (recommended for all severity levels):

• Isometric yoga (seated/supine) (Oka et al. 2014)
• Bed-based passive range-of-motion (caregiver-assisted) (Oka, Wakita, and Kimura 2017)
• Supine gentle stretching
• Supine diaphragmatic breathing

Tier 2 — Low orthostatic demand (constant position; no transitions):

• Recumbent bike (seated constant position)
• Water rowing (chest-deep, thermoneutral, seated) (see Section Thermoneutral Water Temperature (31–32°C) as Prerequisite for Safe Aquatic Exercise in ME/CFS)
• Resistance bands (seated/supine, slow tempo) (see Section Adapted Swimming Protocol for ME/CFS — Backstroke, Thermoneutral, No Turns)
• Adapted backstroke swimming (horizontal, no turns) (see Section Severity-Stratified Starting Parameters Improve Tolerability Over Uniform Rowing Protocol)

Tier 3 — Moderate orthostatic demand (constant position but upright):

• Upright cycling (seated constant position, orthostatic demand of upright posture)
• Seated walking (chair-based stepping)
• Supine-to-seated transitions during mat-based exercises

Tier 4 — High orthostatic demand (positional transitions):

• Walking (upright with movement, frequent start/stop at intersections, stairs)
• Standard yoga (standing-to-floor transitions, balance poses)
• Circuit training (positional changes between stations)
• Swimming with flip turns (rapid rotational movement, breath-holding)

Tier 5 — Contraindicated (PEM-incompatible):

• Graded exercise therapy protocols with progressive overload without continuous PEM monitoring
• Any activity exceeding individual anaerobic threshold
• Any activity producing PEM within 72 hours at any dosage level

Application rules:

• Patients select from their matching severity tier or below
• If a Tier 1 activity produces PEM, the patient should not progress to Tier 2
• Progression to next tier requires: ≥4 weeks tolerance of current tier, zero PEM for 2 consecutive weeks, documented functional stability
• POTS comorbidity: prefer Tier 1–2 (horizontal/constant position). POTS patients should avoid Tier 4 modalities without prior recumbent conditioning (analogous to the Leeds recumbent-to-upright progression, Fu et al. 2010)
• hEDS/HSD comorbidity: add isometric component emphasis within chosen tier; avoid end-range loading

Falsifiable prediction: Patients selecting modalities from their matched severity tier will show ≥50% lower PEM incidence vs. patients selecting modalities above their tier (trial: 12-week observational comparison, N=60).

Limitations: Tier boundaries are theoretically derived, not empirically calibrated. The taxonomy conflates orthostatic demand (physiological) with fall/intensity risk (safety) — these are separate dimensions that may not co-vary perfectly. No validation study exists. Individual variation in tolerance means some patients may exceed or be unable to reach the predicted tier. The taxonomy is a hypothesis, not a validated clinical tool.

2.1.12 Post-Exercise Horizontal Rest for Autonomic Recovery

CautionSpeculation: Mandatory Post-Exercise Horizontal Rest Reduces Delayed PEM via Blood Redistribution and Autonomic Stabilization

Certainty: 0.55. Extends the Klimas program’s supine-rest concept (Section Reduced-Breathing Protocol for Chronic Hypocapnia in ME/CFS) to all exercise modalities. Mechanistically grounded in orthostatic physiology; no controlled trial of post-exercise rest as an independent intervention variable.

Exercise, even at sub-AT intensity, produces transient blood redistribution away from splanchnic circulation and toward working muscles. In ME/CFS patients with global hypoperfusion and orthostatic intolerance, this redistribution may exceed compensatory capacity, contributing to PEM. Post-exercise horizontal rest (supine or semi-recumbent) maximizes blood return to splanchnic circulation and brain, accelerating recovery of perfusion balance and reducing the risk of delayed PEM.

The Klimas program (Ussher 2025) already embeds supine rest between activity bouts. This principle extends to all modalities: every exercise session should conclude with a mandatory 5–10 minute horizontal rest period, independent of whether the patient “feels recovered.” The rest is physiological protection, not symptom palliation.

Protocol:

• Timing: Immediately after any exercise session, transition to supine position within 1 minute
• Position: Supine, legs elevated 15–30 degrees (pillow under calves) to enhance venous return
• Duration: 5 minutes for sessions ≤10 minutes; 10 minutes for sessions 10–20 minutes
• During rest: Diaphragmatic breathing (6 breaths/min) to enhance parasympathetic activation
• Environment: Quiet, darkened room, minimal stimulation
• Completion criteria: HR within 5 bpm of baseline; no orthostatic symptoms; subjective “settled” sensation
• If criteria not met after 10 minutes: Extend rest by 5-minute increments to max 30 minutes. If criteria remain unmet, exercise intensity was too high for that session
• Documentation: Record rest duration needed to meet criteria — serves as an intensity calibration metric for subsequent sessions

Falsifiable prediction: Post-exercise horizontal rest will reduce time to HR baseline (≤5 min vs. 10–15 min without rest), reduce orthostatic symptom duration post-exercise, and reduce PEM incidence (OR below 0.5) in a within-subject crossover (4-week phases, N≥20).

Limitations: The post-exercise rest concept has not been isolated as an independent intervention variable. Benefit may derive from any rest (not specifically horizontal rest). Extending exercise sessions with mandatory rest components increases total time commitment, which may reduce adherence. No ME/CFS-specific trial of post-exercise rest as an independent PEM prevention strategy.

2.2 Gentle Stretching

• Supine or seated: Reduces cardiovascular demand
• Passive range of motion: Maintain joint mobility without resistance
• Avoid ballistic movements: Gentle, sustained stretches only
• Duration: 5–15 minutes may be tolerable
• Daily frequency: If tolerated, maintains flexibility

2.3 Isometric Exercises

Isometric (static muscle contraction without joint movement) may be better tolerated than dynamic exercise:

• Lower cardiovascular demand: Minimal heart rate elevation
• Maintain muscle strength: Prevents complete atrophy
• Short holds: 5–10 second contractions
• Submaximal intensity: Moderate contraction only (30–50% maximal)
• Examples: Wall sits (brief), plank holds (modified), leg presses against bed

2.4 Recumbent Activities

Horizontal or semi-reclined positions reduce orthostatic stress:

• Recumbent bike: Allows cardiovascular activity with lower orthostatic demand
• Supine leg movements: Gentle cycling motions while lying down
• Pool exercises: Buoyancy reduces gravitational stress; hydrostatic pressure augments venous return and parasympathetic tone (Schmid et al. 2007) (if tolerated; some patients worsen in water — see Section Constant Body Position as Primary PEM-Safety Criterion for POTS+ME/CFS for detailed protocol guidance and temperature requirements)
• Resistance bands while seated: Low-impact strength maintenance

2.4.1 Compound vs. Isolated Movement: HR Response in ME/CFS

CautionSpeculation: Compound Movements May Produce Lower HR Response Than Isolated Movements at Equivalent Perceived Exertion in ME/CFS

Certainty: 0.40. Novel extrapolation from global hypoperfusion and cardiac output limitation documented in ME/CFS; zero direct comparative data. Not yet replicated.

In healthy individuals, compound movements (multi-joint, multi-muscle) elicit higher HR than isolated movements because they recruit more total muscle mass. However, in ME/CFS patients with impaired perfusion redistribution and reduced cardiac output, this relationship may reverse: compound movements distribute metabolic load across larger muscle mass, reducing peak local demand and ischemic stress, while isolated movements (e.g., bicep curl) concentrate metabolic demand in small muscle groups, potentially exceeding local perfusion capacity and triggering compensatory sympathetic activation and disproportionate HR elevation.

This hypothesis challenges the intuitive assumption that isolation exercises are “easier” or “safer.” If correct, a seated row (compound: back, biceps, shoulders) may be physiologically less stressful at matched perceived exertion than a bicep curl (isolated: biceps only), because the metabolic demand per gram of active muscle tissue is lower in the compound movement, reducing regional ischemia and the sympathetic HR response that ischemia triggers.

Practical application: When prescribing resistance exercises for ME/CFS patients, prioritize compound movements (seated row, seated chest press, leg press) over isolation movements (bicep curl, tricep extension, leg curl) at equivalent intensity. This applies to resistance band, free-weight, and machine-based exercises alike.

Falsifiable prediction: In ME/CFS patients performing matched-RPE exercises: compound movement (seated row) will show ≥5 bpm lower HR vs. isolated movement (bicep curl); isolated movement will show greater regional deoxygenation on NIRS; compound movement will show lower PEM incidence over 4-week trial.

Limitations: Completely untested in ME/CFS — this is theoretical physiology applied to a novel population. The full hypothesis requires validation of the assumption that local perfusion failure, rather than total metabolic demand, is the dominant HR driver in ME/CFS. Compound movements require more motor coordination, increasing cognitive demand during exercise — this may offset the hypothesized perfusion benefit in patients with cognitive PEM sensitivity. Individual perfusion reserve varies, making individual response unpredictable without NIRS or invasive measurement.

2.5 Monitoring for PEM

Vigilant monitoring prevents inadvertent overexertion:

• Real-time heart rate: Stop if approaching threshold
• Perceived exertion: Use modified Borg scale; stop at first sense of effort
• Post-activity tracking: Log symptoms 12–72 hours after movement
• Adjust based on outcomes: If PEM occurs, reduce intensity/duration for subsequent sessions
• Recovery time: Allow full recovery (minimum 24–48 hours, often longer) between sessions

2.6 Adaptive Progression (If Tolerated)

For patients with stable mild-to-moderate ME/CFS who tolerate current activity levels without PEM:

• Very gradual increases: 1–2 minutes per week, or 1 additional repetition per week
• Sustained tolerance required: Maintain new level for 2–4 weeks before further increase
• Immediate rollback if PEM occurs: Return to previous tolerated level
• Never push through PEM: This worsens condition and should be avoided absolutely
• Realistic expectations: Goal is maintaining current function, not fitness improvement

CautionWarning: Exercise Precautions

Patients with severe ME/CFS (housebound or bedbound) should consult physicians before attempting any structured movement program. Even minimal exertion may trigger severe crashes in this population. For these patients, activities of daily living (personal hygiene, eating) may constitute maximal tolerable exertion, leaving no additional capacity for exercise.

2.6.1 Conceptual Framework: Fitness Maintenance vs. Fitness Improvement

ImportantHypothesis: Fitness Maintenance as a Distinct Goal from Fitness Improvement in ME/CFS

Certainty: 0.50. (0.60→0.50: Mancini/Natelson 2026 NIH-funded null replication brings CPET-specific evidence to equipoise — one positive study and one equivalently powered null. Incoming evidence certainty 0.60. Framework re-anchored to energy envelope theory (Jason, Muldowney, and Torres-Harding 2008), patient-reported harms from GET (Kindlon 2011), and the consistent finding of elevated RPE across ALL CPET studies — though RPE is a subjective self-report measure and should not be equated with the objective VO₂ decline the framework was originally designed to explain. The framework is conceptual, not empirically validated.)

Framework. Exercise physiology distinguishes two fundamental goals: fitness maintenance (preventing further loss from current baseline) and fitness improvement (building capacity above baseline). In healthy populations, these goals coexist — training produces both maintenance and improvement. In ME/CFS, they are fundamentally decoupled. The metabolic dysfunction that underlies PEM (Keller et al. 2024) means that any exercise protocol intended to produce adaptation (fitness improvement) instead produces physiological deterioration — the inverse of the intended effect. The only achievable goal is fitness maintenance: using carefully calibrated movement to prevent the deconditioning, atrophy, joint stiffness, and orthostatic deterioration that complete immobility would cause, without triggering the PEM cascade that would worsen the baseline.

Why “fitness improvement” is contraindicated. Two-day CPET evidence from the largest positive study (Keller et al. 2024, Canadian Consensus Criteria, n=84+71) demonstrates that a single maximal exertion produces 5–8% declines in VO₂peak, work output, and ventilation in ME/CFS, with recovery requiring 13+ days (Keller et al. 2024). A 2026 null replication by Mancini, Natelson et al. (Fukuda criteria, n=58+25) found no significant group-average decline (Mancini et al. 2026), but did find elevated perceived exertion at all workloads and lower maximal HR — corroborating that exercise imposes disproportionate physiological burden even when the group-average VO₂ signal is absent. Regardless of which CPET result is ultimately replicated, the clinical reality that exertion exceeding individual tolerance produces PEM and functional decline is supported by: (a) energy envelope theory, (b) 51% GET harm rates in patient surveys (Kindlon 2011), (c) elevated Borg RPE across ALL CPET studies including the null replication, and (d) chronotropic incompetence as a consistent finding. Every attempt to “increase fitness” through progressive overload risks decreasing functional capacity. Graded exercise therapy (GET) programs, built on the assumption that fitness improvement is possible and desirable, produced 51% harm rates in patient surveys precisely because they applied a fitness-improvement model to a physiology that cannot accommodate it (Kindlon 2011).

What “fitness” means in ME/CFS. In the energy envelope framework (Jason, Muldowney, and Torres-Harding 2008), fitness must be redefined as functional capacity within the envelope — not absolute VO2max, not maximal strength, not cardiovascular endurance in the conventional sense. The relevant metrics are:

• Activities of daily living (ADL) capacity: Can the patient shower, prepare meals, perform basic hygiene without PEM?
• Baseline function stability: Is the patient’s functional level stable over months, or declining?
• PEM threshold: At what exertion level does PEM onset occur, and is this threshold stable or rising?
• Deconditioning prevention: Are joint mobility, muscle mass, and orthostatic tolerance maintained without being actively increased?

A patient whose ADL capacity is steady over 6 months, whose PEM threshold is not declining, and who has not developed contractures or orthostatic worsening is succeeding at fitness maintenance — the highest achievable goal in ME/CFS. This is not “giving up” or “accepting illness limitation.” It is recognizing physiological reality and targeting the intervention to the achievable outcome.

Distinction from deconditioning prevention. Deconditioning prevention is the physical floor — the minimum movement needed to prevent joint contractures, muscle atrophy, and orthostatic deterioration. Fitness maintenance is the operational ceiling — the maximum sustainable activity level within the energy envelope that stabilizes function without triggering PEM. Movement above the envelope produces deterioration (PEM); movement below the deconditioning prevention floor produces gradual iatrogenic loss from immobility. The safe zone between them is narrow and individually calibrated.

Energy envelope expansion through pacing. Anecdotal patient reports describe that months of strict pacing can expand the energy envelope — not through training adaptation, but through removal of the repeated PEM insults that were cumulatively eroding baseline function. This is recovery of latent capacity (unmasking function that was there but suppressed by PEM-induced damage), not acquisition of new capacity (building fitness through training). The distinction is clinically important: a patient whose envelope expands through pacing should not interpret this as evidence that exercise training works. The causal mechanism is cessation of harm, not loading-induced adaptation.

Psychological dimension. The fitness-improvement paradigm is deeply embedded in sports culture, clinical rehabilitation, and the messages ME/CFS patients receive from providers, family, and their own internalized expectations. Accepting that fitness maintenance — not improvement — is the ceiling requires psychological adjustment. Acceptance and commitment therapy (ACT) frameworks that focus on living meaningfully within limitation rather than striving to overcome it are directly applicable to this reframing.

Testable prediction: A 12-week trial randomizing ME/CFS patients to (a) fitness-maintenance-targeted activity (pacing + deconditioning-prevention movement, with explicit expectation-setting that fitness improvement is not a goal) vs. (b) standard pacing advice should show equivalent or better functional stability in the fitness-maintenance group, reduced PEM episode frequency (because patients are not testing the improvement ceiling), and reduced psychological distress (because goal-behavior alignment reduces frustration). The mechanistic prediction is that the fitness-maintenance group experiences fewer PEM events, not that fitness improves.

Limitations: This is a conceptual framework, not an empirically tested model. The distinction between “recovery of latent capacity” and “acquisition of new capacity” cannot be experimentally distinguished in an individual patient without knowing their hypothetical pre-illness ceiling. “Fitness” redefinition is semantic/philosophical — useful for clinical communication but not subject to hypothesis testing. ACT acceptance approaches may not suit all patients; some find the framing of “accepting limitation” demoralizing rather than empowering. Not yet replicated as a structured protocol.

2.6.2 Fitness Maintenance Framework: Activities of Daily Living Capacity as Primary Outcome Metric

CautionSpeculation: ADL Capacity Stability as the Primary Fitness Outcome in ME/CFS

Certainty: 0.35. Extrapolation from existing energy envelope theory and FUNCAP validation; no prospective trial using ADL stability as primary endpoint. Not yet replicated.

If fitness improvement cannot be achieved and fitness maintenance is the ceiling, the natural outcome metric is ADL capacity stability — measured over months to years — rather than conventional fitness metrics (VO2max, strength, endurance). The FUNCAP-27 (Sommerfelt et al. 2024) already provides an 8-domain functional capacity assessment validated in ME/CFS. Serial FUNCAP-27 assessments at 3-month intervals can track whether a patient’s ADL capacity is stable, declining, or (in the rare patient whose latent capacity is recovering through pacing) slowly improving.

This reframing has practical clinical consequences:

• A patient whose FUNCAP scores are stable should be congratulated — they are succeeding at the highest achievable goal. Conventional rehab framing interprets stability as “no progress,” leading to frustration and potentially harmful attempts to “break through the plateau.”
• A patient whose FUNCAP scores are declining should be assessed for remediable causes of envelope shrinkage (new comorbidities, medication side effects, life stressors, overexertion pattern) rather than prescribed “more exercise to build capacity.”
• A patient whose FUNCAP scores are improving should be monitored to distinguish true envelope expansion (sustained improvement with no PEM) from temporary fluctuation (good week followed by crash). Improvement that is not sustained over 2 consecutive 3-month assessments is not envelope expansion.

Falsifiable prediction: A cohort study measuring serial FUNCAP-27 at 3-month intervals over 24 months should show that patients receiving fitness-maintenance-targeted counseling (explicit expectation of ADL stability as primary outcome) have non-inferior 24-month FUNCAP trajectory (difference ≤3 points, pre-specified non-inferiority margin) compared to patients receiving standard pacing advice, with lower rates of “boom-bust” fluctuation (operationally defined as ≥3 excursions of >10% from the patient’s baseline FUNCAP score within a 12-month period). If the stability-targeted group shows worse FUNCAP trajectory (>3 points below control) or equal/higher boom-bust rate, the hypothesis is refuted.

Limitations: FUNCAP-27 sensitivity to detect meaningful within-patient change over 3-month intervals is not established. Stability may reflect the measure’s insensitivity rather than true functional stability. The approach requires a conceptual shift from “improvement” to “stability” that patients, clinicians, and outcome researchers may resist. No trial data exists for this specific reframing.

2.6.3 Psychological Reframing: Acceptance of the Fitness-Maintenance Ceiling

CautionSpeculation: Explicit Fitness-Maintenance Goal-Setting Reduces Exercise-Related Psychological Distress in ME/CFS

Certainty: 0.20. Extrapolation from ACT literature and motivational psychology; no ME/CFS-specific trial. The mechanism is goal-behavior alignment — when achievable goals match prescribed behavior, distress is lower than when unachievable goals are pursued.

ME/CFS patients exist in a rehabilitation culture that equates “progress” with “improvement.” Physiotherapists chart progress toward increased capacity. Patients expect to “get stronger.” Family members ask “are you getting better?” When the achievable ceiling is stability, this mismatch between expectation and reality generates frustration, demoralization, and — importantly — may drive patients to exceed their energy envelope in pursuit of a progress that physiology cannot deliver. Explicitly naming “fitness maintenance” as the goal removes this conflict.

Protocol components:

• Expectation-setting at treatment initiation: “The goal of this movement program is to maintain your current level of function and prevent deconditioning complications. It cannot and should not attempt to build fitness.”
• Outcome reframing: “Stability is success. A flat FUNCAP trajectory is a good result.”
• Monitoring for goal-behavior mismatch: If the patient reports “trying harder” or “wanting to break through,” assess whether this is driving envelope overshoot.
• Integration with ACT: Acceptance and commitment therapy provides frameworks for living meaningfully within limitation, directly applicable to the fitness-maintenance acceptance challenge.

Falsifiable prediction (hierarchical): Patients randomized to fitness-maintenance goal-setting (explicit stability-as-success messaging) vs. standard pacing counseling will show: (1) primary endpoint — lower exercise-related distress at 12 weeks (measured by a validated exercise-avoidance/anxiety scale, ≥0.3 SD effect size); (2) secondary gate — PEM frequency must not increase by >20% vs. control (non-inferiority). If primary endpoint not met, hypothesis refuted regardless of secondary outcome. If primary met but PEM frequency increases >20%, hypothesis is refuted on safety grounds.

Limitations: This is an untested psychological intervention frame. Some patients may find “acceptance of limitation” more distressing than the pursuit of improvement, even if the pursuit is futile. Cultural context matters — the fitness-improvement narrative may be more or less dominant in different healthcare systems. No ME/CFS-specific trial of this framing exists.

2.6.4 Cross-Disease Maintenance Protocols: Lessons from COPD and Heart Failure

CautionSpeculation: COPD and Heart Failure Maintenance Protocols Provide a Transferable Framework for ME/CFS Activity Management

Certainty: 0.50. Cross-disease analogy is strong: both COPD and HF feature pathophysiological exercise intolerance (ventilatory limitation in COPD, cardiac output limitation in HF), and both have developed explicit maintenance-phase protocols distinct from conditioning phases. No ME/CFS-adapted protocol exists.

Rationale. COPD pulmonary rehabilitation and heart failure exercise programs share a critical feature with ME/CFS that is absent in healthy-person exercise: exercise intolerance is pathophysiological, not behavioral. In COPD, ventilatory limitation prevents sustained high-intensity work; in HF, reduced cardiac output caps systemic oxygen delivery. In both fields, the recognition that capacity cannot always be increased has led to standardized maintenance protocols:

• COPD: Puhan et al. (2011) demonstrated that a 12-month maintenance program (unsupervised home-based interval training at 60% peak work rate, with monthly phone coaching) preserved the gains from initial pulmonary rehabilitation better than standard care. The protocol explicitly framed maintenance — not further improvement — as the target.
• Heart Failure: The HF-ACTION trial (O’Connor et al. 2009, n=2331) established that symptom-limited steady-state exercise — not progressive overload — stabilizes functional capacity in HFrEF, with maintenance sustained at 3-year follow-up regardless of whether patients increased training volume. The Taylor et al. (2014) Cochrane review of exercise-based cardiac rehabilitation confirmed that maintenance-phase programs preserve functional gains without requiring intensity progression.
• ICU/Acute Illness Recovery: Post-ICU recovery protocols explicitly distinguish the stabilization phase (maintaining function gained during inpatient rehab) from the recovery phase (active improvement), recognizing that many patients plateau at a level below their pre-illness baseline.

The structural elements of these protocols — interval-based rather than continuous exertion, submaximal rather than progressive intensity, scheduled rather than responsive rest, and maintenance-explicit rather than improvement-implicit framing — are directly transferable to ME/CFS, with the substitution of ME/CFS-specific limits (HR below AT, PEM tracking as the primary safety endpoint, severity-adjusted starting intensity).

Adaptation for ME/CFS. The protocol transfer is structural, not prescriptive: adopt the protocol architecture of COPD/HF maintenance (interval structure, submaximal ceiling, maintenance framing, objective monitoring) and calibrate the parameters to ME/CFS physiology:

• Interval structure: Activity bouts of 1–3 minutes followed by 3–5 minutes of rest (recumbent), substituting the COPD work/rest ratio (1:2 to 1:3) for ME/CFS-specific limits
• Submaximal ceiling: Heart rate strictly below AT –10 bpm (rather than %peak work rate as used in COPD), with PEM tracking as the primary safety signal
• Maintenance framing: Explicit expectation-setting that the goal is functional stability, not capacity gain — borrowed directly from COPD/HF maintenance-phase communication strategies
• Objective monitoring: Continuous HR monitoring (wearable) + symptom logging at 6h, 24h, 48h post-session, analogous to the monitoring protocols used in post-ICU rehabilitation

Falsifiable prediction. ME/CFS patients randomized to a COPD-adapted maintenance protocol (interval-based, HR-monitored, maintenance-framed) vs. standard pacing will show: (1) primary — non-inferior 12-month FUNCAP stability (difference ≤3 points); (2) secondary — ≥30% reduction in PEM frequency vs. control. The hypothesis is refuted if the protocol group shows ≥5-point FUNCAP decline vs. control, or if PEM frequency is not significantly different (regardless of FUNCAP outcome). Two mechanistic endpoints distinguish the components: % of monitored time spent above AT tests the interval-mechanism hypothesis; self-reported envelope-overshoot episodes test the framing-mechanism hypothesis.

Limitations. The cross-disease analogy is not validated for ME/CFS. COPD and HF pathophysiology differs fundamentally from ME/CFS — the transferability of protocol architecture does not guarantee transferability of benefit. The protocol still represents active exertion that may exceed capacity in moderate-to-severe patients. No ME/CFS-adapted protocol has been trialed. The maintenance protocols in COPD/HF research still include moderate exertion levels (60–70% peak work rate) that substantially exceed safe ME/CFS thresholds — the parameters must be calibrated downward, potentially to the point where the protocol is indistinguishable from standard pacing.

2.6.5 Bifurcation Model of Exercise Dose-Response in ME/CFS

CautionSpeculation: Exercise Dose-Response in ME/CFS Is Bifurcated: Maintenance Below Threshold, Deterioration Above

Certainty: 0.40. The concept of a bifurcated dose-response curve is grounded in two-day CPET evidence (contested: Keller 2024 positive vs. (Mancini et al. 2026) null) but no formal mathematical model exists for this pattern in ME/CFS. Not yet replicated as a testable model.

Rationale. Standard exercise physiology assumes a monotonic dose-response curve: more exercise → more adaptation, with a saturation plateau at high doses. The GET model was built on this assumption. Evidence suggests this assumption is inverted in ME/CFS: exercise above a threshold level does not produce adaptation — it can produce deterioration. Two-day CPET findings are contested: Keller et al. (2024, n=84+71, CCC) found 5–8% group-average declines in VO₂peak and doubling of severe impairment on Day 2 (Keller et al. 2024); Mancini, Natelson et al. (2026, n=58+25, Fukuda) found no significant group-average decline (Mancini et al. 2026). Both studies found elevated perceived exertion (Borg RPE) at all workloads in ME/CFS and lower maximal HR (chronotropic incompetence). The bifurcation framework remains valid: regardless of whether deterioration appears as a group-average VO₂ signal, the clinical reality — consistently elevated RPE, lower HR reserve, and individual PEM following exceeding threshold exertion — supports a model where crossing the individual threshold yields net deterioration rather than net adaptation.

This can be formalized as a piecewise function where the sign of the rate of functional capacity change (\(d F\text{/}d t\)) depends on exertion level relative to the individual’s PEM threshold:

• \(d F\text{/}d t \approx \text{maintenance\\_rate}\), when exertion \(\leq\) threshold (net benefit: movement prevents atrophy)
• \(d F\text{/}d t \approx \text{deterioration\\_rate}\), when exertion \(>\) threshold (net harm: PEM produces functional loss)

where F = functional capacity, threshold = the individual’s PEM-triggering exertion level (approximated by anaerobic threshold), maintenance_rate is preventing deconditioning loss (slow decline), and deterioration_rate reflects PEM-induced damage (rapid decline).

• Below threshold: Movement prevents atrophy, maintains joint mobility, preserves orthostatic tolerance — the maintenance_rate is small but positive, reflecting the prevention of disuse loss. The dose-response slope is shallow and net-positive.
• Above threshold: Movement triggers PEM — the deterioration_rate is larger in magnitude and net-negative, because PEM produces more functional loss than the movement prevents. The dose-response slope flips sign.
• The bifurcation point: The AT (or PEM threshold) is the critical value where dF/dt changes sign. This is not a saturation plateau (as in healthy exercise) — it is a structural regime change from maintenance to damage.

Why this matters. The bifurcation model explains why GET fails — it was built on a monotonic positive dose-response assumption that does not hold above the PEM threshold. It explains why pacing works — it keeps exertion below the bifurcation point, in the net-positive maintenance regime. It provides a mathematical justification for HR-based pacing thresholds. And it suggests a clinical protocol: identify the individual bifurcation point (ideally via CPET-derived AT, or via week-by-week PEM threshold tracking) and calibrate all activity to remain below it.

Relationship to existing frameworks. This model is a mathematical formalization of the energy envelope theory (Jason, Muldowney, and Torres-Harding 2008): the envelope is the region below the bifurcation point. The “deconditioning prevention floor” vs. “PEM ceiling” distinction described in the fitness-maintenance hypothesis Fitness Maintenance as a Distinct Goal from Fitness Improvement in ME/CFS maps directly to the two regimes. The model does not add new physiology — it formalizes existing clinical knowledge in a testable mathematical structure.

Testable predictions. 1. A piecewise model (maintenance below AT, deterioration above AT) will better fit longitudinal FUNCAP-27 trajectories (R² improvement ≥0.10) than a linear dose-response model, when tested with continuous wearable HR + activity data over 24 weeks in n≥30 ME/CFS patients. 2. The individually calibrated bifurcation point (AT or PEM threshold) will predict change in FUNCAP score at 6-month follow-up: patients spending >5% of monitored time above threshold will show significant FUNCAP decline. 3. The bifurcation point is lower (more restrictive) in severe vs. mild patients, consistent with 2-day CPET severity-stratification data.

Limitations. The model is conceptual. AT as the bifurcation point is an approximation — PEM may have additional determinants (cognitive load, emotional stress, cumulative sub-threshold overexertion) not captured by a single HR threshold. The piecewise function assumes a sharp bifurcation, but the transition from maintenance to deterioration may be gradual (a sigmoid rather than a step function). Individual calibration requires CPET, which many patients cannot access or tolerate. The model has not been fit to longitudinal data. \(d F\text{/}d t\) magnitudes are unknown for any severity level.] {#spec-bifurcation-model}

2.6.6 Muscle Mass and Joint Preservation in Severe/Bedbound Patients

For severe and bedbound ME/CFS patients, the safe movement approaches described above — even recumbent or aquatic activities — may exceed capacity. In this population, the goal is not movement for deconditioning prevention but passive preservation: preventing the muscle atrophy, bone density loss, contractures, and skin breakdown that prolonged immobility causes, using interventions that consume minimal or zero patient metabolic effort.

CautionSpeculation: Passive Range-of-Motion Protocols Prevent Contractures Without Triggering PEM

Certainty: 0.40. Mechanistic inference from ICU/space medicine deconditioning literature; no ME/CFS-specific trial. Not yet replicated. The Cochrane review of passive stretching for contracture prevention found insufficient evidence to establish effectiveness (Katalinic et al. 2010) — the positive ICU studies (Morris et al. 2008) (Bao et al. 2022) reported feasibility/safety and combined PROM with active components, not pure PROM efficacy.

In ICU patients immobilized for >7 days and in astronauts during microgravity exposure, passive range-of-motion (PROM) exercises performed by a caregiver demonstrably reduce joint contracture development and muscle atrophy compared to no intervention (Morris et al. 2008) (Bao et al. 2022). Large ICU studies confirm that protocolized PROM and early mobility are safe and feasible even in mechanically ventilated patients, with no increase in adverse events (Morris et al. 2008). The energy expenditure cost of PROM is borne by the caregiver, not the patient — the patient’s muscles are not actively contracting, so the metabolic demand that triggers PEM is absent. The risk is not metabolic but positional: incorrect positioning, excessive range, or overly rapid movement can cause pain, subluxation (in hypermobile patients), or autonomic activation.

Protocol for severe/bedbound ME/CFS:

• Caregiver-administered only. Patient remains fully passive
• Position. Supine throughout. No head-of-bed elevation during PROM unless patient already tolerates that position
• Joints targeted. Shoulders, elbows, wrists, hips, knees, ankles — one joint at a time, one limb at a time, 5 repetitions per joint
• Range. Pain-free range only. Never stretch into discomfort. For hypermobile patients: mid-range only, avoid end-range loading
• Tempo. Very slow: 5-second movement through range, brief pause at endpoint
• Duration. 5–8 minutes per session, 2–3 times per week initially
• Monitoring. Symptom tracking at 6h, 24h, 48h post-session. If PEM occurs, reduce to 1 session/week with half the repetitions
• Contraindications. Active PEM, acute pain episode, suspected joint subluxation, elevated resting HR above individual baseline

Falsifiable prediction: Severe/bedbound ME/CFS patients receiving caregiver-administered PROM (2–3 sessions/week for 12 weeks) will show slower joint range-of-motion decline (goniometry) and lower contracture incidence compared to matched bedbound controls without PROM, without increase in PEM frequency.

Limitations: Not tested in ME/CFS; extrapolation from critical care and space medicine populations. Even passive movement may trigger PEM in very severe patients if the proprioceptive input or autonomic response to limb manipulation exceeds tolerance. Caregiver availability is a significant access barrier. Individual baseline flexibility and hypermobility status must guide range limits.

CautionSpeculation: Nutritional Anti-Catabolic Strategies Slow Muscle Wasting in Bedbound ME/CFS

Certainty: 0.35. (0.30→0.35: HMB bed rest RCT externally validates the anti-catabolic principle; general-population evidence, mechanistic, incoming certainty 0.60). Mechanistic inference from sarcopenia, ICU, and space medicine literature; no ME/CFS-specific trial. Not yet replicated in ME/CFS.

When the musculoskeletal system receives no loading stimulus (complete bed rest), muscle protein breakdown exceeds synthesis within days, producing measurable atrophy. Nutritional strategies that have shown benefit in other immobilization contexts — ICU early nutrition, spaceflight countermeasure programs, sarcopenia management — may partially offset this catabolic drive in bedbound ME/CFS. The following strategies are low-risk, do not require patient exertion, and have mechanistic rationale:

• Protein adequacy. 1.2–1.5 g/kg/day protein, distributed across 3–4 small meals/snacks to maximize muscle protein synthesis signaling. For a 60 kg patient: 72–90 g/day. Whey protein isolate (if dairy-tolerant) provides highest leucine content per gram. Plant-based alternatives: pea protein + rice protein blend for complementary amino acid profile
• Leucine threshold. 2.5–3.0 g leucine per meal is the threshold for mTOR-mediated muscle protein synthesis activation. This can be achieved with 25–30 g whey protein or 35–40 g soy protein. Timing: with each meal; supplemental free leucine (2 g) can be added to meals that fall below the threshold
• Creatine monohydrate. 3–5 g/day. Increases intramuscular phosphocreatine, which buffers ATP depletion and may slow muscle wasting by maintaining cellular energy status. Well-established safety profile. Contraindicated in renal impairment
• HMB (beta-hydroxy-beta-methylbutyrate). 3 g/day, divided into 3 doses. HMB is a leucine metabolite that specifically inhibits muscle protein breakdown via the ubiquitin-proteasome pathway. An RCT in healthy older adults during 10 days of complete bed rest showed HMB (1.5 g twice daily) significantly preserved lean body mass compared to placebo (–2.05 ± 0.66 kg vs. –0.17 ± 0.19 kg, p = 0.02, n=19 evaluable) (Deutz et al. 2013). A follow-up mechanistic study from the same cohort (Standley 2017, n=19, same 10-day bed rest protocol, not an independent replication) found HMB maintained mitochondrial OXPHOS complex II protein and preserved mitochondrial dynamics (fission/fusion balance) during the 8-week rehabilitation phase, suggesting anti-catabolic effects may extend to mitochondrial protection (Standley et al. 2017). Evidence in extended complete bed rest (>10 days) is limited; no independent replication of the HMB bed rest finding exists. Do not combine with high-dose leucine (redundant pathway). For severe ME/CFS patients with GI sensitivity or impaired absorption, start at 1 g/day with weekly escalation to 3 g/day as tolerated — the Deutz 2013 efficacious dose may not be tolerable in this population
• Vitamin D3 + calcium. 2,000–4,000 IU/day vitamin D3 + 1,000–1,200 mg/day calcium for bone density preservation. Bedbound patients lose bone density rapidly (1–2 percent per month in weight-bearing sites). Vitamin D also has muscle function effects independent of calcium. Monitor serum 25(OH)D quarterly; target 50–80 nmol/L
• Omega-3 fatty acids. 2–3 g/day EPA+DHA. Anti-inflammatory effect may reduce catabolic cytokine signaling (TNF-alpha, IL-6) that drives muscle wasting in chronic illness. Fish oil or algal oil source

Safety interactions (check before initiating any supplement in patients already on): Creatine: avoid in renal impairment, monitor creatinine. HMB: no significant interactions known; limited safety data beyond 12 months. Omega-3: potentiates anticoagulant/antiplatelet effect — caution with warfarin, clopidogrel, aspirin, high-dose fish oil. Vitamin D: calcium co-administration may increase risk of hypercalcemia in immobility — monitor serum calcium if both are taken.

Falsifiable prediction: Bedbound ME/CFS patients receiving combined nutritional anti-catabolic protocol (protein 1.5 g/kg/day + leucine threshold + creatine 5 g/day) for 12 weeks will show slower mid-thigh muscle cross-sectional area decline (ultrasound or DXA) compared to bedbound controls receiving standard nutrition, without increase in gastrointestinal PEM triggers.

Limitations: No ME/CFS trial. Nutritional interventions in bedbound patients carry aspiration risk for those with severe dysphagia or impaired consciousness (rare in ME/CFS). Gastric emptying may be slowed in severe patients, making large protein doses poorly tolerated — small, frequent servings are essential. Creatine may cause mild GI upset. HMB evidence is modest and primarily from non-ME/CFS populations. Compliance requires caregiver assistance.

CautionSpeculation: Positioning and Postural Management Prevents Contractures and Pressure Injuries in Bedbound ME/CFS

Certainty: 0.45. Well-established in ICU nursing, palliative care, and spinal cord injury rehabilitation; extrapolated to ME/CFS severe/bedbound by analogy. Not yet replicated in ME/CFS specifically.

Prolonged immobility in a single position causes: (a) muscle shortening → contractures (especially hip flexors, knee flexors, ankle plantarflexors); (b) sustained pressure over bony prominences → pressure injuries (sacrum, heels, elbows, occiput); (c) dependent edema in immobile limbs; (d) chest wall restriction from prolonged supine positioning → reduced lung expansion. A structured positioning protocol prevents these complications without metabolic cost to the patient.

Protocol:

• Repositioning schedule. Every 2 hours during waking hours, every 3–4 hours during sleep. A caregiver is required; the patient does not expend energy on repositioning
• Position rotation. Alternate between: supine (default), left side-lying (pillow between knees, pillow behind back), right side-lying (same supports), semi-recumbent (head-of-bed 30 degrees, if orthostatically tolerated — many severe patients may not tolerate this)
• Joint positioning. Hips: neutral rotation, slight abduction (pillow between knees in side-lying, small roll under knees in supine to prevent hyperextension). Knees: slight flexion (10–15 degrees) to avoid hyperextension and posterior capsule stretch. Ankles: neutral (90 degrees) to prevent plantarflexion contracture — use footboard, heel protector boots, or multi-podus boots if the patient tolerates foot contact. Shoulders: slight abduction (pillow under each arm in supine), neutral rotation. Elbows: slight flexion. Wrists: neutral; hand splints if finger flexion contractures are developing
• Pressure relief. Pressure-relieving mattress (alternating air or high-density foam). Heel protectors. Regular skin inspection by caregiver at repositioning times for early signs of pressure injury
• Micro-movement within tolerance. Where possible, encourage the patient to make tiny voluntary position adjustments (finger wiggles, toe curls, gentle head rotation) at each repositioning, provided these do not approach their energy envelope limit
• Contraindications. Active PEM episode → reduce repositioning frequency to essential position changes only (every 4 hours) as tolerated. Positions that worsen orthostatic symptoms (head-of-bed elevation in some POTS patients) → avoid. Pain with specific positions → modify or avoid

Falsifiable prediction: Bedbound ME/CFS patients receiving structured repositioning protocol (every-2-hour rotation + joint positioning) will show lower contracture incidence (goniometry at 12 weeks) and lower pressure injury rate compared to bedbound patients without structured positioning, without PEM frequency increase.

Limitations: Every-2-hour repositioning requires a caregiver present — a significant access barrier for patients living alone or without 24-hour care. Repositioning itself may trigger autonomic symptoms in very sensitive patients (the brief orthostatic stress of side-lying transition). No ME/CFS position-management trial exists. Extrapolation from ICU/Palliative care; applicability to ME/CFS-specific pathophysiology unverified.

2.6.7 Bed-Based Isometric Protocols for Severe Patients

CautionSpeculation: Bed-Based Isometric Holds as the Lowest-Intensity Tier of Safe Movement

Certainty: 0.40. Already partially in Ch. 17 (isometric exercises listed, bed-based micro-movement N3 exists); this extends to explicit severity stratification. Not yet replicated as a structured severity-tiered protocol.

The isometric exercises listed in the general safe movement section — wall sits, plank holds, leg presses against bed — are appropriate for mild-to-moderate patients. For severe/bedbound patients, these must be scaled to the lowest possible intensity tier: holds performed entirely supine, targeting only those muscle groups the patient can voluntarily contract without increasing respiratory rate, without exceeding resting HR by more than 5 bpm, and without producing any sensation of effort.

Severe patient protocol:

• Muscle groups. Start with single-muscle, unilateral holds: one quadriceps (press knee into bed, 3-second hold), one gluteal (squeeze, 3-second hold), one bicep (press forearm into bed, 3-second hold). No compound movements. Progress to bilateral only if single-muscle tolerated without PEM for 2 weeks
• Intensity. Submaximal — the patient should be able to talk normally during contraction. If breathing deepens, intensity is too high
• Hold duration. 3 seconds initially; progress to 5 seconds over 4 weeks if tolerated; 10 seconds maximum. Longer holds recruit a progressively larger fraction of muscle mass as slow-twitch fibers fatigue and fast-twitch fibers are recruited
• Frequency. 1 session/day, 2–3 minutes total contraction time. Rest 30 seconds between holds. Do not attempt on days with active PEM
• Monitoring. HR must remain within 5 bpm of resting HR. If HR increases > 5 bpm, the contraction intensity is too high. Symptom tracking at 6h, 24h, 48h post-session
• Environmental conditions. Quiet, darkened room. Supine position with head supported. Temperature comfortable (not cold, which increases muscle tension). Do not perform isometrics during or immediately after meals or cognitive exertion

Falsifiable prediction: Bedbound ME/CFS patients performing supine single-muscle 3-second isometric holds (2–3 min/day total contraction time) will show slower quadriceps cross-sectional area decline (ultrasound at 12 weeks) vs. bedbound controls, without PEM frequency increase.

Limitations: Even minimal isometric contractions may trigger PEM in very severe patients — this protocol is not guaranteed safe for all bedbound patients. The 3-second, single-muscle, sub-5-bpm-HR-increase parameters are clinician-estimated thresholds, not empirically validated. The protocol may be too minimal to produce measurable muscle preservation in bedbound patients who are already severely atrophied. Oka 2017 recumbent isometric yoga was for severe ME/CFS (n=12, pilot) — closer to this severity tier but not bedbound-severe.

NoteOpen Question: NMES/EMS for Muscle Preservation in Bedbound ME/CFS — Unknown PEM Risk

Certainty: Not applicable — safety threshold unknown.

Neuromuscular electrical stimulation (NMES/EMS) has demonstrated muscle mass preservation in ICU-immobilized patients, spinal cord injury, and prolonged bed rest in healthy volunteers. By activating motor units electrically rather than via voluntary neural drive, NMES can produce muscle contractions that maintain protein synthesis signaling and slow atrophy without requiring patient volitional effort. In fully sedated ICU patients, twice-daily NMES of quadriceps for 7 days prevented muscle atrophy entirely (n=6, within-subject crossover design: control leg: type I fibers –16 percent, type II –24 percent; stimulated leg: no atrophy, mTOR phosphorylation increased) (Dirks et al. 2015).

The risk in ME/CFS is that the electrically-induced contractions are metabolically indistinguishable from voluntary contractions — if the metabolic cost of muscle activation is the PEM trigger (rather than the cognitive effort of voluntary movement), then NMES could trigger PEM identically to volitional exercise. In ICU patients, 30–60 minutes of quadriceps NMES at visible-contraction-threshold intensity can produce measurable muscle preservation. Whether ME/CFS patients can tolerate even minimal NMES parameters (5 minutes, lowest visible contraction intensity, single muscle group) is completely unknown.

No ME/CFS NMES trial exists. The safety question — does NMES trigger PEM? — must be answered before any efficacy question can be asked. A tiered protocol starting from sub-contraction-threshold micro-current stimulation (below visible contraction, to modulate mTOR signaling without metabolic cost) and escalating to minimal visible-contraction NMES only if tolerated, with continuous PEM monitoring, could identify safe thresholds. If future research determines that sub-PEM-threshold NMES parameters exist, this could become a valuable atrophy countermeasure for the most vulnerable patients. Combining NMES with HMB (simultaneously activating mTOR protein synthesis and inhibiting ubiquitin-proteasome breakdown) is mechanistically plausible but completely untested in ME/CFS — PEM risk from the combination is unknown.

CautionWarning: Exercise Precautions

Patients with severe ME/CFS (housebound or bedbound) should consult physicians before attempting any structured movement program. Even minimal exertion may trigger severe crashes in this population. For these patients, activities of daily living (personal hygiene, eating) may constitute maximal tolerable exertion, leaving no additional capacity for exercise.

2.7 Severity-Stratified Muscle Preservation Protocol

CautionSpeculation: Severity-Stratified Muscle Preservation Preserves Function While Minimizing PEM Risk

Certainty: 0.35. Mechanistic inference from energy envelope theory combined with existing individual intervention evidence; no combined protocol tested in ME/CFS. Not yet replicated.

Not all bedbound ME/CFS patients have identical severity or metabolic reserve. A severity-stratified protocol matches intervention intensity to individual capacity:

Tier A — Very Severe (cannot tolerate any movement):

• Positioning only: Every-2-hour repositioning + joint neutral positioning + pressure relief (zero patient metabolic cost)
• Nutrition only: Protein adequacy (1.0–1.2 g/kg/day), vitamin D (2,000–4,000 IU/day), calcium (1,000–1,200 mg/day)
• No isometrics, PROM, NMES. Even passive movement may trigger PEM

Tier B — Severe (can tolerate minimal passive contact):

• All Tier A interventions plus: Caregiver-administered PROM (1 session/week, 3–5 min initially)
• Nutrition: HMB 3 g/day, creatine 3–5 g/day added if GI-tolerated
• Isometrics: Single-muscle 3-second holds only if 2 weeks of PROM produces zero PEM

Tier C — Moderate-Severe (can tolerate minimal voluntary contraction):

• All Tier B interventions plus: Bed-based isometric protocol (2–3 min/day)
• Nutrition: Full anti-catabolic protocol (protein 1.2–1.5 g/kg/day, HMB, creatine, omega-3)
• NMES consideration: Only if 4+ weeks of Tier B+C interventions produce zero PEM; start at 5 min minimal intensity single muscle

Progression and regression rules: Advancement requires ≥4 consecutive weeks at current tier with zero PEM attributable to the intervention. Any PEM episode → regress to prior tier; wait ≥2 weeks before retrying. Two PEM episodes from same tier → that tier is the individual ceiling.

Falsifiable prediction: Bedbound ME/CFS patients managed with severity-stratified protocol will show lower PEM frequency and comparable muscle preservation (CSA decline rate) vs. one-size-fits-all protocol over 24 weeks.

Limitations: Tier definitions based on clinician-estimated capacity thresholds, not empirically derived. Individual PEM sensitivity variation may exceed tier granularity. Protocol complexity increases caregiver burden. The boundary between “passive contact tolerated” (Tier B) and “minimal contraction tolerated” (Tier C) is subjective without objective biomarkers.

2.8 Cross-Disease Bridges for Bedbound Muscle Preservation

CautionSpeculation: Space Medicine and ICU Countermeasure Programs Offer Partially Translatable Frameworks for Bedbound ME/CFS

Certainty: 0.32. Cross-disease analogy — established in disuse atrophy models; no ME/CFS-specific adaptation tested. Not yet replicated in ME/CFS.

Multiple fields with extreme immobilization have developed countermeasure programs that partially translate, provided high-exertion components are removed:

Space medicine (NASA countermeasure program): Nutritional strategies (high protein, vitamin D, calcium), vibration platforms for bone, and resistance exercise principles are translatable; high-intensity resistive exercise (ARED device at 300+ lbs, treadmill with harness, 2+ hours/day) is not. Adaptation: HMB + creatine + vitamin D3 as nutritional analog; recumbent isometric holds (2–3 min/day) as near-zero exercise analog. Astronauts lose 1–2 percent bone density per month in microgravity — comparable to bedbound patients.

ICU early mobility (Morris 2008 n=330): PROM safety/feasibility framework, structured positioning protocols, and caregiver-administered monitored protocols translate; active mobility progression (sitting → standing → ambulation), twice-daily 30-min NMES do not. Adaptation: PROM protocol structure without the active progression ladder; ICU-style repositioning every 2h for contracture prevention (Morris et al. 2008).

Sarcopenia/frailty in elderly: Protein adequacy, HMB 3 g/day, vitamin D, and leucine threshold translate; progressive resistance training and upright mobility do not. Adaptation: full nutritional protocol; no active exercise beyond isometric tier (Deutz et al. 2013).

Spinal cord injury NMES: NMES prevents atrophy when voluntary movement is absent; FES cycling preserves muscle CSA. Translates conceptually but parameters must be titrated to PEM tolerance (see Section NMES/EMS for Muscle Preservation in Bedbound ME/CFS — Unknown PEM Risk) (Dirks et al. 2015).

ALS/neuromuscular disease: Positioning protocols, splinting, PROM routines for progressive weakness translate with the caveat that ALS pathophysiology (motor neuron degeneration) is distinct from ME/CFS (metabolic/immune/neuroimmune); ALS patients may tolerate active PROM where ME/CFS patients require passive-only.

Falsifiable prediction: Bedbound ME/CFS patients receiving adapted countermeasure protocol (nutritional + PROM + positioning, no active components) will show muscle CSA decline rates comparable to the lower end of ICU/space medicine countermeasure efficacy range (–0.3 to –0.5 percent per week CSA), significantly slower than untreated bed rest (–7 to –10 percent per week), without PEM induction.

Limitations: All analogies extrapolated from populations with fundamentally different pathophysiology. The “remove active components” strategy may render countermeasures ineffective — efficacy data exist only for protocols that include active components. No ME/CFS-specific adaptation study exists.

2.9 Muscle Mass Monitoring in Bedbound Patients

CautionSpeculation: Point-of-Care Ultrasound and Urinary Creatinine as Bedside Muscle Monitoring Tools

Certainty: 0.40. Well-validated in ICU/geriatric populations for muscle mass estimation; not yet validated specifically for ME/CFS bedbound monitoring.

Point-of-care ultrasound (POCUS) of quadriceps cross-sectional area and 24-hour urinary creatinine excretion offer two low-cost, bedside methods for monitoring muscle mass in patients who cannot travel for DXA or CT. POCUS requires a handheld ultrasound device (approximately 2000–6000 USD), 5-minute scan, zero radiation, performed supine with no patient movement. Validated against DXA/CT in ICU populations. Urinary creatinine requires accurate 24-hour collection (caregiver-assist needed) and is confounded by diet, renal function, and collection completeness, but serial measurements within the same patient track muscle mass trends.

Falsifiable prediction: Ultrasound quadriceps CSA decline rate will correlate with functional decline (Bell Disability Scale change) over 24 weeks in bedbound ME/CFS, and will detect intervention effects with sensitivity comparable to DXA-measured lean mass.

Limitations: Non-ME/CFS validation only. Neither method measures bone density. Urinary creatinine floor effects possible at low muscle mass. Reliable 24h urine collection challenging in severe patients.

NoteOpen Question: Optimal Intervention Combinations for Synergistic Muscle Preservation

The individual interventions — nutritional anti-catabolic support, PROM, positioning, isometric holds — address complementary pathways: nutrition provides substrate, PROM maintains joint range, positioning prevents contractures, isometrics provide minimal loading stimulus. Whether combining these produces additive or synergistic muscle preservation in bedbound ME/CFS is unknown.

Key open questions: (a) NMES + HMB synergy — one activates mTOR, the other inhibits ubiquitin-proteasome; mechanistically complementary but both introduce PEM risk; (b) PROM + positioning interaction — do both together reduce contracture incidence more than either alone? (c) Nutritional + mechanical synergy — does intact muscle protein synthesis machinery (maintained by nutrition) enable mechanical stimuli to produce an anabolic response? (d) Intervention ceiling — is there a maximum tolerable total intervention intensity beyond which even passive components trigger PEM?

CautionSpeculation: Caregiver Training and Protocol Fidelity Improves Muscle Preservation Outcomes

Certainty: 0.40. Established in chronic disease management for other conditions; extrapolated to ME/CFS by analogy. Not yet replicated in ME/CFS.

All bedbound muscle preservation interventions depend on caregiver execution. Standardized caregiver training — including PROM technique, repositioning procedure, nutritional preparation, and PEM recognition — may improve protocol fidelity and reduce adverse events. Training elements: correct joint positioning and range limits for PROM, safe position transitions and pressure point inspection for repositioning, small frequent meals and GI intolerance monitoring for nutrition, and standardized symptom logs to distinguish intervention-related PEM from background disease fluctuations.

Falsifiable prediction: Caregivers receiving structured training will achieve higher protocol adherence (validated by observation) and their patients will show better muscle preservation outcomes vs. untrained caregivers following written instructions alone, over 12 weeks.

Limitations: No ME/CFS caregiver-training trial exists. Caregiver burden may increase with training requirements. Training availability is a socioeconomic access barrier. Protocol fidelity measurement is inherently intrusive — may not be acceptable to severely ill patients.

CautionSpeculation: PEM Frequency May Accelerate Muscle Atrophy Independently of Disuse in ME/CFS

Certainty: 0.20. (0.15→0.20: caregiver-implemented pacing provides the PEM-reduction tool needed to test this mechanism; the connection is feed-into, not independent confirmation) Theoretical inference only; no direct evidence for PEM-specific atrophy acceleration.

If PEM episodes involve systemic inflammatory activation, oxidative stress, and metabolic crisis — all catabolic stimuli — then repeated PEM may accelerate muscle protein breakdown beyond the rate expected from disuse alone. In this model, muscle CSA decline rate = f(disuse duration, PEM frequency, PEM severity, nutritional status), where frequent PEM increases the slope of atrophy. This would imply that PEM prevention (via aggressive pacing) is itself a muscle preservation intervention.

Falsifiable prediction: In bedbound ME/CFS patients matched for immobility duration, those with higher PEM frequency (≥2 episodes/month) will show faster muscle CSA decline than matched low-PEM patients (≤0.5 episodes/month), independent of nutritional status and positioning adherence.

Limitations: Purely theoretical. No data correlating PEM frequency with muscle mass change. Disuse duration and PEM frequency are likely confounded (patients with more PEM may be more severe and have longer immobility). PEM-specific atrophy beyond standard inflammatory catabolism not established.

CautionSpeculation: Nutritional Status Modulates Atrophy Rate — A Multi-Variable ODE Extension

Certainty: 0.25. Mathematical induction from established atrophy kinetics × nutritional intervention data; no model validated against ME/CFS data.

A multi-variable ODE extension could model muscle mass decline as a function of immobility time, nutritional status, and PEM frequency: dM/dt = –αD – β(1 – S) – γP, where M = muscle CSA, D = disuse duration, S = nutritional status (protein adequacy, HMB, leucine, normalized 0–1), P = PEM frequency, and α, β, γ are patient-specific parameters. Deutz 2013 provides a potential α estimate (–2.05 kg / 10 days) but ME/CFS α may differ substantially (Deutz et al. 2013).

Falsifiable prediction: The multi-variable model will predict muscle CSA decline more accurately (R² improvement ≥0.10) than the disuse-only model when tested against longitudinal ultrasound data from bedbound ME/CFS (n ≥ 15), over 12–24 weeks.

Limitations: No ME/CFS validation data. Parameters unknown for ME/CFS. Model assumes linear additivity of disuse, nutrition, and PEM effects — likely interacting and nonlinear. γ parameter (PEM effect) is purely hypothetical.

NoteOpen Question: Cross-Disease Research Priorities for Bedbound Muscle Preservation in ME/CFS

The following research directions represent the most actionable translational investigations from cross-disease analogies: (a) NASA nutritional countermeasure trial — HMB + creatine + vitamin D3 + omega-3 adapted to bedbound ME/CFS, 12-week pilot (n=10–15), primary outcome ultrasound quadriceps CSA (Deutz et al. 2013); (b) NMES PEM threshold determination — single-arm dose-escalation pilot starting at 1 min lowest-visible-contraction NMES, establishing the fundamental safety question before any efficacy question (Dirks et al. 2015); (c) HMB dose-finding in ME/CFS — pilot (n=10–15) starting at 1 g/day with weekly escalation to 3 g/day, determining whether ME/CFS patients tolerate the Deutz 2013 efficacious dose; (d) PROM safety in ME/CFS — observational cohort (n=20), caregiver-administered PROM 1x/week, establishing safety before larger efficacy trial (Morris et al. 2008); (e) Positioning protocol validation — cohort (n=15) with structured 2-hour repositioning, providing ME/CFS-specific data on an intervention whose safety is well-established in other populations.

CautionSpeculation: Mechanical Vibration for Bone Density Preservation in Bedbound ME/CFS — Risk-Benefit Uncertain

Certainty: 0.15. Extrapolation from general osteoporosis WBV literature; Wollersheim 2017 ICU pilot (n=19) confirms safety and feasibility of supine WBV but demonstrated significant energy expenditure increase — the parameter most concerning for ME/CFS. No ME/CFS trial.

Whole-body vibration (WBV) is proposed for bone density preservation in non-ambulatory populations because mechanical loading stimulates osteocyte Wnt signaling without weight-bearing. However, Wollersheim 2017 found that supine WBV in ventilated ICU patients significantly increased energy expenditure (Wollersheim et al. 2017) — the metabolic cost that ME/CFS patients cannot afford. Very low-intensity mechanical vibration below the threshold that measurably increases energy expenditure remains theoretically possible but has not been tested in any population. Sub-metabolic-threshold vibration is a concept only — no device protocol exists, and whether vibration below the metabolic-threshold-for-non-ME/CFS still triggers PEM in ME/CFS is completely unknown.

For bone preservation in bedbound patients, vitamin D3 + calcium + protein adequacy remain the safest, most evidence-supported strategies. WBV represents an evidence gap; it should not be attempted outside research settings until ME/CFS-specific safety and dosing data exist, and then only in supine application at the lowest parameters tested. Bedbound or severe patients should never use standing WBV platforms.

Falsifiable prediction: Sub-metabolic-threshold supine vibration (parameters below those shown to increase energy expenditure in Wollersheim 2017) applied 10 min/day will increase bone formation marker P1NP without increasing HR, VO2, or PEM frequency in bedbound ME/CFS patients.

Limitations: No ME/CFS WBV trial. Wollersheim 2017 n=19, single-session, no bone outcomes. “Sub-metabolic-threshold” vibration is hypothetical — no device exists. Sensory sensitivities common in ME/CFS may prevent vibration tolerance. Even minimal vibration could trigger autonomic activation or sensory PEM.

NoteOpen Question: Speculative Nutritional and Pharmacological Adjuncts for Muscle Preservation

Several nutritional compounds and repurposed drugs with mechanistic rationale for anti-catabolic or muscle-preserving effects have not been tested in ME/CFS:

Supplement candidates (no ME/CFS data): Ursolic acid (apple peel component, inhibits atrophy via IGF-1 upregulation; preclinical data only), beta-alanine (intramuscular carnosine for pH buffering; athletic performance data only), phosphatidic acid (direct mTOR activation independent of leucine; limited human data), resveratrol (SIRT1/PGC-1alpha mitochondrial biogenesis; mixed human trial results), glycine (rate-limiting for collagen synthesis; connective tissue rationale in EDS+ME/CFS). All are mechanistically speculative with zero ME/CFS evidence. Do not prioritize any of these over the evidence-supported interventions (HMB, creatine, protein, vitamin D).

Repurposed drug candidates (speculative, research-stage only): Myostatin inhibitors (ACE inhibitors like lisinopril have weak myostatin-inhibiting effects — would require prospective study in POTS patients already prescribed lisinopril vs. other POTS medications), low-dose mifepristone (glucocorticoid receptor antagonist; documented HPA dysfunction in ME/CFS but no muscle-specific data; tested in other glucocorticoid-dysfunction conditions), bioidentical hormone replacement (testosterone, DHEA — only if documented deficiency; risks include cancer, cardiovascular, and unknown ME/CFS-specific adverse effects). These are clinical research questions only; no recommendation for clinical use without trial data.

Peptide vs. free-form amino acids: Some ME/CFS patients report improved GI tolerance of peptide supplements compared to free-form amino acids, but no ME/CFS absorption study exists comparing bioavailability. If free-form supplements are poorly tolerated (GI PEM trigger), peptide forms may be trialed with monitoring — but no recommendation can be made on current evidence.

WarningPractical Warning: Speculative Supplements Not Tested in ME/CFS

None of the above compounds have been tested in ME/CFS. They are listed for research prioritization, not clinical recommendation. Any trial in severe/bedbound patients should follow the micro-dosing imperative (1/10 standard dose start, single agent only, 2-week monitoring), carry explicit “uncertainty: untested” labeling, and be discontinued immediately if PEM frequency increases.

References

Bao, W., J. Yang, M. Li, K. Chen, Z. Ma, Y. Bai, and Y. Xu. 2022. “Prevention of Muscle Atrophy in ICU Patients Without Nerve Injury by Neuromuscular Electrical Stimulation: A Randomized Controlled Study.” BMC Musculoskeletal Disorders 23 (1): 780. https://doi.org/10.1186/s12891-022-05739-2.

Boyas, Sébastien, Antoine Nordez, Christophe Cornu, and Arnaud Guével. 2006. “Power Responses of a Rowing Ergometer: Mechanical Sensors Vs. Concept2 Measurement System.” International Journal of Sports Medicine 27 (10): 830–33. https://doi.org/10.1055/s-2006-923774.

Buryk-Iggers, Shonna, Nidhi Mittal, Daniel Santa Mina, et al. 2022. “Exercise and Rehabilitation in People with Ehlers-Danlos Syndrome: A Systematic Review.” Archives of Rehabilitation Research and Clinical Translation 4 (2): 100189. https://doi.org/10.1016/j.arrct.2022.100189.

Campen, C. (Linda) M. C. van, Peter C. Rowe, and Frans C. Visser. 2020. “Heart Rate Thresholds to Limit Activity in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome Patients (Pacing): Comparison of Heart Rate Formulae and Measurements of the Heart Rate at the Lactic Acidosis Threshold During Cardiopulmonary Exercise Testing.” Advances in Physical Education 10: 138–54. https://doi.org/10.4236/ape.2020.102013.

Charlton, Braeden T, Kasper Janssen, David M Systrom, David Putrino, and Rob C I Wüst. 2026. “Post-Exertional Malaise and the Myth of Cardiac Deconditioning: Rethinking the Pathophysiology of Long COVID.” British Journal of Sports Medicine. https://doi.org/10.1136/bjsports-2025-111387.

Clague-Baker, Natalie, Todd Davenport, Katy Wickens, Jo Leeming, Kirsty Dickinson, Heather McBurney, and Nicholas Hilliard. 2025. “Pacing with a Heart Rate Monitor for People with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and Long COVID: A Feasibility Study.” Fatigue: Biomedicine, Health & Behavior. https://doi.org/10.1080/21641846.2025.2565103.

Davenport, Todd E., Staci R. Stevens, J. Mark VanNess, Christopher R. Snell, and Tess Little. 2010. “Conceptual Model for Physical Therapist Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis.” Physical Therapy 90 (4): 602–14. https://doi.org/10.2522/ptj.20090047.

Deutz, N. E. P., S. L. Pereira, N. P. Hays, J. S. Oliver, N. K. Edens, C. M. Evans, and R. R. Wolfe. 2013. “Effect of \({\beta}\)-Hydroxy-\({\beta}\)-Methylbutyrate (HMB) on Lean Body Mass During 10 Days of Bed Rest in Older Adults.” Clinical Nutrition 32 (5): 704–12. https://doi.org/10.1016/j.clnu.2013.02.011.

Dirks, M. L., D. Hansen, A. Van Assche, P. Dendale, and L. J. C. Van Loon. 2015. “Neuromuscular Electrical Stimulation Prevents Muscle Wasting in Critically Ill Comatose Patients.” Clinical Science 128 (6): 357–65. https://doi.org/10.1042/CS20140447.

Flores, Sarah, Abigail Brown, Sarah Adeoye, Leonard A. Jason, and Meredyth Evans. 2013. “Examining the Impact of Obesity on Individuals with Chronic Fatigue Syndrome.” Workplace Health & Safety 61 (7): 299–307. https://doi.org/10.1177/216507991306100705.

Frontiers in Psychiatry Collaborative. 2025. “Effect of Nonpharmacologic Therapies on Depressive Symptoms in Patients with Chronic Fatigue Syndrome: A Network Meta-Analysis.” Frontiers in Psychiatry. https://doi.org/10.3389/fpsyt.2025.1657615.

Fu, Qi, and Benjamin D Levine. 2018. “Exercise and Non-Pharmacological Treatment of POTS.” Autonomic Neuroscience 215: 20–27. https://doi.org/10.1016/j.autneu.2018.07.001.

Fu, Qi, Tiffany B VanGundy, M Galbreath, Shigeki Shibata, Manish Jain, Jeffrey L Hastings, Paul S Bhella, and Benjamin D Levine. 2010. “Cardiac Origins of the Postural Orthostatic Tachycardia Syndrome.” Journal of the American College of Cardiology 55 (25): 2858–68. https://doi.org/10.1016/j.jacc.2010.02.043.

Gloeckl, Rainer, Rembert H. Zwick, Ulrike Fürlinger, Tessa Schneeberger, Daniela Leitl, Inga Jarosch, Uta Behrends, Carmen Scheibenbogen, and Andreas Rembert Koczulla. 2024. “Practical Recommendations for Exercise Training in Patients with Long COVID with or Without Post-Exertional Malaise: A Best Practice Proposal.” Sports Medicine – Open 10: 47. https://doi.org/10.1186/s40798-024-00695-8.

Hanna, R. D., L. M. Sheldahl, and F. E. Tristani. 1993. “Effect of Enhanced Preload with Head-Out Water Immersion on Exercise Response in Men with Healed Myocardial Infarction.” American Journal of Cardiology 71 (12): 1041–44. https://doi.org/10.1016/0002-9149(93)90570-3.

Held, Steffen, Tobias Siebert, and Lars Donath. 2020. “Changes in Mechanical Power Output in Rowing by Varying Stroke Rate and Gearing.” European Journal of Sport Science 20 (3): 357–65. https://doi.org/10.1080/17461391.2019.1628308.

Hofmijster, Mathijs J, Arthur J Van Soest, and Jos J De Koning. 2009. “Gross Efficiency During Rowing Is Not Affected by Stroke Rate.” Medicine and Science in Sports and Exercise 41 (5): 1088–95. https://doi.org/10.1249/MSS.0b013e3181912272.

Jason, Leonard A., Kathleen Muldowney, and Susan Torres-Harding. 2008. “The Energy Envelope Theory and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.” AAOHN Journal 56 (5): 189–95. https://doi.org/10.3928/08910162-20080501-06.

Kane, Daniel A, Sarah J MacKenzie, Randall L Jensen, and Phillip B Watts. 2013. “Effects of Stroke Resistance on Rowing Economy in Club Rowers Post-Season.” International Journal of Sports Medicine 34 (2): 131–37. https://doi.org/10.1055/s-0032-1321721.

Keller, Betsy A, Candace N Receno, Carl J Franconi, Sebastian Harenberg, Jared Stevens, Xiangling Mao, Staci R Stevens, et al. 2024. “Cardiopulmonary and Metabolic Responses During a 2-Day CPET in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Translating Reduced Oxygen Consumption to Impairment Status to Treatment Considerations.” Journal of Translational Medicine 22 (1): 627. https://doi.org/10.1186/s12967-024-05410-5.

Kindlon, Tom. 2011. “Reporting of Harms Associated with Graded Exercise Therapy and Cognitive Behavioural Therapy in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.” Bulletin of the IACFS/ME 19 (2): 59–111.

Mancini, Donna M., Dane B. Cook, Danielle L. Brunjes, Tiffany Soto, Michelle Blate, Patrick Quan, Tadahiro Yamazaki, Anna Norweg, and Benjamin H. Natelson. 2026. “Cardiopulmonary Exercise Test Results Do Not Change over Two Sequential Days in Patients with Chronic Fatigue Syndrome.” Frontiers in Physiology 17: 1816082. https://doi.org/10.3389/fphys.2026.1816082.

Martindale, William O, and D Gordon E Robertson. 1984. “Mechanical Energy in Sculling and in Rowing an Ergometer.” In: Terauds J Et Al, Eds. Biomechanics in Sports, 171–82.

Morris, P. E., A. Goad, C. Thompson, K. Taylor, B. Harry, L. Passmore, A. Ross, et al. 2008. “Early Intensive Care Unit Mobility Therapy in the Treatment of Acute Respiratory Failure.” Critical Care Medicine 36 (8): 2238–43. https://doi.org/10.1097/CCM.0b013e318180b90e.

Natelson, Benjamin H., Daniel L. Brunjes, and Donna Mancini. 2021. “Chronic Fatigue Syndrome and Cardiovascular Disease: JACC State-of-the-Art Review.” Journal of the American College of Cardiology 78 (10): 1056–67. https://doi.org/10.1016/j.jacc.2021.06.045.

National Institute for Health and Care Excellence. 2021. “Myalgic Encephalomyelitis (or Encephalopathy)/Chronic Fatigue Syndrome: Diagnosis and Management.” NICE guideline [NG206]. https://www.nice.org.uk/guidance/ng206.

Naviaux, Robert K., Jane C. Naviaux, Kefeng Li, A. Taylor Bright, William A. Alaynick, Lin Wang, Asha Baxter, Neil Nathan, Wayne Anderson, and Eric Gordon. 2016. “Metabolic Features of Chronic Fatigue Syndrome.” Proceedings of the National Academy of Sciences 113 (37): E5472–80. https://doi.org/10.1073/pnas.1607571113.

Oka, Takakazu, Tokusei Tanahashi, Toshiyuki Chijiwa, Batsaikhan Lkhagvasuren, Nobuyuki Sudo, and Kagari Oka. 2014. “Isometric Yoga Improves the Fatigue and Pain of Patients with Chronic Fatigue Syndrome Who Are Resistant to Conventional Therapy: A Randomized, Controlled Trial.” Biopsychosocial Medicine 8 (1): 27. https://doi.org/10.1186/s13030-014-0027-8.

Oka, Takakazu, Haruyoshi Wakita, and Kanae Kimura. 2017. “Development of a Recumbent Isometric Yoga Program for Patients with Severe Chronic Fatigue Syndrome/Myalgic Encephalomyelitis: A Pilot Study to Assess Feasibility and Efficacy.” Biopsychosocial Medicine 11: 5. https://doi.org/10.1186/s13030-017-0090-z.

Sanderson, Brian, and William Martindale. 1986. “Towards Optimizing Rowing Technique.” Medicine and Science in Sports and Exercise 18 (4): 378–84. https://doi.org/10.1249/00005768-198608000-00005.

Schmid, Jean-Paul, Markus Noveanu, Cyrill Morger, Raymond Gaillet, Mauro Capoferri, Matthias Anderegg, and Hugo Saner. 2007. “Influence of Water Immersion, Water Gymnastics and Swimming on Cardiac Output in Patients with Heart Failure.” Heart 93 (6): 722–27. https://doi.org/10.1136/hrt.2006.094870.

Sommerfelt, Knut, Trude Schei, Kathleen A. Seton, and Simon R. Carding. 2024. “Assessing Functional Capacity in Myalgic Encephalopathy/Chronic Fatigue Syndrome: A Patient-Informed Questionnaire.” Journal of Clinical Medicine 13 (12): 3486. https://doi.org/10.3390/jcm13123486.

Stallard, Jacqueline, Stephan Praet, Sandeep Gupta, and Angela Smith. 2026. “Is the RACGP HANDI Recommendation of Incremental Physical Activity for Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Harming Patients?” Australian Journal of General Practice 55 (3). https://doi.org/10.31128/AJGP-03-25-7614.

Standley, R. A., G. Distefano, S. L. Pereira, M. Tian, O. J. Kelly, P. M. Coen, N. E. P. Deutz, R. R. Wolfe, and B. H. Goodpaster. 2017. “Effects of \({\beta}\)-Hydroxy-\({\beta}\)-Methylbutyrate on Skeletal Muscle Mitochondrial Content and Dynamics, and Lipids After 10 Days of Bed Rest in Older Adults.” Journal of Applied Physiology 123 (5): 1092–1100. https://doi.org/10.1152/japplphysiol.00192.2017.

Stevens, Staci, Christopher Snell, Jared Stevens, Betsy Keller, and J. Mark VanNess. 2018. “Cardiopulmonary Exercise Test Methodology for Assessing Exertion Intolerance in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.” Frontiers in Pediatrics 6: 242. https://doi.org/10.3389/fped.2018.00242.

Treff, Gunnar, Lennart Mentz, Benjamin Mayer, Kay Winkert, Thomas Engleder, and Jürgen M Steinacker. 2022. “Initial Evaluation of the Concept-2 Rowing Ergometer’s Accuracy Using a Motorized Test Rig.” Frontiers in Sports and Active Living 3: 801617. https://doi.org/10.3389/fspor.2021.801617.

Ussher, Jake. 2025. Understanding ME/CFS & Strategies for Healing. Self-published.

Wilshire, Carolyn E, Tom Kindlon, Robert Courtney, Alem Matthees, David Tuller, Keith Geraghty, and Bruce Levin. 2018. “Rethinking the Treatment of Chronic Fatigue Syndrome—a Reanalysis and Evaluation of Findings from a Recent Major Trial of Graded Exercise and CBT.” BMC Psychology 6 (1): 6. https://doi.org/10.1186/s40359-018-0218-3.

Wollersheim, T., K. Haas, S. Wolf, K. Mai, C. Spies, E. Steinhagen-Thiessen, K. D. Wernecke, J. Spranger, and S. Weber-Carstens. 2017. “Whole-Body Vibration to Prevent Intensive Care Unit-Acquired Weakness: Safety, Feasibility, and Metabolic Response.” Critical Care 21 (1): 9. https://doi.org/10.1186/s13054-016-1576-y.

Wormgoor, Marten E. A., and Sanne C. Rodenburg. 2021. “The Evidence Base for Physiotherapy in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome When Considering Post-Exertional Malaise: A Systematic Review and Narrative Synthesis.” Journal of Translational Medicine 19 (1): 1–22. https://doi.org/10.1186/s12967-020-02683-4.

Zabriskie, Heather A. 2022. “Rationale and Feasibility of Resistance Training in hEDS/HSD: A Narrative Review.” Journal of Functional Morphology and Kinesiology 7 (3): 61. https://doi.org/10.3390/jfmk7030061.