Vascular Dysfunction
1 Endothelial Dysfunction
The vascular endothelium regulates vascular tone, coagulation, and inflammation. Endothelial dysfunction is increasingly recognized in ME/CFS.
1.1 Nitric Oxide Bioavailability
Nitric oxide (NO) is a critical vasodilator produced by endothelial NO synthase (eNOS):
- Altered NO metabolism during exercise: ME/CFS patients show significantly elevated nitric oxide metabolites (plasma nitrate up to \(\sim\) 295% above controls) during maximal exercise, consistent with abnormal vascular regulation rather than simple deficiency (Suárez et al. 2010)
- Increased NO scavenging: Oxidative stress may inactivate NO
- eNOS uncoupling: Dysfunctional enzyme produces superoxide instead of NO
- Consequences: Impaired vasodilation, increased vascular resistance
1.2 Flow-Mediated Dilation
Flow-mediated dilation (FMD) measures endothelium-dependent vasodilation of the brachial artery following brief ischemia:
- Reduced FMD in ME/CFS: Multiple studies report impaired endothelium-dependent dilation, with peripheral endothelial dysfunction found in 51% of ME/CFS patients versus 20% of healthy controls (Scherbakov et al. 2020)
- Correlation: Associated with disease severity and severity of immune symptoms (Scherbakov et al. 2020)
- Mechanism: Reflects reduced NO bioavailability, elevated adhesion molecules, or chronic inflammatory state (Appel, Marshall-Gradisnik, and Eaton-Fitch 2024)
1.3 Inflammatory Markers
Endothelial inflammation contributes to dysfunction ((Appel, Marshall-Gradisnik, and Eaton-Fitch 2024)):
- Elevated adhesion molecules: ICAM-1, VCAM-1, E-selectin (Appel, Marshall-Gradisnik, and Eaton-Fitch 2024)
- Increased inflammatory cytokines: IL-6, TNF-\(\alpha\) affect endothelial function
- Oxidative stress markers: Indicate endothelial damage
- Circulating endothelial cells: May be elevated, indicating endothelial injury
1.4 Immune Complex–Mediated Endothelial Injury
The inflammatory markers above document endothelial activation, but the mechanism driving this activation has remained unclear. Liu et al. (2026) (Liu et al. 2026) provided direct experimental evidence that IgG immune complexes from ME/CFS patients can injure endothelial cells. When purified IgG from post-infectious ME/CFS patients (n=39–40) was applied to human umbilical vein endothelial cells (HUVECs), the immune complexes entered the cells and induced mitochondrial fragmentation—a structural disruption visible on confocal microscopy. The IgG also triggered secretion of IL-1\(\beta\), a potent pro-inflammatory cytokine that itself promotes endothelial activation, adhesion molecule upregulation, and vascular permeability.
This finding offers a mechanistic explanation for the elevated adhesion molecules (ICAM-1, VCAM-1, E-selectin) and circulating endothelial cells described above: IgG immune complexes deposited on or internalised by endothelial cells could initiate a self-amplifying inflammatory cycle. IL-1\(\beta\) released by injured endothelial cells recruits further immune cells, amplifies complement activation (HMGB1-S100A8/A9 Double-DAMP Synergy at TLR4/RAGE in Immune System Dysfunction), and sustains the chronic low-grade vascular inflammation characteristic of ME/CFS.
Mass spectrometry proteomics of the IgG complexes revealed disease-subtype-specific signatures with direct vascular relevance:
- Classic ME/CFS IgG: Enriched in extracellular matrix (ECM) reorganisation pathways, consistent with the capillary basement membrane thickening and collagen IV deposition documented by Wüst et al. (Wüst et al. 2024) and confirmed at the ultrastructural level by Charlton, Slaghekke et al. (2025) (Charlton et al. 2025), who additionally documented endothelial microvacuolization, endothelial hypertrophy, and signs of endothelial degeneration on electron microscopy in skeletal muscle capillaries
- Post-COVID ME/CFS IgG: Enriched in hemostasis and blood clot formation pathways, consistent with the micro-clotting and coagulation abnormalities reported in Long COVID cohorts (Coagulation and Rheological Abnormalities)
These distinct proteomic signatures suggest that while the downstream endothelial injury may appear similar across ME/CFS subtypes, the upstream immune complex composition—and therefore the optimal therapeutic target—may differ between post-infectious and post-COVID presentations.
The convergence of IgG-mediated endothelial injury (Liu et al. 2026) with the proteomic evidence for reduced endothelial integrity (vimentin, ANTXR2, CXADR reduction (Hoel et al. 2026)), impaired flow-mediated dilation (Scherbakov et al. 2020) (Haffke et al. 2023), and capillary structural remodelling (Wüst et al. 2024) (Charlton et al. 2025) raises the possibility that immune complex deposition on microvasculature is not a secondary consequence but a primary driver of endothelial dysfunction in ME/CFS. In this model, persistent post-infectious immune activation (Viral Reactivation and Persistence in Immune System Dysfunction) generates pathogenic IgG complexes that continuously injure endothelial cells, producing the chronic microvascular dysfunction that underlies exercise intolerance, orthostatic intolerance, and cognitive impairment.
Certainty: 0.35. The in vitro evidence is strong for the direct IgG → endothelial injury step, but the claim that this represents a central rather than contributory mechanism in vivo requires: (1) demonstration of immune complex deposition in patient microvasculature (biopsy or imaging), (2) correlation between IgG complex composition and endothelial function measures (e.g., FMD), and (3) therapeutic evidence that IgG reduction (immunoadsorption, plasmapheresis) improves endothelial function parameters.
Falsifiable predictions: + Immunoadsorption that removes pathogenic IgG should improve flow-mediated dilation within weeks + Endothelial biopsy or skin punch biopsy should show IgG/complement deposition in ME/CFS microvasculature + Patients whose IgG induces greater mitochondrial fragmentation in vitro should have worse EndoPAT scores
Limitations: Not all patients’ IgG induced endothelial injury (heterogeneity); the HUVEC model may not reflect tissue-specific endothelial responses; female-predominant effect limits generalisability.
The mitochondrial effects of these immune complexes on endothelial cell energetics are discussed in detail in Energy Metabolism and Mitochondrial Function (No Direct Measurement of Intramuscular T3 in ME/CFS).
1.5 Proteomic Evidence for Endothelial Dysfunction and Tissue Hypoxia
Hoel et al. (2026) (Hoel et al. 2026) provided large-scale proteomic confirmation of endothelial and vascular abnormalities in ME/CFS. Several proteins linked to microcirculatory and hypoxia responses were altered:
- Reduced vimentin (VIM): An intermediate filament protein critical for endothelial cell structure and mechanotransduction; its reduction may indicate impaired endothelial integrity
- Reduced ANTXR2 (CMG2): Also known as capillary morphogenesis gene 2, this receptor regulates angiogenesis and endothelial cell chemotaxis; its reduction suggests impaired vascular repair capacity
- Reduced CXADR: The coxsackievirus and adenovirus receptor mediates endothelial responses to fluid shear stress; its decrease may reflect impaired vascular mechanosensing
- Reduced ATP5IF1: ATP synthase inhibitory factor 1 protects against mitochondrial ATP hydrolysis during hypoxia; its reduction implies compromised mitochondrial stress tolerance in endothelial cells
- Elevated LIPG: Endothelial cell-derived lipase, involved in lipoprotein metabolism and vascular biology, was increased—consistent with immune activation and endothelial stress
- Elevated FAM20A: A hematopoietic signaling factor elevated in ME/CFS, suggesting bone marrow-endothelial signaling disruption
Additionally, ME/CFS patients showed increased WASF3 expression, a target of the hypoxia-regulated transcription factor HIF-1\(\alpha\), directly linking the proteomic signature to tissue hypoxia. Together, these findings support a model in which endothelial dysfunction, impaired vascular repair, and chronic tissue hypoxia form an interconnected pathological axis in ME/CFS.
1.6 Ephrin Signaling: A Novel Vascular Pathway
Ligand-receptor interaction analysis from the Hoel et al. proteomics data (Hoel et al. 2026) revealed concordant changes in multiple members of the Ephrin signaling family: four Ephrin subfamily A receptors (EPHA1, EPHA2, EPHA4, EPHA7) and their EFNA ligands showed coordinated alterations in ME/CFS. Ephrin signaling regulates vascular development, endothelial cell migration, and angiogenesis. Disrupted Ephrin-Eph signaling had been previously suggested in ME/CFS based on earlier proteomic studies (Germain, Levine, and Hanson 2021), and the Hoel et al. findings provide independent confirmation across a different cohort and analytical platform. This pathway represents a potential mechanistic link between immune activation (Ephrin signaling also modulates immune cell migration and T cell activation) and the vascular dysfunction observed in ME/CFS.
1.7 Endothelial Senescence and Failed Angiogenic Compensation
Two findings advance the mechanistic understanding of endothelial dysfunction beyond functional impairment to a structural and cellular biology level.
Flaskamp et al. (2022) (Flaskamp et al. 2022) compared the effect of ME/CFS and post-COVID serum on endothelial cell behaviour in vitro. Post-COVID serum significantly enhanced angiogenic tube formation — a compensatory vascular response to hypoperfusion — whereas ME/CFS serum did not. ME/CFS serum selectively abolished this compensatory angiogenic response while also reducing ICAM-1 and E-selectin expression. If confirmed in vivo, this failed angiogenic compensation would explain why microvascular hypoperfusion becomes structural and self-sustaining in ME/CFS, consistent with the collagen IV deposition and basement membrane thickening documented by Wüst et al. (Wüst et al. 2024).
Nunes et al. (2026) (M. Nunes et al. 2026) proposed that virus-induced endothelial senescence at the BBB, cerebral arteries, gut vasculature, and skeletal muscle capillaries produces a SASP (senescence-associated secretory phenotype) that is pro-inflammatory, pro-oxidative, procoagulant, and vasoconstrictive. Senescent endothelial cells lose the capacity for vascular regeneration, consistent with the failed angiogenic response above. NK cell dysfunction and T cell exhaustion (Chapter Immune System Dysfunction, Section Virus-Induced Endothelial Senescence) prevent clearance of senescent cells, establishing a self-perpetuating loop. The therapeutic implication is that senolytic agents (dasatinib + quercetin) may reduce the senescent cell burden and break this cycle; no ME/CFS trial has yet tested this approach.
The ACHTSAM outreach study (Fricke et al. 2026) is deploying EndoPAT (peripheral arterial tonometry) as a non-invasive, home-based endothelial function assessment in severe ME/CFS patients (Bell score \(\leq\) 30)—a population typically excluded from vascular studies. Results, expected mid-2026, will provide the first endothelial function data specifically from severe and very severe patients assessed in their home environment.
1.8 HIF-2α-Mediated Endothelial Dysfunction as a Post-Viral Mechanism
A distinct line of evidence points to sustained hypoxia-inducible factor 2\(\alpha\) (HIF-2\(\alpha\), encoded by EPAS1) activation as a molecular mechanism for endothelial dysfunction specifically in post-viral ME/CFS, with implications for the ~70% of ME/CFS patients who lack \(\beta_2\)-adrenergic receptor autoantibodies and whose endothelial dysfunction mechanism has remained unexplained (Sandvik et al. 2023) (Stein et al. 2025).
Ribeiro et al. (2026) (Ribeiro et al. 2026) demonstrated that SARS-CoV-2 spike S1 protein produces a dichotomous HIF response in retinal endothelial cells: transient HIF-1\(\alpha\) activation (resolving within hours) followed by sustained HIF-2\(\alpha\) activation persisting for days, driving VEGF production, intercellular gap formation, and increased permeability. Critically, belzutifan — a selective HIF-2\(\alpha\) inhibitor approved for VHL disease-associated tumours — rescued endothelial barrier integrity in this model. This establishes a direct molecular pathway from viral protein exposure to chronic endothelial dysfunction via sustained HIF-2\(\alpha\), distinct from the acute HIF-1\(\alpha\) response that is normally adaptive (Ribeiro et al. 2026) (Bishop and Ratcliffe 2025).
The pathophysiological significance of sustained HIF-2\(\alpha\) (rather than HIF-1\(\alpha\)) as the mediator of post-viral endothelial injury is supported by convergent evidence. In pulmonary endothelium, acute hypoxia drives HIF-1\(\alpha\)-dependent barrier disruption, while chronic hypoxia produces HIF-2\(\alpha\)-dependent vascular remodelling — a temporal dissociation that mirrors the pattern triggered by spike protein (Reiterer et al. 2019). Mechanistically, antiviral signalling intersects directly with HIF regulation: interferon regulatory factor 3 (IRF3) retains both HIF-1\(\alpha\) and HIF-2\(\alpha\) in the cytoplasm under resting conditions, and its depletion during viral infection increases HIF signalling (Deng et al. 2026). The HIF-2\(\alpha\) isoform is exploited by multiple viruses for replication — hepatitis C virus requires HIF-2\(\alpha\) for production of infectious lipoviroparticles (Couteaudier et al. 2025), and respiratory syncytial virus stabilises HIFs under normoxic conditions (Morris et al. 2025) — suggesting that post-viral HIF-2\(\alpha\) activation may be a general phenomenon rather than SARS-CoV-2-specific.
This mechanism may explain several observations in post-viral ME/CFS. First, it provides a molecular basis for endothelial dysfunction in the \(\beta_2\)AR-autoantibody-negative majority: approximately 70% of ME/CFS patients have documented endothelial dysfunction (impaired FMD, reduced PORH) without detectable \(\beta_2\)-adrenergic receptor autoantibodies (Sandvik et al. 2023) (Stein et al. 2025). HIF-2\(\alpha\)-mediated VEGF production and barrier disruption would produce the same net tissue hypoxia as autoantibody-mediated vasoconstriction, but through increased capillary permeability rather than reduced vessel calibre — a mechanistic distinction with therapeutic implications (HIF-2\(\alpha\) inhibition vs immunomodulation). Second, sustained HIF-2\(\alpha\) activation could drive the immunothrombotic microclot phenotype documented in ME/CFS: HIF-2\(\alpha\) target VEGF triggers Weibel-Palade body exocytosis, releasing ultra-large von Willebrand factor multimers that promote platelet adhesion and microclot formation (Linden et al. 2023) (K. Wirth and Löhn 2024). Third, the NRF2-HIF-2\(\alpha\) axis offers a framework for understanding why endothelial dysfunction persists: NRF2 normally competes with HIF-2\(\alpha\) for transcriptional co-activators and maintains endothelial junction integrity (Shen et al. 2024), but NRF2 depletion from prolonged oxidative stress (documented in ME/CFS) removes this brake, permitting unchecked HIF-2\(\alpha\) dominance (Zhang et al. 2026).
The ~70% of ME/CFS patients without detectable \(\beta_2\)AR autoantibodies (Stein et al. 2025) nonetheless show endothelial dysfunction comparable to the autoantibody-positive subgroup (Sandvik et al. 2023). This paradox may reflect two distinct proximal mechanisms converging on the same distal phenotype: autoantibody-driven vasoconstriction in the ~30% vs HIF-2\(\alpha\)-driven capillary permeability in the ~70%. The latter would explain normal or elevated VEGF in this subgroup, responsiveness to compression therapy (which counteracts capillary leak), and the association between post-viral onset and elevated endothelial activation markers (vWF, ICAM-1). This framework generates a testable prediction: \(\beta_2\)AR-autoantibody-negative ME/CFS patients will show elevated plasma VEGF, vWF, and EPO compared to autoantibody-positive and healthy controls, with VEGF levels correlating with FMD impairment. Certainty: 0.55 (indirect evidence from separate literatures converged; direct ME/CFS testing is pending).
Falsifiable predictions: + In a cohort of ME/CFS patients stratified by \(\beta_2\)AR autoantibody status, the autoantibody-negative subgroup will show higher plasma VEGF and vWF than the autoantibody-positive subgroup + VEGF levels will correlate inversely with FMD in the autoantibody-negative subgroup (r \(<\) -0.5) but not in the autoantibody-positive subgroup + Belzutifan (HIF-2\(\alpha\) inhibitor) will improve FMD and reduce VEGF preferentially in the autoantibody-negative subgroup
HIF-2\(\alpha\) target VEGF induces Weibel-Palade body exocytosis, releasing ultra-large vWF multimers that promote platelet adhesion without requiring coagulation cascade activation (Ribeiro et al. 2026). This provides a direct mechanism for the fibrinolysis-resistant amyloid-type microclots documented in ME/CFS and Long COVID (Linden et al. 2023) (Davis et al. 2023). The HIF-2\(\alpha\)-driven pathway does not require autoantibodies and would operate in any post-viral state where viral proteins or nucleic acids persist sufficiently to maintain HIF-2\(\alpha\) activation. Certainty: 0.50 (VEGF-vWF link is established; microclot composition data are consistent; direct demonstration of HIF-2\(\alpha\) dependence in ME/CFS microclots is pending).
Falsifiable predictions: + Plasma vWF antigen and vWF propeptide (marker of acute Weibel-Palade release) will be elevated in post-viral ME/CFS and correlate with microclot burden measured by fluorescence microscopy + vWF levels will correlate with VEGF but not with \(\beta_2\)AR autoantibody titres + In vitro, belzutifan reduces spike-induced vWF release from cultured endothelial cells (HUVEC or HREC)
Acute viral infection depletes NRF2 antioxidant reserve (GSH consumption, lipid peroxidation). NRF2 normally maintains endothelial junction integrity and competes with HIF-2\(\alpha\) for CBP/p300 transcriptional co-activators (Shen et al. 2024). NRF2 exhaustion removes this brake, allowing unchecked HIF-2\(\alpha\) activation and sustained VEGF production — a bimodal switch: moderate oxidative stress recruits protective NRF2-HIF-2\(\alpha\) signalling, but severe or prolonged stress depletes NRF2, permitting pathological HIF-2\(\alpha\) dominance (Zhang et al. 2026). This provides a mechanistic link between the oxidative stress documented in ME/CFS and the sustained endothelial dysfunction: the same reactive species that mark the disease may also lock the HIF-2\(\alpha\) system in its pathological state. Certainty: 0.50 (Shen2024NRF2HIF2Senescence and Zhang2026HIFSwitchICH mechanistic studies; NRF2 depletion in ME/CFS is plausible but not directly measured in endothelial cells).
Falsifiable predictions: + NRF2 target genes (NQO1, HO-1, GCLC) will be suppressed in post-viral ME/CFS endothelial cells relative to HIF-2\(\alpha\) targets + Sulforaphane (NRF2 activator) will rescue barrier integrity in spike-stimulated endothelial cells in vitro + NRF2 activation status will predict VEGF response: patients with low NRF2 activity show the greatest VEGF reduction with NRF2 activator treatment
The HIF-2\(\alpha\) environments in this section assemble a single candidate mechanism for the subset of ME/CFS endothelial dysfunction that follows viral infection: SARS-CoV-2 spike drives a dichotomous hypoxia-inducible-factor response — transient HIF-1\(\alpha\) followed by sustained HIF-2\(\alpha\) that persists after the trigger clears (HIF-2\(\alpha\) Sustained Activation as the Molecular Basis for Post-Viral Endothelial Dysfunction, Immune Complex–Endothelial Injury as a Central Vascular Mechanism). This section’s contribution is to trace that single (in-vitro, unvalidated-in-ME/CFS) finding into three interpretive implications — not three independent findings — for endothelial dysfunction, complementary to and deliberately not duplicating the VEGF/angiogenic-failure and sFlt-1/PlGF-biomarker arm already synthesised in the cross-disease chapter (VEGF Pathway Dysfunction Converges Across HIF-1α/HIF-2α Dichotomous Responses: Angiogenic Failure, Sustained HIF-2α, and the sFlt-1/PlGF Biomarker Model). First, sustained HIF-2\(\alpha\) offers one candidate mechanism — not the only one — for the roughly 70% of ME/CFS endothelial dysfunction that is independent of \(\beta\) 2-adrenergic-receptor autoantibodies (Sustained HIF-2\(\alpha\) as a Mechanistic Explanation for \(\beta_2\)AR-Autoantibody-Negative Endothelial Dysfunction): a non-autoimmune route to the same functional deficit, parallel in logic to the structural-vs-functional dichotomies elsewhere in the paper. It competes with, rather than supersedes, the IgG immune-complex endothelial-injury model advanced in this same section (Immune Complex–Endothelial Injury as a Central Vascular Mechanism), which proposes a different non-autoantibody route (internalised IgG complexes driving mitochondrial fragmentation and IL-1\(\beta\) release); the two are experimentally separable — the HIF-2\(\alpha\) account predicts correlation with VEGF/vWF and HIF-2\(\alpha\)-target expression, the IgG account predicts correlation with circulating immune-complex/IgG proteomics — and it is unknown whether they operate in the same patients or define distinct subgroups. Second, via VEGF-driven Weibel-Palade exocytosis of von Willebrand factor, it links the transcriptional state to the microclot phenomenon (HIF-2\(\alpha\) as the Proximal Mechanism for Post-Viral Microclot Formation), tying a gene-regulation abnormality to a coagulation readout. Third, it is permitted rather than caused in isolation: NRF2 antioxidant exhaustion is proposed as the condition under which HIF-2\(\alpha\) escapes normal restraint (NRF2 Exhaustion Permits Unchecked HIF-2\(\alpha\) Dominance in Post-Viral Endothelial Cells), making the mechanism a two-factor (loss-of-brake plus gain-of-drive) rather than single-hit story. Because HIF-2\(\alpha\) is pharmacologically tractable, this mechanism carries an unusually concrete — though entirely investigational — therapeutic corollary: HIF-2\(\alpha\) inhibition (belzutifan) and HIF-2\(\alpha\)-blocking repurposed agents (darunavir) are discussed as mechanistic probes in the treatment chapters, with the explicit caution that HIF-PH inhibitors could worsen a HIF-2\(\alpha\)-dominant endotype and therefore require stratification.
The honest status is that the entire mechanistic chain is inferential for ME/CFS: the dichotomous-HIF finding is from in vitro spike-stimulated endothelial cells (Ribeiro 2026), the CPET-dynamics and sFlt-1/PlGF proposals (Exercise-Induced HIF-2\(\alpha\) Dynamics Differentiate Post-Viral from Gradual-Onset ME/CFS) are untested, and no study has measured endothelial HIF-2\(\alpha\) status in ME/CFS patients. What makes the mechanism attractive is not evidential weight but that it is falsifiable and stratifiable: it predicts a measurable HIF-2\(\alpha\)-target signature that should be present in post-viral but not gradual-onset endothelial dysfunction, should correlate with vWF/microclot burden, and should track NRF2 status — a set of predictions a single well-designed endothelial-biopsy or circulating-endothelial-cell study could confirm or refute. The central open question is whether sustained HIF-2\(\alpha\) is a driver of post-viral endothelial dysfunction or a marker of the same upstream inflammatory insult that damages the endothelium by other routes.
Consequence: This offers a testable reason why some post-viral ME/CFS patients have “leaky,” clot-prone blood vessels without the autoantibodies usually blamed — a stuck oxygen-sensing switch in the vessel lining — and, unusually, it points at an existing drug class (HIF-2\(\alpha\) inhibitors) that could be trialled to test the idea; but the whole chain is so far shown only in cultured cells and has never been measured in ME/CFS patients, so its immediate value is telling researchers exactly which measurement would confirm or kill it, not guiding treatment.
Several 2025–2026 studies extend the endothelial dysfunction evidence. Retinal vessel analysis revealed reduced venular flicker-induced dilation in ME/CFS and post-COVID patients, providing a non-invasive measure of persistent microvascular dysfunction (2026) (Appendix Ongoing and Planned ME/CFS Research Studies, Section Cardiovascular, Autonomic, and Vascular Research). Azcue et al. (2025) documented small fibre neuropathy in both ME/CFS and post-COVID patients using corneal confocal microscopy and electrochemical skin conductance (Azcue et al. 2025) (Section Cardiovascular, Autonomic, and Vascular Research). The MCAM multi-site study confirmed that autonomic dysfunction (assessed by COMPASS-31 and NASA lean test) is pervasive across diverse ME/CFS cohorts (Issa et al. 2025) (Section Cardiovascular, Autonomic, and Vascular Research).
2 Blood Volume Abnormalities
Blood volume deficits are among the most consistently documented abnormalities in ME/CFS (see also Section Primary Site of Autonomic Dysfunction Unresolved for impact on cardiac preload). This section expands on measurement methods and pathophysiological mechanisms.
2.1 Measurement and Magnitude
- Measurement methods: Radioisotope dilution (gold standard), carbon monoxide rebreathing, dye dilution
- Plasma volume: Typically 10–20% below predicted (Streeten and Bell 1998)
- Red cell mass: Variable; some studies report proportional reduction, others find preserved red cell mass with disproportionate plasma volume loss
- Total blood volume: 10–15% below normal in most studies (Newton et al. 2016)
- Hemoglobin/hematocrit: May appear normal or elevated due to hemoconcentration
2.2 Mechanisms of Volume Depletion
Renin-Angiotensin-Aldosterone System Dysfunction Studies document a paradoxical RAAS response in POTS and ME/CFS, with elevated angiotensin II despite hypovolemia ((Raj et al. 2005), (Stewart, Glover, and Medow 2006)):
- Blunted aldosterone response to hypovolemia (Raj et al. 2005)
- Impaired sodium retention leading to inappropriate natriuresis
- Elevated angiotensin II may contribute to symptoms through vasoconstriction (Stewart, Glover, and Medow 2006)
Miwa (2016) quantified this paradox directly, demonstrating that renin and aldosterone output were abnormally low in ME/CFS patients despite simultaneous hypovolemia—a pattern Raj had previously identified in POTS and termed the “renin-aldosterone paradox” (Miwa 2016) (Raj et al. 2005). Under normal physiology, RAAS activity rises sharply to correct low blood volume; that it fails to do so in ME/CFS indicates a specific pathological mechanism suppressing the system rather than simple deconditioning.
Vasoactive Substance Spillover and Bradykinin-Mediated RAAS Suppression
Falsifiability: weakly — Falsified if (a) systemic bradykinin levels are normal in ME/CFS patients during and after exertion, or (b) kallikrein-kinin system blockade fails to restore RAAS activity in ME/CFS
Wirth and Scheibenbogen (2020) proposed that the RAAS paradox originates upstream in skeletal muscle hypoperfusion (K. J. Wirth and Scheibenbogen 2020). Under their model, \(\beta_2\)-adrenergic receptor dysfunction (from sympathetic desensitisation, chronic vasoconstriction, and autoantibody blocking) impairs vasodilation in muscle vasculature. The body compensates by upregulating local vasoactive substances—bradykinin, prostaglandins, adenosine, ATP, and prostacyclin—to force blood vessel opening. When this compensatory production exceeds local clearance capacity, it spills from muscle interstitium into the systemic circulation.
Bradykinin reaching the kidneys produces two hypovolemia-inducing effects:
- Increased renal blood flow forces filtration of more solute (salt and water) than normal.
- Kallikrein-kinin system (KKS) opposition to RAAS: bradykinin directly suppresses renin release in the juxtaglomerular apparatus. Because renin is the rate-limiting first step in the RAAS cascade, its suppression propagates downward to reduce angiotensin~II and aldosterone output.
The net result is chronic inappropriate natriuresis—the body dumps salt despite needing to retain it—producing progressive plasma volume contraction.
Testable predictions:
- (a): Systemic bradykinin or kallikrein levels are elevated in ME/CFS patients at rest and rise disproportionately after exertion compared to controls.
- (b): Pharmacological KKS blockade (e.g., icatibant, a bradykinin B2 receptor antagonist) restores appropriate RAAS upregulation in response to orthostatic challenge.
- (c): The magnitude of RAAS suppression correlates with the severity of muscular hypoperfusion measured by NIRS or invasive CPET.
Treatment implication: If bradykinin spillover drives RAAS suppression, then treatments restoring muscular vasodilation (e.g., \(\beta_2\)-AR resensitisation via Mitodicure/MDC002) would address the upstream cause, while oral rehydration solutions and fludrocortisone provide downstream compensation. Certainty: 0.50—the model is mechanistically coherent and consistent with the Miwa (2016) and Raj (2005) RAAS paradox data, but direct measurement of systemic bradykinin in ME/CFS has not been performed.
Natriuretic Peptide Effects
- Elevated ANP or BNP promoting sodium/water excretion
- May result from atrial stretch due to cardiac filling abnormalities
Capillary Permeability and Microvascular Leakage
Increased vascular permeability shifts fluid to the interstitium and may be inflammation-mediated (cytokines increase endothelial permeability), explaining peripheral oedema in some patients despite intravascular hypovolemia. Wirth and Scheibenbogen (2020) identify bradykinin as a specific mediator of microvascular leakage: at sufficiently high concentrations, bradykinin disrupts endothelial tight junctions, creating microscopic vascular leaks through which blood components escape into the interstitial space (K. J. Wirth and Scheibenbogen 2020). The leaked fluid is eventually returned via lymphatic drainage, but sluggish lymphatic function may delay this return, transiently worsening hypovolemia after exertion.
2.3 Hypovolemia-Driven Thirst
The brain contains two distinct thirst centres: an osmotic centre (activated by cellular dehydration from high plasma osmolality) and a hypovolemic centre (activated by low circulating volume). In ME/CFS, the dominant thirst driver may be hypovolemic: patients are thirsty because they lack blood volume, not because their cells are dehydrated (Ussher 2025) (this interpretation is proposed by Ussher based on the Wirth-Scheibenbogen model; direct comparison of osmotic versus hypovolemic thirst responses in ME/CFS has not been performed). The hypovolemic thirst threshold is approximately a 10% drop in plasma volume (280mL), a deficit commonly exceeded in ME/CFS patients.
This distinction has important clinical consequences. Patients responding to hypovolemic thirst by drinking plain water do not resolve the problem: without adequate sodium reabsorption capacity (suppressed RAAS), water passes through the kidneys without expanding plasma volume. The result is a vicious cycle of drinking and urinating without benefit—some patients report consuming 8–20 litres per day without relief (patient-reported; no clinical measurement data available).
Huhmar et al. (2024) measured vasopressin levels in 111 ME/CFS patients after overnight fasting and 10~hours of fluid deprivation. 82% had vasopressin below the detection limit, and 66.7% had lower-than-normal urine osmolality—indicating that the kidneys fail to concentrate urine even under dehydration conditions (Huhmar et al. 2024). This is consistent with secondary vasopressin suppression downstream of RAAS dysfunction rather than primary diabetes insipidus.
Oral rehydration solutions (ORS) containing glucose, sodium, and potassium exploit the intestinal sodium-glucose cotransporter (SGLT1) to achieve efficient electrolyte absorption, effectively bypassing the renal retention deficit. Clinical evidence for ORS efficacy is discussed in Chapter Supplements and Nutraceuticals.
2.4 The CO2–Bohr–Bradykinin Triangle
The bradykinin spillover model (Section Bradykinin Spillover as the Mechanism of RAAS Suppression) and the low CO2 findings (Chapter Lifestyle and Non-Pharmacological Interventions) interact through a previously unrecognised feedback loop. Low pCO2 shifts the oxygen-haemoglobin dissociation curve leftward via the Bohr effect, reducing oxygen release to tissues. This worsens tissue hypoxia—precisely the condition that drives compensatory bradykinin production. The triangle is: impaired muscular respiration → low CO2 → reduced O2 delivery (Bohr effect) → worsened tissue hypoxia → increased bradykinin production → more RAAS suppression → more hypovolemia → worse perfusion → worse muscular respiration. Interventions raising CO2 (e.g., adapted Buteyko breathing, Chapter Lifestyle and Non-Pharmacological Interventions) would cut this triangle at the CO2–hypoxia link, reducing bradykinin production as a secondary benefit beyond direct vascular effects.
Certainty: 0.30 — each individual link (low CO2, Bohr effect, bradykinin production) is physiologically established, but their integration into a self-reinforcing triangle in ME/CFS is a novel synthesis that has not been directly tested.
3 Arterial Stiffness and Vascular Compliance
Beyond endothelial function, arterial mechanical properties influence cardiovascular regulation. Pulse wave velocity (PWV), a measure of arterial stiffness, affects blood pressure regulation through its impact on baroreceptor function. The relationship between arterial stiffness and ME/CFS has not been extensively studied, but related conditions provide insight.
In hypermobile Ehlers-Danlos syndrome (hEDS), which frequently co-occurs with ME/CFS (Section Prospective Phenotyping as Harm Reduction), central pulse wave velocity is significantly lower than controls (4.73 m/s versus normal values), indicating excessive arterial elasticity (Miller et al. 2020). This increased compliance paradoxically impairs blood pressure regulation: stretch receptors in vessel walls (baroreceptors) cannot accurately detect pressure changes when arterial walls are too compliant. The result is impaired baroreflex function and orthostatic intolerance despite (or because of) increased rather than decreased arterial elasticity.
Whether ME/CFS patients without connective tissue disorders show altered arterial stiffness remains unclear. Chronic inflammation typically increases arterial stiffness over time, while autonomic dysfunction could affect vascular smooth muscle tone. The interaction between these competing influences likely varies across patient subgroups. Pulse wave velocity measurement is non-invasive and could provide additional phenotyping data, though its clinical utility in ME/CFS management has not been established.
4 Microcirculation
4.1 Capillary Perfusion
The microcirculation delivers oxygen and nutrients to tissues and removes metabolic waste:
- Red cell deformability: Impaired RBC flexibility may impede capillary transit (Saha et al. 2019)
- Microclot obstruction: Amyloid-resistant fibrinaloid microclots may occlude microcapillaries, impairing perfusion (J. M. Nunes et al. 2022)
4.2 Oxygen Extraction
Peripheral oxygen extraction findings in ME/CFS appear contradictory across studies, likely reflecting patient heterogeneity or methodological differences:
- Contradictory findings: Some studies report widened arteriovenous O2 difference (increased extraction compensating for reduced cardiac output), while others report impaired extraction consistent with mitochondrial limitation; these contradictions may reflect distinct patient subgroups
- Near-infrared spectroscopy: Abnormal muscle oxygenation kinetics during exercise and delayed recovery are documented by NIRS, consistent with both delivery and utilization abnormalities (Keller et al. 2024)
4.3 Tissue Hypoxia
Inadequate oxygen delivery produces tissue hypoxia:
- Muscle hypoxia: Contributes to weakness and post-exertional symptoms
- Cerebral hypoperfusion: Causes cognitive dysfunction (see Chapter Neurological and Neurocognitive Dysfunction)
- Lactate accumulation: Results from anaerobic metabolism
- Symptom generation: Hypoxia-sensitive nociceptors may trigger pain
4.4 Cerebral Blood Flow During Orthostatic Stress
While tissue hypoxia affects multiple organs, the brain is particularly vulnerable to perfusion deficits during orthostatic challenge. Van Campen and colleagues have systematically characterized cerebral blood flow (CBF) abnormalities in ME/CFS through a series of rigorous studies using transcranial Doppler during tilt-table testing (C. L. M. C. van Campen et al. 2020) (C. L. M. C. van Campen, Rowe, and Visser 2021) (C. L. M. C. van Campen, Rowe, and Visser 2023) (C. L. M. C. van Campen et al. 2024).
Van Campen et al. (C. L. M. C. van Campen et al. 2020) demonstrated that ME/CFS patients show reduced cerebral blood flow during head-up tilt testing even in the absence of hypotension or tachycardia. The findings are striking in their consistency across orthostatic phenotypes (percentages represent distinct, non-overlapping subgroups stratified by vital sign response):
- 82% of patients with normal HR/BP showed abnormal CBF reduction
- 98% of patients with delayed orthostatic hypotension showed abnormal CBF
- 100% of patients meeting POTS criteria showed abnormal CBF
- End-tilt CBF reduction: 26% in ME/CFS vs. 7% in controls (3.7-fold greater)
Abnormal CBF reduction thus occurs across all orthostatic presentations—even in patients with entirely normal vital signs. In the largest study to date (n=534), 91% of ME/CFS patients with normal HR and BP responses demonstrated abnormal cardiac output and CBF reduction during tilt (C. L. M. C. van Campen et al. 2024), indicating that standard orthostatic vital signs miss the primary pathology in most patients.
All van Campen CBF studies originate from a single Dutch referral center using the same tilt-table protocol and transcranial Doppler methodology (C. L. M. C. van Campen et al. 2020) (C. L. M. C. van Campen, Rowe, and Visser 2021) (C. L. M. C. van Campen, Rowe, and Visser 2023) (C. L. M. C. van Campen et al. 2024). While internally consistent across multiple publications (n=107 to n=534), no independent research group has replicated these findings using equivalent methodology. The sequential publication pattern from one center may give the appearance of independent replication but represents a single group’s cohort. Independent replication with standardized protocols at other centers is essential. See also the related CBF findings in Chapter Neurological and Neurocognitive Dysfunction (Achievement Near-Universal CBF Decline During Orthostatic Challenge).
Clinical Implications of CBF Findings
The cognitive symptoms during orthostatic stress—including brain fog, difficulty concentrating, and word-finding problems—correlate directly with the degree of cerebral hypoperfusion (C. L. M. C. van Campen, Rowe, and Visser 2023). Patients often report that cognitive function worsens progressively during prolonged standing and improves rapidly upon assuming a recumbent position. This positional dependence of cognitive symptoms provides clinical evidence for the cerebrovascular contribution to ME/CFS neurological dysfunction.
CBF reduction persists even after returning to supine position. Van Campen et al. (C. L. M. C. van Campen, Rowe, and Visser 2021) documented CBF reduction of \(-29%\) at end-tilt, improving to only \(-16%\) post-tilt. The degree of recovery correlated with disease severity rather than hemodynamic parameters, suggesting the CBF abnormality reflects intrinsic cerebrovascular or metabolic dysfunction rather than simple hemodynamic failure.
Absence of Compensatory Vasodilation
A particularly significant finding is the near 1:1 relationship between cardiac output reduction and CBF reduction in ME/CFS patients (C. L. M. C. van Campen et al. 2024). In healthy individuals, reduced cardiac output triggers compensatory cerebral vasodilation to maintain brain perfusion. The absence of this compensation in ME/CFS suggests possible endothelial dysfunction affecting cerebrovascular autoregulation. This may represent a critical vulnerability: the brain cannot protect itself from systemic hemodynamic perturbations.
Mechanisms of Cerebral Hypoperfusion
Multiple mechanisms likely contribute to orthostatic cerebral hypoperfusion in ME/CFS. Evidence strength varies: (documented) = directly measured in ME/CFS studies; (inferred) = logically derived from observed relationships; (hypothesized) = proposed mechanism not yet directly tested.
- Reduced cardiac output (documented): Preload failure and chronotropic incompetence limit systemic perfusion pressure; directly measured in tilt studies showing parallel CO and CBF reduction (C. L. M. C. van Campen et al. 2024) (see Section Primary Site of Autonomic Dysfunction Unresolved)
- Impaired cerebral autoregulation (inferred): Failure of compensatory vasodilation during reduced perfusion pressure; inferred from near 1:1 CO-CBF relationship where healthy controls show compensatory vasodilation (C. L. M. C. van Campen et al. 2024) (Medow and Stewart 2024)
- Endothelial dysfunction (hypothesized): May impair nitric oxide-mediated vasodilation; suggested by absence of compensatory response but not directly measured in CBF studies
- Autonomic dysregulation (documented): Impaired sympathetic vasoconstriction in peripheral vascular beds allows excessive venous pooling; documented via HRV and catecholamine studies (see Chapter Neurological and Neurocognitive Dysfunction Section Autonomic Nervous System Dysfunction)
- Blood volume deficit (documented): Reduced circulating volume exacerbates orthostatic hemodynamic stress; documented in multiple studies showing 10–15% blood volume reduction (see Section blood volume)
- Orthostatic hypocapnia (documented): Van Campen et al. (2023) (C. M. C. van Campen et al. 2023) quantified PETCO2 during tilt in 535 female ME/CFS patients: end-tilt PETCO2 was 26–30~mmHg vs. 36±3~mmHg in controls. CO2 reactivity slopes were preserved, meaning the hypocapnia drives CBF reduction via normal cerebrovascular reactivity — orthostatic hypocapnia functions as a physiological amplifier: reduced cardiac output causes ventilation/perfusion mismatch, PETCO2 falls, CO2-sensitive cerebral vessels constrict, and CBF falls further than systemic haemodynamics alone would predict (Badhwar et al. 2025) (Medow and Stewart 2024). This mechanism may explain why CBF reductions (26–31%) consistently exceed systemic cardiac output reduction in ME/CFS tilt studies
In mast cell disorder patients, Novak et al. documented 20–24% reduction in orthostatic cerebral blood flow velocity using transcranial Doppler (Novak et al. 2022). Given the substantial overlap between mast cell activation and ME/CFS, histamine-mediated vasodilation during orthostatic stress may contribute to cerebral hypoperfusion in some patients. The combination of reduced blood volume, impaired vasoconstriction, and potentially histamine-induced vasodilation creates multiple mechanisms converging on inadequate cerebral perfusion during upright posture.
Integration with Selective Energy Dysfunction Hypothesis
The profound and consistent CBF reduction during orthostatic challenge exemplifies the broader pattern of preserved baseline function with impaired challenge response characteristic of ME/CFS (see Chapter Energy Metabolism and Mitochondrial Function Section Selective Energy Dysfunction: The CNS-Dependency Hypothesis). Resting cerebral perfusion may be adequate, but the system cannot maintain CBF during the increased demand of orthostatic stress. The brain—with its high energy demands and critical dependence on continuous perfusion—may serve as the “canary in the coal mine” for systemic energy coordination dysfunction.
See Chapter Neurological and Neurocognitive Dysfunction Section Central Catecholamine Findings: Expanding Evidence, Remaining Gaps for discussion of brain-centric pathophysiology and the role of CBF abnormalities in the broader ME/CFS disease model.