Inflammatory Cytokine-Induced Somnolence and Fatigue
Pro-inflammatory cytokines (IL-1\(\beta\), IL-6, TNF-\(\alpha\), IFN-\(\alpha/\gamma\)) directly induce fatigue and somnolence through central mechanisms independent of peripheral tissue damage. This section reviews the cytokine-fatigue literature, the specific profile documented in ME/CFS (elevated IL-1\(\beta\), IL-6, IFN-\(\alpha\) in subsets), and the downstream signaling pathways by which circulating cytokines access the brain to suppress arousal circuits and activate fatigue-signaling pathways.
1 Cytokine-to-Brain Signaling Routes
Peripheral cytokines communicate with the brain through three complementary routes (Dantzer et al. 2000) (McCusker and Kelley 2013). The humoral pathway operates via circumventricular organs—notably the organum vasculosum of the lamina terminalis (OVLT) and the area postrema—where an incomplete blood-brain barrier permits direct cytokine diffusion into adjacent hypothalamic tissue. A saturable transport pathway operates through carrier-mediated influx systems at the BBB endothelium for IL-1\(\beta\), IL-6, and TNF-\(\alpha\) (McCusker and Kelley 2013). The neural pathway exploits vagal afferents: cytokines at peripheral infection sites activate cytokine receptors on paraganglia of the vagus nerve, transmitting immune signals to the nucleus tractus solitarius and thence to hypothalamic circuits within minutes (Dantzer et al. 2000) (Huerta, Masters, and Matheny 2025).
At the BBB endothelium and within perivascular spaces, circulating cytokines trigger prostaglandin E2 (PGE2) synthesis, which diffuses into the hypothalamic parenchyma to activate EP3 receptors, suppressing wake-active orexin neurons and promoting sleep-pressure circuits. Microglia—the brain’s resident immune cells—amplify this signal via TLR4/NF-\(\kappa\)B activation, producing local IL-1\(\beta\), IL-6, and TNF-\(\alpha\) that sustain the fatigue state long after peripheral cytokine levels normalize (McCusker and Kelley 2013).
2 Quantitative Human Cytokine-Fatigue Relationships
Vollmer-Conna et al. (Vollmer-Conna et al. 2004) demonstrated in prospective cohorts with documented EBV, Q fever, and Ross River virus infection that IL-1\(\beta\) and IL-6 levels correlate directly and quantitatively with fatigue severity, malaise, pain, mood disturbance, and cognitive slowing. Higher peripheral cytokine burdens predict worse sickness symptom scores, establishing a dose-dependent relationship in humans analogous to animal model data (study: multiple infection cohorts, natural infection setting, certainty: Medium-High). However, Omdal et al. (2026, n=96) found that severely fatigued long-COVID patients (fVAS 63 vs 5) show completely normal peripheral cytokines (Omdal et al. 2026), indicating that the cytokine-fatigue correlation is state-dependent: cytokines drive fatigue during the acute phase of infection but may normalize while fatigue persists through CNS-confined maintenance mechanisms. Both findings are compatible under a biphasic model where acute cytokine surge initiates hypothalamic microglial priming, which then sustains fatigue independently of ongoing peripheral cytokine elevation. Capuron et al. (Capuron et al. 2002) demonstrated in a double-blind randomized trial (n=40 melanoma patients receiving IFN-\(\alpha\) therapy) that fatigue and anorexia emerge within two weeks of cytokine exposure through mechanisms distinct from serotonergic pathways: the neurovegetative syndrome (fatigue, somnolence, anorexia) was substantially less responsive to paroxetine than the concurrent mood syndrome. This dissociation indicates that IFN-\(\alpha\)-induced fatigue operates via non-monoaminergic circuits, consistent with direct hypothalamic cytokine signaling (RCT, n=40, certainty: High). ## TLR4/NF-κB as PEM Amplifier
ME/CFS patients may exhibit exaggerated and prolonged TLR4/NF-\(\kappa\)B activation following physical exertion, driving the post-exertional cytokine surge that underlies PEM. Light et al. (Light et al. 2009) demonstrated that moderate exercise elicits significantly greater increases in leukocyte TLR4 gene expression in CFS patients versus controls, with elevations persisting 48 hours post-exercise and correlating with fatigue and pain severity (n=19 CFS, n=18 controls). Moneghetti et al. (Moneghetti et al. 2018) independently documented that IL-1\(\beta\) and IFN-\(\alpha\) are specifically elevated 18 hours post-exercise in ME/CFS but not sedentary controls (n=24 vs. n=24), providing the cytokine correlate of the gene expression changes. Che et al. (Che et al. 2025) further demonstrated that this heightened innate immune response worsens after exercise in ME/CFS, implicating TLR4/NF-\(\kappa\)B drive as a central mechanism.
Testable prediction: TLR4 blockade (e.g., TAK-242) or anti-IL-1\(\beta\) therapy (anakinra (Roerink et al. 2017)) before planned exertion should attenuate or delay PEM onset, with effect size proportional to pre-exercise TLR4 expression.
Limitations: Both studies use small samples; the causal direction (TLR4 drives PEM vs. PEM drives TLR4) remains unresolved; peripheral leukocyte TLR4 is a proxy for microglial TLR4 activity. Certainty: Medium (single studies, small n, no replication by independent groups at time of writing).
3 ME/CFS-Specific Cytokine Profile: Evidence and Limitations
Several large studies document cytokine abnormalities in ME/CFS subsets. Hornig et al. (Hornig et al. 2015) identified distinct plasma immune signatures early in illness (elevated pro-inflammatory cytokines) that shift toward an exhaustion pattern in longer-duration disease. Montoya et al. (Montoya et al. 2017) correlated 17 cytokines with disease severity in 192 CFS patients, with IL-17F and TGF-\(\beta\) showing the strongest severity correlations. Giloteaux et al. (Giloteaux et al. 2023) conducted comprehensive proteomics analysis, identifying dysregulated cytokine networks in ME/CFS.
However, a systematic review of 15 case-control studies (Corbitt et al. (Corbitt et al. 2019), screening 16,702 publications) concluded that cytokine findings are heterogeneous and inconclusive as diagnostic markers. No universal cytokine signature has been validated. Elevations in IL-1\(\beta\), IL-6, and IFN-\(\alpha/\gamma\) are documented in subsets of ME/CFS patients and should not be overstated as universal features.
This heterogeneity is consistent with ME/CFS being a syndrome of multiple convergent pathomechanisms: cytokine-driven fatigue may predominate in some patient subsets while other mechanisms (adenosine dysregulation, mitochondrial dysfunction, central sensitization) predominate in others.
Fatigue Without Detectable Peripheral Inflammation. Omdal et al. (2026, n=96) directly tested whether fatigue severity in long COVID correlates with peripheral inflammatory biomarkers — and found a complete dissociation: fVAS 63 vs 5 (p = 0.001) despite zero differences in CRP, TNF-α, IL-6, HSP90α, Serpin F1, hemopexine, or APOA4 (Omdal et al. 2026). Multivariable regression confirmed no biomarker-fatigue association. This null is important because it constrains the cytokine-fatigue model: if peripheral cytokines are not elevated in severely fatigued patients, then either (a) fatigue is driven by CNS-confined inflammation not reflected in blood, (b) fatigue is maintained by non-inflammatory mechanisms (epigenetic, metabolic, predictive processing), or (c) cytokines were elevated earlier and triggered persistent downstream effects after normalising. The sickness behavior model (Section Sickness Behavior as Overarching Integrative Framework) and the locked-sickness-behavior framework predict option (c): hypothalamic microglial priming, once established by an acute cytokine storm, maintains fatigue signalling via CNS circuits independently of ongoing peripheral immune activation. This CNS-confined, circuit-mediated account is independently reinforced by Komaroff and Dantzer (2025), who propose that symptom persistence in long COVID and ME/CFS is generated by dedicated sickness-behavior and torpor neural circuits downstream of (localized) neuroinflammation rather than by ongoing systemic cytokines (Komaroff and Dantzer 2025).