This page summarises blood, tissue, imaging and autonomic markers that published studies have reported as altered in Long COVID and in post-acute COVID-19 vaccination syndrome (PACVS). Each finding is a group-level research result, often from small studies, and none of them is a diagnostic test or reference range for an individual patient. Each entry links to the papers it comes from, and null or contested findings are labelled as such.
Spike protein & exposure markers Persistent antigen
Spike protein has been detected in blood, monocytes and tissues after mRNA vaccination, in some reports months later. Serology and sequence-level tests are proposed to separate vaccine exposure from infection.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| S1 spike subunit in CD16+ monocytes | ↑ | PACVS | Detected by flow cytometry in people with post-vaccine syndrome (n=50) vs asymptomatic vaccinated controls (n=26), up to 245 days after vaccination; prior infection excluded by history and anti-nucleocapsid testing.[1] |
| Free circulating full-length spike (post-vaccine myocarditis) | ↑ | PACVS | Unbound spike 33.9±22.4 pg/mL in 16 adolescents/young adults with post-mRNA-vaccine myocarditis; none detected in 45 vaccinated controls (p<0.0001). Myocarditis cohort, not a PACVS cohort.[2] |
| Spike protein in cerebral artery tissue | + | PACVS | Immunostaining found spike in the arterial intima of 43.8% of vaccinated hemorrhagic-stroke patients (19 cases), up to 17 months after vaccination; nucleocapsid negative. Stroke series, not a PACVS cohort.[3] |
| Anti-spike (anti-S) antibodies | + | PACVS | Positive in a PACVS case series (n=17, never infected); an exposure marker rather than a disease marker, since vaccinated people generally develop them.[4,5] |
| Anti-nucleocapsid (anti-N) antibodies | ↓ | PACVS | Expected to be negative in vaccine-only PACVS (the vaccines do not contain nucleocapsid); used to rule out prior infection.[1,5] |
| Vaccine-derived spike or mRNA sequences | + | PACVS | Proposed candidate markers (vaccine-sequence mRNA; mass-spectrometry differentiation of vaccine vs viral spike). LC-MS was used to confirm S1 in one cohort; no validated case-control assay yet.[1,5] |
GPCR & renin-angiotensin autoantibodies Autoimmunity
Antibodies against G-protein-coupled receptors and renin-angiotensin system molecules are frequently positive in PACVS series; Long COVID data are weaker and overlap with controls.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Anti-MAS1 receptor autoantibodies | + | PACVS | Positive in 16/17 (94%) PACVS patients; associated with widespread burning sensation. Case series without a control group.[4] |
| Anti-ACE2 autoantibodies | + | PACVS | Positive in 11/17 (65%); associated with ecchymosis, skin oedema and rash; proposed as a candidate PACVS biomarker.[4] |
| Anti-AT1R (angiotensin II type 1 receptor) autoantibodies | ↑ | PACVS | Reported as increased in PACVS vs the normal post-vaccination state (registry, n=191); in a case series associated with lymphadenopathy/tonsillitis.[4,6] |
| Anti-α2B-adrenergic receptor autoantibodies | ↓ | PACVS | Reported as decreased in PACVS vs the normal post-vaccination state.[6] |
| Anti-α2A-adrenergic receptor (ADRA2A) autoantibodies | ↕ | PACVS | Levels inversely associated with memory loss/brain fog in a 17-patient case series.[4] |
| Anti-STAB1 (stabilin-1) autoantibodies | + | PACVS | Associated with skin oedema and rash in a 17-patient case series.[4] |
| Anti-muscarinic M3 (CHRM3) autoantibodies | + | PACVS | Above threshold in 8/17 (47%) PACVS patients, usually together with anti-CHRM4.[4] |
| Anti-GPCR autoantibody panel (16 targets) | + | PACVS | A high share of patients were positive for several GPCR autoantibodies (most often MAS1, ACE2, CHRM4, ADRA1A, ADRB2).[4] |
| Anti-idiotype antibodies (ACE2 vs anti-S) | ↕ | PACVS | ACE2 autoantibody levels correlated with anti-S titres, consistent with a proposed anti-idiotype mechanism; hypothesis-level evidence.[4,7] |
| Anti-ETAR and anti-β2-adrenergic receptor autoantibodies | ↕ | Long COVID | No significant difference: ETAR antibodies were higher in Long COVID (n=100) than spike-antibody-negative controls, but equally raised in asymptomatic post-COVID and vaccinated controls (n=20 each); β2-AR only trended higher.[8] |
Inflammation & tryptophan-kynurenine pathway Immune dysregulation
Inflammatory cytokines are raised in both conditions. In Long COVID, tryptophan is shunted down the kynurenine pathway, one of the best-replicated metabolic findings (confirmed by meta-analysis).
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Kynurenine/tryptophan (KYN/TRP) ratio | ↑ | Long COVID | Meta-analysis of 14 studies (480 Long COVID vs 687 controls): SMD 0.755; highest in a severe Long COVID subgroup.[9,10,11] |
| Kynurenine | ↑ | Long COVID | Meta-analysis SMD 1.176 (after imputing missing studies); raised two years after infection (n=48 vs 37). One study found it normalised in PASC.[9,11,12] |
| Tryptophan | ↓ | Long COVID | Meta-analysis SMD -0.520. Severity-dependent: lower in severe Long COVID (8.81 vs 11.65 ng/mL in controls, n=20 vs 30) but not in milder cases.[9,10,12] |
| IL-6 and IL-8 | ↑ | PACVS | Elevated in >80% of a PACVS registry cohort (n=191).[6] |
| Pro-inflammatory markers (IL-1α, IL-6, TNF-α, sCD14, Flt-1) | ↑ | Long COVID | Raised in Long COVID with ME/CFS-like symptoms vs recovered and unexposed controls, 12 months after infection.[13] |
| IL-17 | ↑ | Long COVID | Higher in Long COVID patients with a more complicated course, two years after infection.[11] |
| C-reactive protein (CRP) | ↑ | Long COVID | Rises stepwise with Long COVID severity (90 patients vs 30 controls); with KYN/TRP and insulin resistance explains about 40% of symptom variance.[10] |
Energy metabolism & fatty-acid oxidation Mitochondria & metabolism
Resting plasma in Long COVID shows lower TCA-cycle and glycolytic intermediates with a build-up of acylcarnitines and free fatty acids, consistent with incomplete fat oxidation. During exercise, lactate rises early.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| TCA-cycle intermediates (citrate, succinate, malate) | ↓ | Long COVID | Lower in PASC plasma (n=29) than healthy controls (n=30) or recovered individuals (n=16).[12] |
| Lactate (resting plasma) | ↓ | Long COVID | Lower at rest in PASC plasma (n=29 vs 30 controls); the lactate/pyruvate ratio was also lower two years after infection.[11,12] |
| Pyruvate | ↕ | Long COVID | Lower in resting PASC plasma in one study, but higher two years after infection in another (adjusted p=0.01).[11,12] |
| Blood lactate during exercise (lactate threshold) | ↑ | Long COVID, PACVS | Higher lactate accumulation and lower fat oxidation during graded exercise in PASC; reviews describe a markedly reduced lactate threshold in PACVS too.[12,14,15] |
| Medium- and long-chain acylcarnitines (C8-C16, incl. C14:1, C16:1, hydroxylated) | ↑ | Long COVID | Significantly higher in PASC plasma than controls, consistent with incomplete β-oxidation.[12] |
| Free fatty acids (C10-C22, incl. very-long-chain polyunsaturated) | ↑ | Long COVID | Significantly higher in PASC plasma than controls or recovered individuals.[12] |
| Plasma ATP | ↓ | Long COVID | Reduced in Long COVID with ME/CFS-like symptoms.[13] |
| Glucose | ↑ | Long COVID | Higher two years after infection (n=48 vs 37 controls, adjusted p=5×10⁻⁵).[11] |
Amino acids, polyamines & one-carbon metabolism Mitochondria & metabolism
Homocysteine is clearly raised, and several plasma amino acids (taurine, sarcosine, serine, branched-chain amino acids, tyrosine) are lower. Polyamine findings differ between studies.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Homocysteine | ↑ | Long COVID | Mean 19.07 vs 11.31 µmol/L (62 recovered ≥3 months vs 64 controls); strongly correlated with lower cognitive (MoCA) scores (r=-0.705).[16] |
| Taurine | ↓ | Long COVID | Lower in PASC plasma (n=29 vs 30 controls) and two years after infection (n=48 vs 37, adjusted p=1.3×10⁻³).[11,12] |
| Sarcosine and serine | ↓ | Long COVID | Reduced in Long COVID and inversely correlated with depression, anxiety and cognitive-dysfunction scores; sarcosine also lower two years after infection (adjusted p=5.4×10⁻⁴).[11,13] |
| Glutamic acid (plasma) | ↓ | Long COVID | Lower two years after infection (n=48 vs 37, adjusted p=2.9×10⁻⁴); glutamine/glutamate ratio raised.[11] |
| Branched-chain amino acids (leucine/isoleucine, valine) | ↓ | Long COVID | Lower in PASC plasma than in healthy controls (n=29 vs 30).[12] |
| Tyrosine | ↓ | Long COVID | Lower in PASC plasma than in healthy controls and in recovered individuals.[12] |
| Spermine (polyamine) | ↓ | Long COVID | Lower in PASC plasma; recovered individuals had higher levels.[12] |
| Putrescine (polyamine) | ↑ | Long COVID | Higher two years after infection (adjusted p=0.01).[11] |
Mitochondrial structure & oxidative stress Mitochondria & metabolism
Small tissue studies show swollen mitochondria with disrupted cristae and altered fusion/fission and mitophagy proteins in Long COVID; circulating cell-free mitochondrial DNA is reduced. Not all oxidative-stress markers are raised.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Circulating cell-free mitochondrial DNA (ccf-mtDNA) | ↓ | Long COVID | Significantly reduced vs healthy controls; proposed as a novel Long COVID biomarker.[17] |
| Superoxide dismutase 1 (SOD1) | ↑ | Long COVID | Raised in tissue immunostaining, suggesting an oxidative-stress response (small sample, 5 patients vs 5 controls).[17] |
| MFN2 and DRP1 (fusion/fission proteins) | ↑ | Long COVID | Both raised (MFN1 and FIS1 unchanged), indicating disrupted fusion/fission balance; 5 vs 5 tissue samples.[17] |
| ATG4B (mitophagy) | ↑ | Long COVID | Raised, suggesting disrupted mitophagy; 5 vs 5 tissue samples.[17] |
| Malondialdehyde (serum) | ↕ | Long COVID | No significant difference: serum MDA did not differ between Long COVID (n=124) and ex-COVID (n=24) participants.[18] |
Nitric oxide & vascular metabolism Vascular & clotting
Markers of nitric-oxide supply suggest a smaller circulating NO reservoir in Long COVID, although arginine and ADMA are unchanged.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Serum nitrite | ↓ | Long COVID | 1.96 vs 2.56 µM in Long COVID (n=124) vs ex-COVID (n=24); inversely correlated with anxiety.[18] |
| Arginine and asymmetric dimethylarginine (ADMA) | ↕ | Long COVID | No significant difference: circulating and urinary arginine and ADMA did not differ from ex-COVID controls. Separately, lower arginine tracked with more symptoms two years after infection.[11,18] |
Lipids, iron, thyroid & insulin markers Mitochondria & metabolism
Routine blood markers that shift at group level: lower HDL-cholesterol and insulin resistance in severe Long COVID, and impaired iron storage and low-normal free T3 in PACVS.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| HDL-cholesterol | ↓ | Long COVID | 1.5 (controls, n=29) vs 1.3 mmol/L (mild/moderate) vs 1.1 mmol/L (severe Long COVID); severity-linked.[19] |
| LDL and total cholesterol | ↕ | Long COVID | No significant difference: LDL (3.1 vs 3.0 mmol/L) and total cholesterol (5.2 vs 5.1 mmol/L) did not differ from controls.[19] |
| Insulin resistance (HOMA-IR) | ↑ | Long COVID | Higher in severe Long COVID (about doubled vs mild cases and controls); insulin resistance also tracks symptom burden.[10,19] |
| Ferritin (Long COVID) | ↑ | Long COVID | About doubled in severe Long COVID (207 vs 108 µg/L in controls).[19] |
| Iron storage (ferritin index, transferrin saturation) | ↓ | PACVS | 61% of a PACVS registry cohort (n=191) had a low ferritin index with subnormal transferrin and transferrin saturation.[6] |
| Free T3 (tri-iodothyronine) | ↓ | PACVS | Low or low-normal in >80% of a PACVS registry cohort (n=191).[6] |
Brain neurochemistry & neuronal injury Neuroinflammation
Imaging studies show lower striatal dopamine-terminal markers and lower cortical GABA in Long COVID; glutamate is not clearly raised. Serotonin findings are disputed. Neurofilament light is raised in some PACVS patients.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Striatal VMAT2 binding (dopamine terminals, PET) | ↓ | Long COVID | Lower (+)[11C]DTBZ binding in 24 Long COVID vs 24 matched controls (p=4×10⁻⁵); correlated with apathy, motor slowing and memory decline.[20] |
| GABA (occipital cortex, MRS) | ↓ | Long COVID | Lower GABA+/water than matched controls; correlated with depression and poor sleep.[21] |
| Glutamate (anterior cingulate cortex, 7T MRS) | ↕ | Long COVID | No significant difference: pregenual ACC 14.12 vs 13.78 and dorsal ACC 13.43 vs 13.36 (Long COVID vs controls, n=17-24).[22] |
| Total choline (dorsal anterior cingulate, 7T MRS) | ↓ | Long COVID | Lower in Long COVID (n=25) than healthy controls (n=24).[22] |
| Serotonin (peripheral) | ↓ | Long COVID | Reported as reduced in PASC; contested, because a commentary argues the published plasma values are 30-60× too high given the method.[23,24] |
| Neurofilament light chain (sNfL) | ↑ | PACVS | Above the 90th percentile in about 27-30% of a PACVS registry cohort (n=191).[6] |
Autonomic function Autonomic & nerve
In Long COVID with POTS, orthostatic tachycardia and low end-tidal CO2 are common, but many autonomic measures overlap with non-COVID POTS and controls.
| Analyte | Change | Condition | Notes |
|---|---|---|---|
| Orthostatic heart-rate increment (head-up tilt) | ↑ | Long COVID | ΔHR 42±18.6 in Long COVID POTS (n=43) vs 25±12.9 in post-COVID controls (n=27). Preprint.[25] |
| End-tidal CO2 at end-tilt | ↓ | Long COVID | 22.1 vs 29.7 in controls (hypocapnia). Preprint.[25] |
| Plasma norepinephrine | ↕ | Long COVID | No significant difference: mean norepinephrine was similar to controls, although 42.5% of Long COVID POTS patients had a hyperadrenergic response on tilt. Preprint.[25] |
References
- Patterson BK, Yogendra R, Francisco EB, et al.. Detection of S1 spike protein in CD16+ monocytes up to 245 days in SARS-CoV-2-negative post-COVID-19 vaccine syndrome (PCVS) individuals. Hum Vaccin Immunother 21:2494934, 2025. doi:10.1080/21645515.2025.2494934
- Yonker LM, Swank Z, Bartsch YC, et al.. Circulating Spike Protein Detected in Post-COVID-19 mRNA Vaccine Myocarditis. Circulation 147:867-876, 2023. doi:10.1161/circulationaha.122.061025
- Ota N, Itani M, Aoki T, et al.. Expression of SARS-CoV-2 spike protein in cerebral Arteries: Implications for hemorrhagic stroke Post-mRNA vaccination. J Clin Neurosci 136:111223, 2025. doi:10.1016/j.jocn.2025.111223
- Mantovani M, Bellavite P, Fazio S, et al.. Autoantibodies Targeting G-Protein-Coupled Receptors and RAS-Related Molecules in Post-Acute COVID Vaccination Syndrome: A Retrospective Case Series Study. Biomedicines 12:2852, 2024. doi:10.3390/biomedicines12122852
- Halma M, Varon J. Restoring trust in vaccination: listening to patients and acknowledging Post-Acute COVID Vaccine Syndrome. Front Med (Lausanne) 12:1688170, 2025. doi:10.3389/fmed.2025.1688170
- Mundorf AK, Semmler A, Heidecke H, et al.. Clinical and Diagnostic Features of Post-Acute COVID-19 Vaccination Syndrome (PACVS). Vaccines (Basel) 12:790, 2024. doi:10.3390/vaccines12070790
- Federico M. The Immunologic Downsides Associated with the Powerful Translation of Current COVID-19 Vaccine mRNA Can Be Overcome by Mucosal Vaccines. Vaccines (Basel) 12:1281, 2024. doi:10.3390/vaccines12111281
- Han E, Müller-Zlabinger K, Hasimbegovic E, et al.. Circulating Autoantibodies Against Vasoactive Biomarkers Related to Orthostatic Intolerance in Long COVID Patients Compared to No-Long-COVID Populations: A Case-Control Study. Biomolecules 15:300, 2025. doi:10.3390/biom15020300
- Almulla AF, Thipakorn Y, Zhou B, et al.. The tryptophan catabolite or kynurenine pathway in long COVID disease: A systematic review and meta-analysis. Neuroscience 563:268-277, 2024. doi:10.1016/j.neuroscience.2024.10.021
- Al-Hakeim HK, Khairi Abed A, Rouf Moustafa S, et al.. Tryptophan catabolites, inflammation, and insulin resistance as determinants of chronic fatigue syndrome and affective symptoms in long COVID. Front Mol Neurosci 16:1194769, 2023. doi:10.3389/fnmol.2023.1194769
- López-Hernández Y, Monárrez-Espino J, López DAG, et al.. The plasma metabolome of long COVID patients two years after infection. Sci Rep 13:12420, 2023. doi:10.1038/s41598-023-39049-x
- Guntur VP, Nemkov T, de Boer E, et al.. Signatures of Mitochondrial Dysfunction and Impaired Fatty Acid Metabolism in Plasma of Patients with Post-Acute Sequelae of COVID-19 (PASC). Metabolites 12:1026, 2022. doi:10.3390/metabo12111026
- Saito S, Shahbaz S, Luo X, et al.. Metabolomic and immune alterations in long COVID patients with chronic fatigue syndrome. Front Immunol 15:1341843, 2024. doi:10.3389/fimmu.2024.1341843
- Gracidas C, Levy R, Varon J, et al.. Lactate, Capnia, and Fat Oxidation as Therapeutic Axes for SARS-CoV-2 Spike Protein-Induced Sequelae. Horm Metab Res 58:90-102, 2026. doi:10.1055/a-2794-9646
- Halma M, Varon J. Metabolic modulation as a therapeutic strategy for post-acute vaccination syndrome (PACVS): A review of pathomechanisms and existing therapeutic components. Biomed Pharmacother 193:118864, 2025. doi:10.1016/j.biopha.2025.118864
- Oner P, Yilmaz S, Doğan S. High Homocysteine Levels Are Associated with Cognitive Impairment in Patients Who Recovered from COVID-19 in the Long Term. J Pers Med 13:503, 2023. doi:10.3390/jpm13030503
- Szögi T, Borsos BN, Masic D, et al.. Novel biomarkers of mitochondrial dysfunction in Long COVID patients. Geroscience 47:2245-2261, 2025. doi:10.1007/s11357-024-01398-4
- Mikuteit M, Baskal S, Klawitter S, et al.. Amino acids, post-translational modifications, nitric oxide, and oxidative stress in serum and urine of long COVID and ex COVID human subjects. Amino Acids 55:1173-1188, 2023. doi:10.1007/s00726-023-03305-1
- Al-Zadjali J, Al-Lawati A, Al Riyami N, et al.. Reduced HDL-cholesterol in long COVID-19: A key metabolic risk factor tied to disease severity. Clinics (Sao Paulo) 79:100344, 2024. doi:10.1016/j.clinsp.2024.100344
- Liu YK, Persaud D, Vieira EL, et al.. Loss of vesicular monoamine transporter 2 in striatum of long COVID and relationship to neuropsychiatric symptoms. EBioMedicine 130:106339, 2026. doi:10.1016/j.ebiom.2026.106339
- Marinkovic K, White DR, Alderson Myers A, et al.. Cortical GABA Levels Are Reduced in Post-Acute COVID-19 Syndrome. Brain Sci 13:1666, 2023. doi:10.3390/brainsci13121666
- Godlewska BR, Sylvester AL, Emir UE, et al.. Brain and muscle chemistry in myalgic encephalitis/chronic fatigue syndrome (ME/CFS) and long COVID: a 7T magnetic resonance spectroscopy study. Mol Psychiatry 30:5215-5226, 2025. doi:10.1038/s41380-025-03108-8
- Wong AC, Devason AS, Umana IC, et al.. Serotonin reduction in post-acute sequelae of viral infection. Cell 186:4851-4867.e20, 2023. doi:10.1016/j.cell.2023.09.013
- Anderson GM, Cook EH, Blakely RD, et al.. Long COVID-19 and Peripheral Serotonin: A Commentary and Reconsideration. J Inflamm Res 17:2169-2172, 2024. doi:10.2147/jir.s456000 Counter-evidence
- Larsen NW, Taha HB, Seliger J, et al.. Deep Phenotyping Long-COVID Postural Tachycardia Syndrome. medRxiv (preprint), 2025. doi:10.1101/2025.04.28.25326587 Preprint
Research findings at group level. They are not diagnostic reference ranges for individuals.