This page summarises published research on individual dietary supplements and nutraceuticals that have been studied, or proposed on mechanistic grounds, for Long COVID and post-COVID-vaccination syndrome (PACVS). Each ingredient is described on its own: what it does biologically, which proposed disease mechanism it relates to, and what the clinical evidence currently shows. Much of the evidence is early. It comes from small trials, from acute COVID-19 or other conditions such as ME/CFS, critical illness and heart failure, or from laboratory and metabolomic findings that have not yet been tested in controlled trials in these populations. Doses are reported only as they appear in the cited studies. They are not recommendations.
Safety. This content is for education only and is not medical advice. Supplements can interact with medicines and other supplements, and can be unsuitable for people with kidney or liver disease, during pregnancy or breastfeeding, or alongside some prescription drugs. Doses listed are those reported in published studies, not recommended intakes. Talk to a qualified healthcare professional before starting, stopping or combining any supplement.
Conditionally essential amino acid · Gut barrier integrity; immune-cell fuel; glutathione precursor
- How it may act
- Primary fuel for rapidly dividing immune cells (lymphocytes, macrophages) and intestinal epithelial cells; conditionally essential during illness
- Maintains tight junctions of the intestinal epithelium; depletion may allow bacterial products to translocate and amplify systemic inflammation
- Rate-limiting substrate for glutathione synthesis
- Feeds alpha-ketoglutarate into the TCA cycle
- Long COVID
- Plasma glutamine was lower in the long-COVID phase, and urea-cycle metabolism was among the most dysregulated pathways in a subgroup, about 2 years after infection (López-Hernández et al. 2023). In acute COVID-19, a non-randomised comparison found glutamine 10 g three times daily was associated with shorter hospitalisation (Cengiz et al. 2020). No verified interventional Long COVID trials of glutamine identified.
PACVS, other evidence, doses & safety
- PACVS
- No PACVS-specific studies cited.
- ME/CFS
- Gut microbiome disruption and intestinal permeability are reported in post-COVID and ME/CFS cohorts. No glutamine intervention trial in ME/CFS cited.
- Other conditions
- In a small RCT in severely burned patients, IV glutamine reduced Gram-negative bacteremia (8% vs 43%) and markers of inflammation (Wischmeyer et al. 2001).
- Studied doses
- 10 g three times daily, oral (acute COVID-19; Cengiz et al. 2020)
- Continuous IV infusion (burn ICU RCT; Wischmeyer et al. 2001)
- Safety
- Check standard references, especially for people with liver or kidney disease.
References [1,2,3]
Non-proteinogenic amino acid (urea-cycle intermediate; nitric oxide precursor) · Nitric oxide production; vascular function; arginine recycling
- How it may act
- Converted to arginine in the kidney, which nitric oxide synthase (NOS) then uses to make nitric oxide (NO)
- Avoids first-pass intestinal and hepatic arginase breakdown, so it raises plasma arginine and NO more steadily than oral arginine
- NO is a vasodilator with anti-platelet and anti-inflammatory signalling effects; reduced NO is a feature of endothelial dysfunction
- Reported to reduce arterial stiffness, improve microvascular flow, inhibit platelet aggregation and endothelial activation, and reduce exercise-induced fatigue
- Preclinical data in mice: 250 mg/kg daily for 15 days increased skeletal-muscle PGC-1α (Villareal et al. 2018)
- Long COVID
- Endothelial dysfunction and reduced NO production are reported in Long COVID. An increased ornithine/citrulline ratio, with urea-cycle metabolism among the five most dysregulated pathways, was reported in one PASC subgroup (López-Hernández et al. 2023). No verified citrulline intervention studies in Long COVID identified. Citrulline has been tested in acute COVID-19 respiratory failure and ARDS (NCT04570384, IV L-citrulline, n=65; NCT04404426, CACOLAC, n=33; both completed).
PACVS, other evidence, doses & safety
- PACVS
- Endothelial dysfunction is described in PACVS. No PACVS-specific intervention studies cited.
- ME/CFS
- No ME/CFS-specific studies cited.
- Other conditions
- In healthy volunteers, oral L-citrulline raised plasma arginine and the L-arginine/ADMA ratio more effectively than L-arginine (Schwedhelm et al. 2008). In trained men, 2.4 g/day for 7 days modestly improved cycling time-trial performance (Suzuki et al. 2016).
- Studied doses
- Up to 3 g twice daily for 1 week (healthy volunteers; Schwedhelm et al. 2008)
- 2.4 g/day for 7 days (Suzuki et al. 2016, cycling performance)
- 250 mg/kg/day for 15 days (mice; Villareal et al. 2018)
- Safety
- As an NO precursor it may lower blood pressure. Take care alongside antihypertensives, nitrates or PDE5 inhibitors.
- Studied alongside L-arginine as a precursor in the same NO pathway (Schwedhelm et al. 2008).
References [1,4,5,6,7,8,9]
Conditionally essential amino acid (direct NOS substrate) · Direct nitric oxide substrate; creatine synthesis; immune modulation
- How it may act
- Direct substrate for nitric oxide synthase
- Supplies the guanidino group for creatine biosynthesis
- Takes part in the urea cycle for nitrogen disposal
- Modulates macrophage polarisation and T-cell activity
- Arginase and eNOS compete for arginine (the 'arginine paradox'; Elms et al. 2013). About 31–38% of dietary arginine is taken up by the splanchnic region on first pass in healthy adults (Castillo et al. 1993)
- Long COVID
- One single-blind RCT tested L-arginine 1.66 g plus liposomal vitamin C 500 mg twice daily against placebo for 28 days in adults with Long COVID and persistent fatigue (50 randomised, 46 completed). The active group improved 6-minute walk distance (+30 m vs +0 m, p=0.001), handgrip strength and flow-mediated dilation; fatigue at day 28 was reported by 2 (8.7%) vs 21 placebo participants (p<0.0001) (Tosato et al. 2022; NCT04947488). The vitamin C and arginine effects cannot be separated. A secondary analysis reported increased serum L-arginine and L-arginine/ADMA ratio vs placebo (Calvani et al. 2023).
PACVS, other evidence, doses & safety
- PACVS
- No PACVS-specific studies cited.
- ME/CFS
- No ME/CFS-specific studies cited.
- Other conditions
- None
- Studied doses
- 1.66 g twice daily with 500 mg liposomal vitamin C, for 28 days (Long COVID RCT; Tosato et al. 2022)
- Safety
- Substantial first-pass splanchnic uptake limits how much oral arginine reaches plasma (Castillo et al. 1993). Citrulline raised plasma arginine more effectively in a head-to-head study (Schwedhelm et al. 2008).
- As an NO precursor it may lower blood pressure.
References [4,5,10,11,12,13]
Acetylated cysteine derivative (glutathione precursor; mucolytic) · Glutathione replenishment; antioxidant; anti-inflammatory
- How it may act
- Replenishes intracellular cysteine, the rate-limiting substrate for glutathione synthesis
- Scavenges reactive oxygen species directly
- Inhibits NF-κB-mediated inflammatory gene expression
- Reduces platelet aggregation and hyperactivation
- Mucolytic
- Computational docking suggests NAC binds the SARS-CoV-2 main protease (Mpro) catalytic site (in silico only)
- Long COVID
- A retrospective case series (n=9) reported von Willebrand factor normalising in 3 of 3 patients taking oral NAC (600–1200 mg twice daily) vs 0 of 6 controls, with improvements in breathlessness, brain fog and fatigue (Bellone, Siegel & Santin 2025). Raised plasma acylcarnitines and impaired fatty-acid oxidation in PASC are part of the metabolic rationale (Guntur et al. 2022). No verified controlled NAC trials in Long COVID identified.
PACVS, other evidence, doses & safety
- PACVS
- No PACVS intervention studies cited. A systematic review reported vaccine-derived mRNA or spike protein detected in bodily fluids and tissues, in blood up to 709 days after vaccination (Selem, Raszek, Varon & Halma 2026; OSMF co-authored). This is context for the persistent-antigen hypothesis, not evidence for NAC.
- ME/CFS
- Glutathione depletion and raised oxidative stress are reported in ME/CFS.
- Other conditions
- Decades of clinical use in paracetamol (acetaminophen) overdose, COPD and idiopathic pulmonary fibrosis. In acute COVID-19, a meta-analysis of 12 RCTs (1,125 patients in the mortality analysis) reported lower mortality (RR 0.59) (Varikasuvu et al. 2025, CoViNAC). A double-blind RCT preprint (n=63) reported faster SGRQ improvement with NAC 600 mg twice daily for 6 months after hospital admission for acute COVID-19, not Long COVID (Li et al. 2025, preprint).
- Studied doses
- 600–1200 mg twice daily (Long COVID case series; Bellone et al. 2025)
- 600 mg twice daily for 6 months (acute COVID-19 RCT preprint; Li et al. 2025)
- Safety
- Long record of clinical use, including as the antidote for paracetamol (acetaminophen) overdose.
- Because of its anti-platelet effect, take care alongside anticoagulants or antiplatelet drugs. NAC may also potentiate nitroglycerin.
References [14,15,16,17,18,19]
Water-soluble vitamin (antioxidant; enzyme co-factor) · Antioxidant; carnitine-synthesis co-factor; endothelial support
- How it may act
- Main water-soluble antioxidant in plasma; scavenges ROS and regenerates vitamin E in lipid membranes
- Essential co-factor for two hydroxylation steps in endogenous L-carnitine biosynthesis (Rebouche 1991)
- Drives collagen synthesis through prolyl and lysyl hydroxylases, supporting endothelial repair
- Reduces oxidative inactivation of nitric oxide, which may enhance NO-pathway effects
- Supports neutrophil, lymphocyte and epithelial-barrier function (Carr & Maggini 2017)
- Long COVID
- The main Long COVID evidence is the single-blind RCT of L-arginine plus liposomal vitamin C (500 mg twice daily) for 28 days (50 randomised): fatigue at day 28 in 2 (8.7%) vs 21 placebo participants, p<0.0001, with better walking distance and endothelial function (Tosato et al. 2022; NCT04947488). The vitamin C component cannot be separated from the arginine.
PACVS, other evidence, doses & safety
- PACVS
- No PACVS-specific studies cited.
- ME/CFS
- No ME/CFS-specific studies cited.
- Other conditions
- No verified clinical studies in other conditions cited. Vitamin C is reviewed as supporting immune function in infection (Carr & Maggini 2017).
- Studied doses
- 500 mg liposomal vitamin C twice daily with 1.66 g L-arginine, for 28 days (Long COVID RCT; Tosato et al. 2022)
- Safety
- Higher oral doses can cause gastrointestinal upset.
- High doses may raise the risk of oxalate kidney stones and can interfere with some lab tests, including glucose meters. Avoid high-dose IV use in G6PD deficiency.
References [10,13,20,21]
Amino-acid derivative (phosphagen energy buffer) · Rapid ATP buffering; energy support independent of mitochondria; cognitive support
- How it may act
- Phosphocreatine regenerates ATP from ADP in under a second without involving mitochondria, giving an alternative energy route when mitochondrial function is impaired
- Reduces lactate accumulation during exertion
- Neuroprotective: maintains neuronal ATP under metabolic stress
- Supports mitochondrial membrane potential and mitochondrial biogenesis
- Long COVID
- Several small RCTs. A 6-month double-blind RCT at 4 g/day (n=12, 6 per group) raised muscle and brain creatine, reduced general fatigue at 3 months vs baseline and improved several symptoms, including loss of taste, body aches and concentration, at 6 months (Slankamenac et al. 2023). An 8-week RCT (n=15) of creatine 8 g/day with or without glucose found significant symptom improvements (body aches, breathing problems, concentration, headache) in the creatine–glucose group (Slankamenac et al. 2024). A single-blind pilot RCT (n=67) of 6 g/day, 18 g/day or placebo for 4 weeks found fatigue improved by 2.05 points (95% CI −3.47 to −0.63; p=0.005) in the 6 g/day arm only (Dos Santos et al. 2026). CPET in PASC shows reduced fat oxidation and altered lactate production, which is the mechanistic context (de Boer et al. 2022).
PACVS, other evidence, doses & safety
- PACVS
- Post-exertional malaise is described as a central feature of PACVS. No PACVS-specific creatine trials cited.
- ME/CFS
- Proposed for post-exertional malaise on mechanistic grounds. No verified ME/CFS trials cited.
- Other conditions
- Large safety and efficacy literature in healthy and clinical populations for exercise performance, fatigue and cognition.
- Studied doses
- 4 g/day for 6 months (Long COVID RCT; Slankamenac et al. 2023)
- 8 g/day for 8 weeks, with or without 3 g glucose (Long COVID RCT; Slankamenac et al. 2024)
- 6 g/day or 18 g/day for 4 weeks (Long COVID pilot RCT; Dos Santos et al. 2026)
- Safety
- Among the most extensively safety-tested nutritional supplements.
- Creatine can raise serum creatinine without any change in kidney function, which can confuse kidney-function tests. Use caution in existing kidney disease.
References [22,23,24,25]
Acetylated amino-acid derivative (carnitine ester) · Fatty-acid oxidation; mitochondrial metabolism; neuronal support
- How it may act
- Carnitine carries long-chain fatty acids across the inner mitochondrial membrane for beta-oxidation
- The acetyl form crosses the blood-brain barrier and donates acetyl groups for acetylcholine synthesis and neuronal metabolism
- Reduces oxidative damage to mitochondrial membranes and supports complex I activity
- Long COVID
- Mixed. In a post-COVID case-control study, ALCAR 500 mg plus rehabilitation improved pain, quality of life and depression more than rehabilitation alone, with no significant difference in breathlessness or functional capacity (Scaturro et al. 2022). Raised plasma acylcarnitines and free fatty acids in PASC suggest impaired fatty-acid oxidation (Guntur et al. 2022), and fat oxidation during graded exercise is reduced (de Boer et al. 2022). In hospitalised mild-to-moderate acute COVID-19, L-carnitine 3 g/day for 5 days lowered inflammatory markers, with no deaths vs 6 in controls (Talebi et al. 2022).
PACVS, other evidence, doses & safety
- PACVS
- No PACVS-specific studies cited.
- Studied doses
- 500 mg ALCAR alongside rehabilitation (post-COVID case-control study; Scaturro et al. 2022)
- 3 g/day L-carnitine in three divided doses for 5 days (acute COVID-19; Talebi et al. 2022)
- Safety
- Carnitine products may interact with warfarin and thyroid hormone therapy.
References [15,25,26,27]
Sulfur-containing amino acid (non-proteinogenic; osmolyte) · Mitochondrial membrane stability; cardioprotection; cytoprotection
- How it may act
- Stabilises the inner mitochondrial membrane and reduces ROS production by the electron transport chain
- Modulates calcium handling in cardiac and skeletal muscle, which may be relevant to arrhythmias and dysautonomia/POTS; has anti-arrhythmic effects on cardiac conduction
- Anti-inflammatory through inhibition of neutrophil ROS production
- Intracellular osmolyte that protects cells under metabolic stress; conjugates bile acids
- Long COVID
- In a cohort followed for 6 months, plasma taurine was negatively associated with post-COVID symptoms and was lower in people with more than three symptoms than in those who recovered; rising taurine was associated with fewer adverse clinical events (Khoramjoo et al. 2024). No verified taurine intervention trials in Long COVID identified.
PACVS, other evidence, doses & safety
- PACVS
- Proposed on mechanistic grounds for mitochondrial, cardiac and autonomic symptoms. No PACVS-specific studies cited.
- ME/CFS
- No ME/CFS-specific studies cited.
- Other conditions
- In prehypertension, 1.6 g/day for 12 weeks lowered blood pressure, improved vascular function and raised plasma taurine (Sun et al. 2016). Multi-species work linked declining taurine to ageing, and supplementation reduced markers of mitochondrial dysfunction and inflammation in animals (Singh et al. 2023).
- Studied doses
- 1.6 g/day for 12 weeks (prehypertension RCT; Sun et al. 2016)
- Safety
- No specific safety data summarised here. Check interactions with a pharmacist, especially with liver or kidney disease.
References [28,29,30]
NAD+ precursor (vitamin B3 derivative) · NAD+ repletion; mitochondrial biogenesis; sirtuin activation
- How it may act
- Direct precursor to NAD+, which carries electrons in mitochondrial complexes I and III
- Substrate for sirtuins (SIRT1/SIRT3), which regulate inflammation, mitochondrial biogenesis and stress responses
- Supports PARP-mediated DNA repair
- Long COVID
- No verified NMN studies in Long COVID identified. Indirect evidence: an RCT of nicotinamide riboside (2000 mg/day), a related NAD+ precursor, raised NAD+ 2.6–3.1-fold but did not improve cognition, fatigue, sleep or mood vs placebo; exploratory within-group benefits appeared after 10 weeks, and dropout was high (Wu et al. 2025; NCT04809974). A single-arm pilot of NAD+ iontophoresis plus low-dose naltrexone reported Chalder fatigue scores falling from 25.9 to 17.4 (p<0.0001), with 52% responders, but the NAD+ effect cannot be separated (Isman et al. 2024).
PACVS, other evidence, doses & safety
- PACVS
- No PACVS-specific studies cited.
- ME/CFS
- No ME/CFS-specific studies cited.
- Other conditions
- In overweight or obese postmenopausal women with prediabetes, NMN 250 mg/day for 10 weeks increased muscle insulin sensitivity and insulin signalling (Yoshino et al. 2021).
- Studied doses
- 250 mg/day for 10 weeks (Yoshino et al. 2021)
- Nicotinamide riboside 2000 mg/day (related precursor; Long COVID RCT; Wu et al. 2025)
- Safety
- Regulatory status of NMN as a supplement differs by country.
References [31,32,33,34]
Conditionally essential amino acid · Neuronal membrane synthesis; myelin support; one-carbon metabolism
- How it may act
- Precursor to phosphatidylserine (a neuronal membrane phospholipid), glycine, cysteine and sphingolipids
- Required for myelin sphingolipid synthesis and normal glial function; serine deficiency causes progressive peripheral neuropathy and cognitive impairment
- Feeds the one-carbon/folate cycle that supports methylation, neurotransmitter synthesis and DNA repair
- Long COVID
- Serine reported depleted in PASC, with levels inversely correlated with neuropsychiatric symptom scores (Saito et al. 2024). No interventional Long COVID trials cited.
PACVS, other evidence, doses & safety
- PACVS
- Small fibre neuropathy and cognitive impairment are reported as common in PACVS cohorts. No serine intervention studies in PACVS cited.
- ME/CFS
- Some ME/CFS cohort studies report reduced plasma serine. No specific citation given.
- Other conditions
- L-serine has been tested for safety in ALS at doses from 0.5 g to 15 g twice daily (NCT01835782; status unknown, estimated enrolment 20). No verified results are cited.
- Studied doses
- 0.5–15 g twice daily (L-serine; ALS safety study NCT01835782)
- Safety
- No verified safety data are cited for L-serine here. Check standard references.
References [35,36]