We graded 12 mitochondrial-supplement claims — here’s what holds up
·Walk into any wellness clinic or scroll any longevity feed and you’ll hear the same promises: this precursor “boosts mitochondrial function,” that capsule “creates new mitochondria,” this drip “repairs your cells.” We took the 12 most repeated marketing claims about supplements and mitochondrial health — drawn from product labels, clinic marketing, and social media, not straw men — and graded each one against the human evidence.
The rule was simple: a claim that something “supports mitochondrial health” is only meaningful if it changes something measurable — mitochondrial respiration, ATP output, mitochondrial protein content, mitophagy, or a clinical outcome tied to energy metabolism. Mechanism isn’t evidence. Animal results aren’t human results. And a blood marker isn’t mitochondrial function. Here’s what holds up — and what doesn’t. We revisit and update this report every quarter as new human trials publish.
How we graded the claims
- Claim inventory: the 12 most-repeated marketing claims about supplements and mitochondrial health, drawn from product labels, clinic marketing, and social media — not straw men.
- Evidence standard: each claim graded on our five-label system — Established / Preliminary / Extrapolated / Contested / Absent — with one rule: a claim “supports mitochondrial health” is only meaningful if it changes something measurable (mitochondrial respiration, ATP output, mitochondrial protein content, mitophagy, or a clinical outcome tied to energy metabolism).
- Weighting: human randomized trials with direct mitochondrial endpoints (muscle biopsies, respirometry) weighted above trials with blood biomarkers; animal and cell data flagged as such and never counted as human evidence; company funding disclosed on every industry-sponsored trial.
- Cutoff: literature reviewed through our 2026 research program, anchored by the NAD+ evidence synthesis (Gallagher et al. 2026 systematic review, 113 studies) and our evidence-graded supplement guide.
The 12 claims, ranked
Ranked from weakest to strongest evidence. Nine of the twelve grade out as Absent — the marketing is doing the work that the human trials never did.
1. Oral NR/NMN “boosts mitochondrial function” Absent
Four human muscle-biopsy randomized trials (Dollerup 2020, Remie 2020, Elhassan 2019, Stocks 2021) confirmed target engagement via NAD+ metabolites — whole-blood NAD+ rose >2-fold in Elhassan 2019, while muscle or urinary NAD+ metabolites moved in Dollerup, Remie, and Stocks, which did not measure blood NAD+ — yet mitochondrial respiration, mitochondrial content, and bioenergetics did not improve in any of them. (One open-label exception: a 5-month twin trial (Lapatto et al. 2023, Science Advances) did report improved mitochondrial biogenesis.) Raising a blood marker is not the same as improving mitochondrial function.
2. Oral NAD+ capsules “replenish cellular NAD+” Absent
On October 13, 2025, the US National Advertising Division found there is no human clinical evidence that orally ingested NAD+ raises NAD+ levels in the body — “liposomal or otherwise.”
3. IV NAD+ for detox, addiction, and “cellular repair” Absent
Grant 2019 showed infused NAD+ is rapidly cleared and metabolized, with no mitochondrial or clinical endpoints established. The addiction-treatment claims rest on an unpublished, uncontrolled pilot — not a trial.
4. Blood NAD+ tests reveal your “mitochondrial health” Absent
Blood NAD+ is a poor proxy for tissue and mitochondrial NAD+, and it degrades rapidly in poorly processed samples. A 2026 multi-cohort analysis (Trętowicz et al., Nature Metabolism) found whole-blood NAD+ remarkably stable across age and lifestyle interventions — challenging blood NAD+ as a biomarker of aging.
5. Resveratrol “activates longevity pathways” Absent
In a human trial at 300 mg, resveratrol did not activate the AMPK/SIRT1 “longevity axis” in muscle or fat — and it actually inhibited mitochondrial respiration in ex vivo muscle. This one isn't just unproven; the human data point the other way.
6. PQQ “creates new mitochondria” Absent
At 20 mg/day combined with training (n=23), PQQ raised only the PGC-1α marker — and oral PQQ did not stimulate biogenesis in human cells in vitro. A marker moved; no new mitochondria were shown.
7. L-carnitine “optimizes mitochondrial fat-burning” in healthy adults Absent for healthy
The physiology is textbook (Established), but 4 g/day for 3 months in healthy men produced no change in muscle carnitine, mitochondrial enzymes, or performance. Benefits concentrate in deficient or diseased populations — not in healthy adults buying the promise.
8. CoQ10 improves energy and mitochondrial function in healthy adults Absent for healthy
Established biochemistry (it's a core component of the electron transport chain) and Established benefit in genetic CoQ10 deficiency. But in healthy, non-deficient adults, human trials show CoQ10 raises plasma levels without generally improving muscle CoQ10, VO₂max, or exercise economy.
9. “Mega-dosing” magnesium and B vitamins supercharges mitochondria Absent beyond sufficiency
Established as essential cofactors — ATP must bind Mg²⁺, and vitamins B1, B2, and B3 work inside the TCA cycle and electron transport chain. But there is no demonstrated mitochondrial benefit from super-dosing people who are already replete. Sufficiency matters; mega-dosing doesn't add more.
10. Urolithin A supports healthy aging via mitophagy Preliminary
Singh et al., Cell Reports Medicine 2022 (n=88, 4 months): ~10% greater leg-strength gains at 1,000 mg/day (≈11–12% was the 500 mg group), increased mitophagy-protein expression, and VO₂ peak improved within-group but not vs placebo (p=0.058). The caveats, stated plainly: the prespecified primary endpoint (peak power) missed statistical significance, the placebo group’s leg strength declined ~9.8% over the trial, and the product's manufacturer sponsored the trials. The only claim here with a genuine human mitophagy signal — preliminary beats overhyped.
11. Creatine supports mitochondrial energy metabolism Preliminary / mixed
Strong rationale via phosphocreatine buffering of cellular energy; small trials in mitochondrial cytopathy are mixed, with the pooled high-quality evidence (Cochrane, Kley 2013) showing no effect; and a four-ingredient cocktail of creatine, carnitine, leucine, and vitamin D3 improved lean mass and strength in healthy older adults in an 8-week randomized trial — but creatine cannot be isolated from the cocktail. Promising, not proven.
12. Alpha-lipoic acid is a proven mitochondrial antioxidant Preliminary / inconclusive
At 300 mg/day combined with acetyl-L-carnitine for 4 weeks in cyclists, there was no aerobic or anaerobic benefit. Inconclusive rather than disproven — the trial was small and short, and “proven” is doing far more work than the data allow.
What surprised us most
Most overhyped: the NAD+ category as a whole. Best-funded, best-marketed — and four direct human measurements of mitochondrial function all came back null, while a regulator ruled oral NAD+ has no human clinical evidence of even raising NAD+ levels. Blood-marker marketing is doing all the work.
Most underappreciated: urolithin A. The only claim with a genuine human mitophagy signal from muscle biopsies — with its caveats (missed primary endpoint, company sponsorship) stated plainly. Preliminary beats overhyped.
The meta-finding: across all 12 claims, the pattern holds — deficiency and disease states respond; healthy, replete adults almost never do. The entire category sells deficiency-correction biology at super-dose prices.
The numbers that matter
- “In four human muscle-biopsy trials of nicotinamide riboside, target engagement was confirmed via NAD+ metabolites (whole-blood NAD+ rose only in Elhassan 2019) — but not one found improvement in measured mitochondrial respiration, content, or bioenergetics.” (Dollerup 2020; Remie 2020; Elhassan 2019; Stocks 2021)
- “Oral NR raised whole-blood NAD+ up to 142% within two weeks at 1,000 mg/day (Conze et al. 2019), sustained at ~139% at eight weeks — yet in a 12-week biopsy trial at 2,000 mg/day, muscle NAD+ was unchanged.” (Blood ≠ tissue.)
- “Urolithin A at 1,000 mg/day was linked to ~10% greater leg-muscle strength gains than placebo (Singh et al., Cell Reports Medicine 2022, n=88; ≈11–12% was the 500 mg group), with VO₂ peak improved within-group but not vs placebo (p=0.058) — and the trial’s prespecified primary endpoint missed statistical significance, and the product’s manufacturer sponsored the trials.”
- “On October 13, 2025, the US National Advertising Division found there is no human clinical evidence that orally ingested NAD+ — ‘liposomal or otherwise’ — raises NAD+ levels in the body.”
- “A 2026 systematic review of 113 studies (Gallagher et al.) found zero eligible outcomes trials of IV or intramuscular NAD+ for anti-aging or wellness.”
- “A 2026 multi-cohort analysis (Nature Metabolism) found whole-blood NAD+ remarkably stable across age and lifestyle interventions — challenging blood NAD+ as a biomarker of aging.”
- “In a human trial, 300 mg of resveratrol did not activate the AMPK/SIRT1 ‘longevity axis’ in muscle or fat — and actually inhibited mitochondrial respiration in ex vivo muscle.”
- “The longest published NR trials run ~5–6 months (a 24-week long-COVID RCT with 20 weeks at 2,000 mg/day; a 5-month twin study up to 1,000 mg/day) — no safety data exist beyond ~6 months for NR or NMN, and multi-year long-term safety data are absent.”
Key studies cited
- Dollerup et al. 2020; Remie et al. 2020; Elhassan et al. 2019; Stocks et al. 2021 — human muscle-biopsy randomized trials of nicotinamide riboside: target engagement confirmed via NAD+ metabolites (whole-blood NAD+ rose >2-fold only in Elhassan 2019; the others measured muscle or urinary metabolites); mitochondrial respiration, content, and bioenergetics did not improve in any.
- Conze et al. 2019 — oral nicotinamide riboside raised whole-blood NAD+ up to 142% within two weeks at 1,000 mg/day, sustained at ~139% at eight weeks; a separate 12-week biopsy trial at 2,000 mg/day found muscle NAD+ unchanged.
- Singh et al., Cell Reports Medicine 2022 — urolithin A randomized trial (n=88, 4 months): ~10% greater leg-strength gains at 1,000 mg/day (≈11–12% was the 500 mg group), increased mitophagy-protein expression, VO₂ peak improved within-group but not vs placebo (p=0.058); prespecified primary endpoint (peak power) missed significance; manufacturer-sponsored.
- Gallagher et al. 2026 — systematic review of 113 studies: zero eligible outcomes trials of IV or intramuscular NAD+ for anti-aging or wellness.
- Trętowicz et al., Nature Metabolism 2026 — multi-cohort analysis: whole-blood NAD+ remarkably stable across age and lifestyle interventions, challenging blood NAD+ as a biomarker of aging.
- US National Advertising Division, October 13, 2025 — finding: no human clinical evidence that orally ingested NAD+, “liposomal or otherwise,” raises NAD+ levels in the body.
- Grant 2019 — IV NAD+ pharmacokinetics: infused NAD+ rapidly cleared and metabolized; no mitochondrial or clinical endpoints established.
Where the evidence stands
- Established: The underlying physiology — magnesium, B vitamins, CoQ10, and carnitine as essential cofactors and components of energy metabolism — is textbook. What is not established is any benefit from super-dosing healthy, replete people.
- Preliminary: Urolithin A has the only genuine human mitophagy signal (one 88-person randomized trial, with caveats). Creatine and alpha-lipoic acid have mixed or inconclusive human data — not disproven, not proven.
- Extrapolated: Much of the category’s marketing extrapolates deficiency-correction biology — what helps a deficient or diseased population — to healthy adults. That extrapolation is where the claims break down.
- Contested: Whether whole-blood NAD+ even declines with healthy aging — the core premise of repletion marketing — is now disputed by large-scale human data.
- Absent: Nine of the 12 claims graded Absent, including every mitochondrial-function claim for oral NAD+ precursors in healthy adults: four direct human measurements of mitochondrial function all came back null.