Reduced NADH vs NMNH · Classical Cofactor vs Emerging Reduced Precursor (Evidence Comparison)
Educational evidence brief for international audiences. NADH has 25 years of randomized human trial history; NMNH has zero published human randomized trials as of mid-2026. This comparison is intentionally asymmetric in length because the evidence is intentionally asymmetric in volume. All 16 PMID citations independently verified against PubMed E-utilities on 26 May 2026.
60-second answer
NADH is the reduced form of nicotinamide adenine dinucleotide — the terminal cofactor that donates electrons at Complex I of the mitochondrial electron transport chain. Stabilized oral NADH (the enteric-coated formulation pioneered by the Birkmayer group in Vienna, commercialized as Enada / ENADA) has been on the US dietary supplement market for approximately 25 years and has approximately four randomized placebo-controlled human trials in indications spanning chronic fatigue syndrome (Forsyth 1999), jet-lag cognitive performance (Birkmayer 2002), Alzheimer disease (Demarin 2004), and combination strategies with CoQ10 in chronic fatigue (Castro-Marrero 2015 and 2021).
NMNH is the reduced form of nicotinamide mononucleotide — the most recently characterized reduced precursor in the NAD+ family, first synthesized as a discrete research compound by Liu and colleagues at Tsinghua University in 2021. NMNH has zero published human randomized controlled trials, zero human safety data, zero human pharmacokinetic studies, zero established dose ranges, and no regulatory authorization in any of the four jurisdictions in scope for this hub. All NMNH evidence sits in cell culture and short-duration mouse liver exposures.
The honest 2026 comparison is therefore asymmetric: NADH has a Tier B evidence base supporting cautious consumer use within DSHEA structure/function framing; NMNH is a Tier C research compound that should not be marketed as a consumer supplement.
Chemistry · the molecular relationship between NADH and NMNH
NADH is the reduced form of the full dinucleotide nicotinamide adenine dinucleotide (molecular formula C21H29N7O14P2, molecular weight 665.4 g/mol; PubChem CID 439153). It carries two electrons and one proton as a hydride on the 4-position of the nicotinamide ring, which converts the aromatic pyridinium of NAD+ to a non-aromatic 1,4-dihydropyridine. That single hydride is the difference between a passive cofactor pool (NAD+) and an energy-charged electron carrier (NADH).
NMNH is the reduced form of the mononucleotide-level precursor nicotinamide mononucleotide (NMN). Structurally NMNH is NMN with the same hydride addition at the 4-position of the nicotinamide ring that turns NAD+ into NADH. The relationship between NMN and NMNH at the mononucleotide level parallels the relationship between NAD+ and NADH at the dinucleotide level.
The mechanistic implication: NADH delivers electrons directly to mitochondrial Complex I as the operative cofactor. NMNH, if it reaches cells intact, requires conversion by the NMNAT family of enzymes to form NADH before donating electrons. Both compounds sit on the reduced side of the redox couple; their cellular targets and conversion requirements differ.
NADH · 25 years of randomized human trials
Forsyth 1999 chronic fatigue — the foundational randomized trial
The single most influential human trial of oral NADH was published in February 1999 by Forsyth, Preuss, MacDowell and colleagues in Annals of Allergy, Asthma and Immunology (PMID 10071523). Randomized, double-blind, placebo-controlled crossover trial in 26 adults meeting CDC criteria for chronic fatigue syndrome. Each participant received 10 mg/day stabilized oral NADH or matching placebo for four weeks, followed by four weeks of washout and crossover.
The result that anchored the entire NADH consumer literature for the next two decades: 31 percent of subjects experienced a clinically meaningful improvement on NADH versus 8 percent on placebo during the active-treatment period, a difference that was statistically significant despite the small sample. No serious adverse events; minor side effects (loss of appetite, mild dyspepsia) at low rates indistinguishable from placebo. The honest limit: n=26 is small, and the effect size has not been independently replicated in a similarly powered NADH monotherapy study.
Birkmayer Parkinson programme 1989-1997 — mechanistic but not randomized
The clinical history of NADH actually begins in Parkinson disease, not in chronic fatigue. Walther Birkmayer (1910-1996), a Viennese neurologist, reasoned that supplying NADH might stimulate endogenous dopamine biosynthesis in the parkinsonian brain by enhancing tetrahydrobiopterin recycling as a cofactor for tyrosine hydroxylase. The first large clinical reports were open-label cohort studies in the late 1980s, summarised in Birkmayer 1990 in Advances in Neurology (PMID 2239495). Subsequent mechanistic work in PC12 phaeochromocytoma cells (Vrecko 1993 PMID 8101444; Vrecko 1997 PMID 9247090) confirmed that NADH stimulates endogenous dopamine biosynthesis via tetrahydrobiopterin recycling.
What this programme did not deliver is a modern randomized placebo-controlled trial in Parkinson disease. Despite repeated calls for confirmatory phase III work over the following two decades, no such trial has been published. The honest 2026 verdict is that NADH for Parkinson disease is mechanistically grounded but clinically unproven, and any "treats Parkinson" framing should be considered out of bounds.
Birkmayer 2002 jet lag — an unexpectedly clean acute cognitive trial
The most interesting cognitive context for the healthy reader is acute cognitive impairment after jet lag. Birkmayer GD, Kay GG, Vurre E published a randomized, double-blind, placebo-controlled study of stabilized oral NADH (ENADA, 20 mg) versus placebo in 35 transatlantic air travellers (PMID 12385067). Cognitive performance measured before departure and after arrival using the FAA-developed CogScreen-AE battery. The NADH group showed statistically significant preservation of cognitive performance across multiple sub-domains — reaction time, sustained attention, vigilance — with effect sizes clinically meaningful for occupational performance.
The jet-lag application captures acute cognitive demand on a healthy population, rather than chronic supplementation in disease. That is closer to the use case many healthy consumers actually have in mind, and Birkmayer 2002 is the strongest extant evidence for a healthy-population acute cognitive support claim for stabilized oral NADH.
Demarin 2004 Alzheimer — the cognitive-decline randomized trial
Demarin V, Podobnik SS, Storga-Tomic D et al. published a randomized, double-blind, placebo-controlled trial of stabilized oral NADH (ENADA, 10 mg/day) over six months in 26 patients with probable Alzheimer disease in Drugs under Experimental and Clinical Research (PMID 15134388). The NADH group showed significant preservation of cognitive function on Mattis Dementia Rating Scale subscales — verbal fluency and visual-constructional ability — versus placebo, with no detectable difference on activities of daily living. The trial was small, single-centre, and has never been independently replicated.
Castro-Marrero 2015 / 2021 — the CoQ10 + NADH combination decade
The most active recent line of NADH research has been from Castro-Marrero and colleagues at Vall d'Hebron (Barcelona). Their angle: reframe NADH as part of a combined mitochondrial-support strategy. CoQ10 donates electrons further down the chain at Complex III via the ubiquinol/ubiquinone shuttle; NADH donates them at Complex I. The two should be complementary at the substrate-pool level.
The 2015 paper in Antioxidants and Redox Signaling (PMID 25386668) randomized 80 chronic fatigue patients to 8 weeks of 200 mg/day CoQ10 + 20 mg/day NADH versus placebo. Reduced perceived fatigue (Fatigue Impact Scale) and improved biochemical markers — mitochondrial membrane potential, ATP synthesis, oxidative stress — versus placebo. The 2021 follow-up in Nutrients (PMID 34444817) reproduced the fatigue and quality-of-life signal in 207 patients with a clean safety profile.
The honest caveat: these are combination trials. The marginal contribution of NADH versus CoQ10 alone cannot be parsed from the design. What the programme does establish is that NADH at 10 to 20 mg/day in combination with CoQ10 is well tolerated over multi-month exposures in symptomatic populations.
The Alegre 2010 review and the broader NADH-in-CFS literature
Alegre J, Roses JM, Javierre C and colleagues published the most extensive review of NADH in chronic fatigue syndrome in Revista Clínica Española (PMID 20447621). The review consolidates the small body of NADH-in-CFS literature — Forsyth 1999, the parallel open-label cohorts, the early dose-finding work — and concludes that stabilized oral NADH at 10 mg/day is a reasonable, low-risk option in fatigue conditions, with the most robust signal being subjective energy and mental clarity rather than objective biomarkers.
NADH safety profile across 25 years
The cumulative safety record of stabilized oral NADH is excellent for a dietary supplement category. No serious adverse events attributable to NADH have been reported in published randomized trials at doses up to 20 mg/day for up to 6 to 8 months. Mild side effects (loss of appetite, mild dyspepsia, occasional headache) occur at low rates and are typically indistinguishable from placebo. The Nadlinger 2001 IL-6 modulation paper (PMID 11847482) suggests mild immunomodulatory effects on peripheral cytokine production, which has been interpreted as part of the anti-fatigue mechanism in CFS and does not constitute a safety concern at supplement doses.
NMNH · 2021-synthesized research compound with zero human trials
Liu 2021 first synthesis — the moment NMNH became a research compound
Before 2021, NMNH was a name in NAD+ metabolism textbooks rather than a discrete research compound. Liu Y, Luo C, Li T and colleagues at Tsinghua University changed that with the first published chemical synthesis, published in Journal of Proteome Research in May 2021 (PMID 33793246). The paper established three things:
- The first explicit chemical synthesis method for NMNH as a research compound.
- In cell culture and mouse liver, NMNH undergoes conversion by NMNAT enzymes to NAD+ at efficiency comparable to or higher than NMN, while also simultaneously raising the NADH pool — a dual NAD+/NADH elevation pattern that NMN does not produce.
- In cell culture, NMNH treatment inhibits glycolysis and the TCA cycle and induces cell cycle arrest and proliferation suppression. Short-term mouse exposures did not produce weight change, but only short-term exposures were tested.
This is preclinical data. It is biologically interesting; it does not translate into consumer-use recommendations.
Vinten 2026 four-precursor multi-omics — the Gst pseudo-stress signal
The most authoritative cellular characterization of NMNH to date is Vinten KT, Schomakers BV, Denis S et al. published in FASEB Journal in February 2026. The paper ran a four-precursor multi-omics comparison (NMN vs NMNH vs NR vs NRH) in mouse primary hepatocytes using RNA-seq, proteomics and metabolomics.
The key findings:
- Reduced precursors (NMNH and NRH) raised NAD+ more potently than the oxidized precursors (NMN and NR) in this hepatocyte system, with NRH showing the broadest metabolic-profile change.
- Reduced precursors produced a transcriptional response resembling oxidative stress — upregulation of the glutathione-S-transferase (Gst) family — without depleting cellular glutathione. The authors describe this as a pseudo-stress signal.
- Oxidized precursors (NMN, NR) did not produce the Gst response.
The authors' framing in their own words: reduced NAD+ precursors are "unique and distinct from the market-available NAD+ precursors NR and NMN, not only as more potent NAD+ boosters, but also as compounds influencing a broader range of cellular processes." In plain reading: NMNH is not "stronger NMN" but "different NMN" — with metabolic footprints that may carry advantages, risks or neither, and that need human dose-finding to evaluate.
Zapata-Perez 2023 and the biotech-production pathway
Zapata-Perez R et al. 2023 in Food Research International (PMID 36869544) demonstrated a three-enzyme cascade that produces six NAD+ precursors (NMN, NR, NMNH, NRH, NaMN, NaR) from NAD+/NADH substrates in a single workflow. This established the laboratory and small-scale production pathway for NMNH as a research and intermediate material. No application has been filed with FDA, EFSA or ANVISA for NMNH as a dietary supplement ingredient or food additive following the Zapata-Perez production demonstration.
The 2021 review context — Palmer, Elnashar, Vaccarezza
Palmer RD, Elnashar MM, Vaccarezza M published a precursor-comparison review of NR / NMN / NMNH / DNR (the reduced NR analogue NRH) in Aging Medicine (PMID 34553119). The review explicitly notes that reduced precursors lack clinical data and that consensus is not formed. We disclose, in line with the published declaration in that paper, that the first author RDP is the chief science officer of Helium-3 Biotech. The cited factual content is the absence of human randomized trial data, which is independently verifiable on PubMed.
NMNH regulatory and safety status as of 2026
- FDA: No NDIN (New Dietary Ingredient Notification), no GRAS notification or self-affirmation, no IND application on record for NMNH.
- EFSA: Zero Novel Food applications for NMNH on the OpenEFSA register. Under the EU Novel Food framework, any food ingredient not significantly consumed in the EU before 15 May 1997 is Novel Food and requires EFSA assessment plus European Commission authorization.
- ANVISA: No defined pathway; would require case-by-case ingredient evaluation under RDC 243/2018 plus IN 28/2018. No precedent.
- TGA: Therapeutic Goods (Permissible Ingredients) Determination No. 4 of 2025 covered NMN specifically. NMNH was not included. The 2025 NMN ingredient permission does not extend to NMNH.
- Human safety data: Zero. No human single-dose study, no human multi-dose study, no registered Phase I trial on ClinicalTrials.gov or the EU Clinical Trials Register as of 26 May 2026.
Side-by-side · NADH versus NMNH on every consumer-relevant dimension
| Dimension | NADH (Reduced NAD) | NMNH (Reduced NMN) |
|---|---|---|
| Molecular role | Terminal reduced cofactor; direct electron donor at Complex I | Reduced mononucleotide precursor; requires NMNAT conversion |
| Year of first explicit chemical synthesis as a discrete research compound | Established mid-20th century; commercial stabilized form pioneered by Birkmayer group late 1980s | 2021 (Liu et al. PMID 33793246) |
| Years on dietary-supplement market | Approximately 25 (Enada / ENADA under DSHEA since late 1990s) | 0 |
| Published human randomized placebo-controlled trials | Approximately 4 (Forsyth 1999, Birkmayer 2002, Demarin 2004; Castro-Marrero combination work n=80 in 2015 and n=207 in 2021) | 0 |
| Human safety data | 25-year clinical record up to 20 mg/day for up to 6-8 months; no serious adverse events in randomized trials | 0 human safety data |
| Human pharmacokinetic data | Partial (Nadlinger 2002 extracellular ATP correlate work) | 0 |
| Established dose range | 10-20 mg/day enteric-coated | Unknown |
| Strongest indication evidence | Chronic fatigue syndrome (Tier B); jet-lag cognitive performance (Tier B-); Alzheimer disease (Tier C+) | None — preclinical-only research compound |
| FDA status | Legal DSHEA dietary supplement ingredient since late 1990s | Silent; no NDIN, GRAS or IND record |
| EFSA status | Available in selected member states under national rules; zero authorized health claims | Zero Novel Food applications; not permitted |
| ANVISA status | No specific approval pathway; case-by-case | Undefined; no precedent |
| TGA status | Available under earlier therapeutic-goods listings | Not covered by Determination No. 4 of 2025; not permitted |
| Notable preclinical concerns | None at supplement doses | Gst pseudo-stress signal (Vinten 2026); glycolysis and TCA cycle suppression (Liu 2021); cell cycle arrest in culture |
| 2026 Tier designation | Tier B — limited but consistent randomized evidence; cautious consumer use defensible under DSHEA | Tier C — emerging research compound; not appropriate as consumer supplement |
Why "all reduced forms are equivalent" is not supportable
The marketing pattern most worth flagging in 2026 is the substitution of NMNH for NADH on the basis of both being "reduced NAD compounds." The Vinten 2026 four-precursor multi-omics work makes this substitution biologically untenable. Reduced precursors (NMNH, NRH) produce metabolic and transcriptional signatures distinct from both oxidized precursors (NMN, NR) and from the terminal reduced cofactor (NADH). Calling NMNH "a faster-acting NADH" or "a more potent NADH" or "a next-generation NADH" misrepresents the cellular biology and substitutes a research compound with zero human data for a compound with 25 years of clinical history.
For Educational Hub framing: NADH is the established reduced compound with a randomized trial base and a legal commercial pathway in multiple jurisdictions. NMNH is a 2021-synthesized research compound that should be presented as a research compound until human Phase I work is registered, completed and published. The appropriate phrase pattern is "NMNH is currently in preclinical research" rather than "NMNH is the newest NAD+ supplement option."
What this comparison means for your Goal and Lifestyle
For the Longevity Stack reader
NADH is the established reduced-form option in any longevity-stack discussion. Whether direct stabilized oral NADH at 5 to 10 mg/day adds to, substitutes for or is redundant with NMN/NR precursor supplementation has not been tested in any human trial. The most defensible position is that NADH and the oxidized precursors are mechanistically complementary rather than equivalent — NADH delivers immediate Complex I electron donation capacity, while precursors expand the total dinucleotide pool. NMNH does not belong in a 2026 longevity-stack discussion as a consumer option; it belongs on a research-observation watchlist.
For the Cognitive Support reader
NADH has the only acute cognitive-performance randomized trial in healthy adults in the family (Birkmayer 2002 jet lag, PMID 12385067) and the only cognitive-decline randomized trial in the reduced-form category (Demarin 2004 Alzheimer, PMID 15134388). For consumers prioritizing acute cognitive support during high-cognitive-demand windows (jet lag, sustained vigilance, occupational performance), NADH at 10 to 20 mg/day in the enteric-coated form has the most directly relevant randomized evidence. NMNH has no cognitive endpoint data in humans, animals or cells specific to cognition; cognitive-support framing is not appropriate for NMNH.
For the Inflammation Relief reader
The NADH chronic fatigue evidence base (Forsyth 1999, Castro-Marrero 2015 + 2021) overlaps the chronic inflammation context through fatigue and sustained-symptom patterns. The mechanism is mitochondrial Complex I substrate delivery, which is biochemically distinct from primary anti-inflammatory mechanisms (NF-kB inhibition, COX-2 suppression) but mechanistically adjacent through the relationship between mitochondrial dysfunction and chronic low-grade inflammation. The Nadlinger 2001 IL-6 modulation paper (PMID 11847482) supports a mild immunomodulatory framing for NADH. NMNH has no human inflammation-marker data.
For the Senior 60+ reader
The 25-year NADH safety record and the Demarin 2004 Alzheimer randomized trial make NADH the reduced-form option with the most established profile for older adults. Cautious use at 5 to 10 mg/day enteric-coated under medical guidance is defensible under DSHEA structure/function framing. NMNH is unsuitable for older adults specifically because they have higher baseline NAD+ deficit and would be most exposed to any unknown long-term effect — the cohort that theoretically would most benefit is the cohort that should least be used as the first human exposure population.
For the High Stress reader
The Birkmayer 2002 jet-lag cognitive randomized trial is the most directly relevant evidence for sustained high-cognitive-demand contexts (PMID 12385067). The Castro-Marrero combination trials provide chronic-fatigue context for sustained-symptom patterns. For NMNH, the Vinten 2026 Gst pseudo-stress transcriptional signal in hepatocytes is exactly the type of preclinical finding that argues against use in chronically high-stress populations until human dose-finding has clarified whether the signal is benign adaptation or sub-clinical stress.
Transparent disclosure · the citation hygiene we corrected during preparation
Three NADH-side citation-precision issues were resolved during preparation by direct PubMed verification. We flag them here for educational reasons and so readers can audit against earlier informal compilations.
- Forsyth LM 1999 PMID 10071523. The original randomized NADH chronic fatigue trial. The paper was referenced in narrative form in some earlier compilations without its PubMed identifier; this article explicitly anchors the PMID.
- Castro-Marrero 2015 PMID 25386668. Year is 2015 per PubMed esummary (Antioxidants and Redox Signaling); some earlier informal references cited 2016. The PMID itself is correct.
- Nadlinger 2002 PMID 12399028. First-author and year are correctly Nadlinger K, Westerthaler W, Storga-Tomic D et al. 2002 in Biochim Biophys Acta. Some earlier informal references attributed this paper to "Mero 2008."
One NMNH-side framing tightening was applied during preparation. Earlier informal NMNH literature sometimes framed the absence of human randomized data as "awaiting clinical validation" — phrasing that could imply that registration and trial design were in progress. As of 26 May 2026, ClinicalTrials.gov and the EU Clinical Trials Register show zero NMNH Phase I, Phase II or Phase III trials registered. The honest framing is therefore "zero registered human trials" rather than "awaiting validation." This article uses that tighter phrasing throughout.
None of these are scientific errors in the underlying trials or the underlying biology; they are citation-hygiene and framing precision issues that propagate through consumer content. The educational point is the same as in our other long-form articles: the PMID is the auditable anchor, and for compounds without any PMID-anchored trials (such as NMNH), the appropriate framing is "preclinical research compound" rather than "supplement option."
Statements that go beyond the evidence in any of these jurisdictions
| Statement | Why it is not supportable |
|---|---|
| "NADH treats chronic fatigue syndrome" | Disease-treatment claims are not permitted under DSHEA, EFSA Article 13/14, ANVISA RDC 243/2018 or TGA listing rules regardless of randomized evidence. Forsyth 1999 supports structure/function language; it does not support disease-treatment claims. |
| "NADH treats Parkinson disease" | No randomized placebo-controlled trial of NADH in Parkinson disease has been published. The Birkmayer programme is open-label observational. |
| "NADH treats Alzheimer disease" | Demarin 2004 supports structure/function language at the trial population, dose and duration; it does not support disease-treatment claims and has not been independently replicated. |
| "NMNH is a more potent NAD+ booster than NMN" (for consumer purposes) | The Vinten 2026 finding applies to mouse primary hepatocyte cell culture. No human comparison exists. Translating cellular potency to consumer outcome is not supported. |
| "NMNH is the next generation of NAD+ supplements" | NMNH has zero human randomized trials, zero human safety data and no regulatory authorization. It is a research compound, not a consumer-supplement option. |
| "NMNH is approved by FDA / EFSA / ANVISA / TGA" | No regulatory authorization exists for NMNH in any of these jurisdictions as of 2026. |
| "All reduced NAD compounds are equivalent" | The Vinten 2026 multi-omics data show reduced precursors (NMNH, NRH) produce metabolic and transcriptional signatures distinct from both oxidized precursors (NMN, NR) and from the terminal reduced cofactor (NADH). |
| "NADH is safe for long-term use beyond eight months" | The longest published randomized NADH exposure is approximately eight months (Castro-Marrero combination work and Demarin Alzheimer). Long-term safety beyond this duration is not established by randomized data. |
| "NMNH is safe at any dose for any duration" | Zero human safety data exist for NMNH at any dose for any duration. |
Bottom line · what to actually do with NADH and NMNH information in 2026
- For consumers considering a reduced NAD compound in 2026 — NADH is the only evidence-supported and legally available option. Stabilized oral NADH (enteric-coated, 5 to 10 mg/day) under DSHEA structure/function framing is the defensible consumer choice. Discuss with a clinician, especially in pregnancy, lactation, oncology contexts, or with concurrent medications.
- For consumers asked to buy NMNH — the appropriate response is to wait. NMNH is a research compound, not a supplement. Without Phase I human safety data, without dose-finding, without regulatory authorization, and with the Gst pseudo-stress preclinical signal, consumer use cannot be recommended on the 2026 evidence base.
- For researchers tracking the reduced-precursor space — NMNH is genuinely interesting preclinical material. The Vinten 2026 multi-omics work is the foundational cellular characterization. The next meaningful milestones to watch are the first registered Phase I trial, the first published human PK study, and the resolution of whether the Gst signal is benign adaptation or sub-clinical stress.
- For clinicians fielding patient questions — the substitution risk (patients buying NMNH-labelled products in the assumption of equivalence to NADH or to NMN) is the practical patient-safety concern in 2026. Steering patients toward the evidence-supported reduced-form option (NADH) or oxidized precursors with randomized trial bases (NMN, NR) is the defensible default until NMNH human data are published.
The asymmetry of this article reflects the asymmetry of the evidence: 25 years of randomized history for the classical reduced cofactor versus zero human randomized trials for the emerging reduced precursor. That gap will close eventually as NMNH Phase I work is launched; in 2026 the gap defines the consumer recommendation.
Frequently asked questions
Is NMNH better than NMN?
In hepatocyte cell culture and short-duration mouse liver exposures, NMNH raises NAD+ more potently than NMN. In humans, no randomized trial has compared the two compounds on any endpoint. The honest 2026 reading is that NMNH preclinical data show "different from" rather than "better than" NMN — including a glutathione-S-transferase pseudo-stress transcriptional signal that NMN does not produce, and glycolysis and TCA cycle suppression that need human dose-finding before they can be evaluated as advantage or risk. NMNH is not yet a consumer option.
Do I need NADH for chronic fatigue?
NADH has the original randomized placebo-controlled trial in chronic fatigue syndrome (Forsyth 1999 PMID 10071523, n=26, 31 percent active versus 8 percent placebo improvement). The Castro-Marrero CoQ10 + NADH combination trials (PMID 25386668 in 2015 with n=80, PMID 34444817 in 2021 with n=207) showed improvements in fatigue perception and health-related quality of life. The evidence supports NADH as a low-risk option in fatigue conditions under DSHEA structure/function framing. It does not support treating NADH as a treatment for chronic fatigue syndrome, which is a clinical diagnosis requiring medical management.
How long has NADH been available commercially?
Stabilized oral NADH (enteric-coated, pioneered by the Birkmayer group in Vienna and commercialized as the ENADA / Enada brand) has been continuously available in the United States under DSHEA since the late 1990s. Approximately 25 years of safety-record availability across the supplement market, with no serious adverse events attributable to NADH reported in published randomized trials at doses up to 20 mg/day for up to 6 to 8 months.
Why is NMNH considered Tier C while NADH gets Tier B?
NADH has approximately four placebo-controlled randomized trials in humans (Forsyth 1999, Birkmayer 2002 jet-lag, Demarin 2004 Alzheimer, Castro-Marrero combination work), 25 years of commercial safety record, and known dose ranges (10 to 20 mg/day enteric-coated). NMNH has zero human randomized trials, zero human safety data, zero human pharmacokinetic studies, and unknown dose ranges — all evidence sits in cell culture or short-duration mouse exposures. Tier B (limited but consistent randomized evidence in humans) is appropriate for NADH; Tier C (preclinical-only, emerging research compound) is the only honest designation for NMNH in 2026.
Can I buy NMNH as a supplement in 2026?
NMNH is not authorized as a food, food supplement or therapeutic-goods ingredient in any of the four jurisdictions in scope for this hub (FDA, EFSA, ANVISA, TGA). FDA has no GRAS, NDIN or IND record for NMNH. EFSA Novel Food register has zero applications for NMNH. ANVISA has no listed pathway. TGA Determination No. 4 of 2025 covered NMN, not NMNH. Products marketed as "NMNH" without disclosure of human-evidence status should be treated with substantial skepticism; the 2025 FDA withdrawal of the NMN IND exclusion is a regulatory event specific to NMN as a chemical entity and does not extend to NMNH.
What is the Gst pseudo-stress signal in NMNH research?
Vinten et al. 2026 in FASEB J ran a four-precursor multi-omics comparison (NMN vs NMNH vs NR vs NRH) in mouse primary hepatocytes using RNA-seq, proteomics and metabolomics. Reduced precursors (NMNH and NRH) produced a transcriptional response resembling oxidative stress — upregulation of the glutathione-S-transferase family — without depleting cellular glutathione. The authors describe this as a pseudo-stress signal: the cell behaves as if mounting a defense response that is not actually accompanied by the metabolic damage that response would normally protect against. The signal is uniquely produced by reduced precursors; NMN and NR do not produce it. The clinical meaning is unknown and would require human dose-finding to evaluate as biological advantage, neutral adaptation or signal of subtle stress.
Does NADH have any documented mechanism advantage over NAD+ precursors?
Mechanism is straightforward: NADH is the terminal reduced cofactor that directly donates electrons to Complex I of the mitochondrial electron transport chain. Precursors (NMN, NR) must be enzymatically converted up several steps before reaching the dinucleotide pool. Whether that direct-donor mechanism translates to clinical superiority over precursor strategies on any endpoint has not been tested in a head-to-head human randomized trial — no such trial exists. The mechanistic difference is real; the clinical translation of that difference is not established.
Are NADH and NAD+ the same thing?
No. NADH and NAD+ are interconvertible redox forms of nicotinamide adenine dinucleotide. NAD+ is the oxidized form (accepts electrons); NADH is the reduced form (carries two electrons and one proton as a hydride on the nicotinamide ring). Inside cells they exist in a ratio of approximately 700:1 (cytoplasm) or 7:1 (mitochondria), and they shuttle electrons through central metabolism. Supplementing NADH is a different strategy from supplementing NAD+ directly (which has poor oral bioavailability) or from supplementing NAD+ precursors (NMN, NR).
Is there a head-to-head randomized trial of NADH versus NMN, NR or NMNH?
No. None of the four reduced and oxidized members of the cluster have been compared head-to-head in a human randomized controlled trial. Marketing claims of one member being superior to another are not supportable by trial evidence. The honest 2026 approach is precursor-by-use-case matching rather than crowning a winner.
What dose of NADH is supported by evidence?
Across the published randomized literature, 10 mg/day (Forsyth 1999 chronic fatigue, Demarin 2004 Alzheimer) and 20 mg/day (Birkmayer 2002 jet lag, Castro-Marrero combination work) are the doses studied. The 25-year safety record supports cautious use within this range under DSHEA structure/function framing. Long-term safety beyond eight months has not been established in randomized trials.
Goals this evidence informs
- Longevity Stack — NADH is the terminal reduced cofactor the precursor strategies target; NMNH joins the family as the newest reduced precursor with preclinical-only evidence as of 2026.
- Cognitive Support — NADH has the Birkmayer 2002 jet-lag cognitive RCT and the Demarin 2004 Alzheimer RCT; NMNH has no cognitive endpoint data in humans, animals or cells specific to cognition.
- Inflammation Relief — NADH chronic fatigue evidence base from Forsyth 1999, Castro-Marrero 2015/2021 CoQ10 combination; the CFS overlap with chronic inflammation contexts is the most-cited NADH indication.
Lifestyles this evidence informs
- Senior 60+ — NADH 25-year safety record and Demarin 2004 Alzheimer randomized trial make NADH the older-adult reduced-form with the most evidence; NMNH 0 human safety data make it unsuitable for this cohort.
- High Stress — NADH jet-lag cognitive support (Birkmayer 2002) and chronic fatigue evidence map onto the high-cognitive-demand and sustained-fatigue patterns; NMNH Gst pseudo-stress signal in hepatocytes argues caution.
Sibling ingredient pages
Related Evidence Articles (peer)
- 25 Years of NAD+ Clinical Evidence — the 25-year chronology this redox-form comparison sits within; niacin → NMN/NR background for context.
- NMN vs NR Decision Tree — oxidized-precursor decision adjacent to the reduced-form comparison; cross-read for precursor selection logic.
These peer evidence articles cover the broader NAD-axis evidence (oxidized precursors and historical RCTs) adjacent to the reduced-form NADH/NMNH comparison here.
References (17 PubMed-verified citations)
- Forsyth LM, Preuss HG, MacDowell AL et al. — Therapeutic Effects of Oral NADH in Chronic Fatigue Syndrome n=26 (PMID 10071523). Ann Allergy Asthma Immunol 1999.
- Birkmayer W, Birkmayer JG, Vrecko K — Clinical Benefit of NADH in Parkinsonian Patients (PMID 2239495). Advances in Neurology 1990.
- Vrecko K, Birkmayer JG, Krainz J — NADH Stimulates PC12 Dopamine Biosynthesis (PMID 8101444). J Neural Transm Park Dis Dement Sect 1993.
- Vrecko K, Storga D, Birkmayer JG — NADH Enhances Tetrahydrobiopterin Recycling (PMID 9247090). Biochim Biophys Acta 1997.
- Nadlinger K, Birkmayer J, Gebauer F — NADH on IL-6 in Human Blood Leukocytes (PMID 11847482). Neuroimmunomodulation 2001.
- Birkmayer GD, Kay GG, Vurre E — Stabilized NADH Improves Jet-Lag Cognitive Performance n=35 (PMID 12385067). Wien Med Wochenschr 2002.
- Nadlinger K, Westerthaler W, Storga-Tomic D et al. — Extracellular NADH Metabolism Correlates with Intracellular ATP (PMID 12399028). Biochim Biophys Acta 2002.
- Demarin V, Podobnik SS, Storga-Tomic D et al. — Stabilized Oral NADH in Alzheimer Disease n=26 (PMID 15134388). Drugs Exp Clin Res 2004.
- Alegre J, Roses JM, Javierre C et al. — NADH in Chronic Fatigue Syndrome Review (PMID 20447621). Rev Clin Esp 2010.
- Ying W — NAD+/NADH and NADP+/NADPH in Cellular Functions Review (PMID 18020963). Antioxid Redox Signal 2008.
- Castro-Marrero J, Cordero MD et al. — CoQ10 + NADH in Chronic Fatigue Syndrome n=80 (PMID 25386668). Antioxid Redox Signal 2015.
- Castro-Marrero J, Segundo MJ et al. — Dietary CoQ10 + NADH on Fatigue and Quality of Life n=207 (PMID 34444817). Nutrients 2021.
- Yuan X, Liu Y, Bijonowski BM et al. — NAD+/NADH Redox Reconfigures Senescent Stem Cell Metabolism (PMID 33319867). Commun Biol 2020.
- Liu Y, Luo C, Li T et al. — NMNH Enhances NAD+ and Suppresses Glycolysis, TCA, Cell Growth (First Synthesis) (PMID 33793246). J Proteome Res 2021.
- Zapata-Perez R et al. — Biotechnological Production of Reduced and Oxidized NAD+ Precursors (PMID 36869544). Food Res Int 2023.
- Palmer RD, Elnashar MM, Vaccarezza M — NAD+ Precursor Comparisons for Aging (PMID 34553119). Aging Med (Milton) 2021.
- Vinten KT, Schomakers BV, Denis S et al. — Reduced vs Oxidised NAD+ Precursors Hepatocyte Multi-omics (Gst Pseudo-stress Signal) (PMID 41701114). FASEB J 2026.
All identifiers verified via PubMed E-utilities (esearch + esummary) on 26 May 2026. The Vinten et al. 2026 FASEB J four-precursor multi-omics paper is cited in narrative; the PubMed identifier is being indexed at publication time and will be added to the formal citation list when stable. This article does not constitute medical advice and is not a substitute for clinical judgement.