NAD+, NMN and NR: what the human trials actually show

NAD+ is not a peptide. It is a coenzyme every cell needs to turn food into usable energy, and it sits on peptide menus because clinics sell it by injection and drip. The human trials that exist are mostly of two oral precursors, NR and NMN, and they land on a clear finding: these supplements raise NAD+ in blood, and in one 10-week trial NMN improved muscle insulin sensitivity in women with prediabetes by 25%. Beyond that the health effects are modest and mixed, and the injected route that clinics sell carries the least trial support of the three ways to raise NAD+.

What it is

Nicotinamide adenine dinucleotide shuttles electrons in the reactions that make ATP, and it is used up by enzymes involved in DNA repair and by the sirtuins, a family of proteins linked to ageing in animal studies. In mice, tissue NAD+ falls with age, and one enzyme, CD38, appears to drive much of that decline; mice lacking CD38 did not show it [11]. The idea that topping NAD+ back up will slow human ageing comes from this animal work, and it is the hypothesis the human trials below are built to test.

You can try to raise NAD+ three ways. You can take a precursor by mouth: nicotinamide riboside (NR), nicotinamide mononucleotide (NMN), or older forms of vitamin B3 such as niacin. You can infuse or inject NAD+ itself. Or you can do neither and rely on diet, exercise and sleep. Only the first route has meaningful trial data.

Even the oral route is less direct than the marketing suggests. In mice, only a small part of oral NMN or NR was absorbed intact from the small intestine; most was converted by gut bacteria into nicotinic acid and reached NAD+ by a roundabout route through the liver [12]. Whether humans handle oral precursors the same way is the question the trials below have not settled.

The human evidence: oral precursors

NR raises NAD+ by about 60%

The trial most often quoted gave 1,000 mg a day of NR or placebo, six weeks each in random order, to healthy adults aged 55 to 79 [1]. Thirty were randomised and 24 finished. NAD+ in white blood cells rose by roughly 60% compared with placebo, and NR was well tolerated with no meaningful changes in blood safety tests [1]. The authors also saw hints of lower blood pressure and arterial stiffness in people who started with higher blood pressure. Those were secondary, exploratory findings, and the paper says so.

NMN and muscle insulin sensitivity

The most rigorous positive result comes from Washington University in St Louis. Twenty-five postmenopausal women with prediabetes and overweight or obesity took 250 mg a day of NMN (13 women) or placebo (12) for 10 weeks [2]. Muscle insulin sensitivity, measured with the hyperinsulinaemic-euglycaemic clamp (the reference method), rose by 25 ± 7% on NMN and did not change on placebo [2]. Body composition, blood pressure, blood lipids, fasting glucose and liver insulin sensitivity did not change in either group. So this is one real, well-measured effect in one tissue, in 13 treated women.

NMN dose-response

A 60-day trial in 80 healthy adults (mean age 49) compared placebo with 300, 600 or 900 mg of NMN a day [3]. Blood NAD+ rose in every NMN group, highest at 600 and 900 mg. Six-minute walk distance improved more in all NMN groups than on placebo, and a questionnaire-based health score improved; insulin resistance (HOMA-IR) did not change [3]. No safety problems were seen. The trial was run with an NMN manufacturer (the lead author’s listed address is a company email), and a walking-test gain in healthy middle-aged people over two months is the kind of result that needs independent repetition.

Where it didn’t work

The null results matter as much. In 13 overweight or obese adults, six weeks of 1,000 mg/day NR raised muscle NAD+ markers but did not improve insulin sensitivity, mitochondrial function, liver fat, blood pressure or inflammation [4]. In 12 older men, three weeks of NR reached muscle but did not change mitochondrial energy output, though it lowered some circulating inflammatory markers [5].

A clinical population

The largest outcome trial so far enrolled 90 people with peripheral artery disease, who walk poorly because of narrowed leg arteries [6]. After six months, NR improved six-minute walk distance by 17.6 metres more than placebo (+7.0 vs −10.6 m), a meaningful but modest change measured with a one-sided 90% confidence interval; adding resveratrol did not help [6]. The authors call for a larger trial to confirm it.

The trials side by side

The pattern is consistent: the biochemistry moves every time, the health measures move sometimes.

Pooled evidence

A 2025 meta-analysis of 12 NMN trials (513 participants) found NMN reliably raised blood NAD+ but did not significantly change fasting glucose or blood lipids, and rated the risk of bias as having “some concerns” or “high” in every study [7]. A 2026 systematic review of 33 human intervention studies reached the same place: good evidence that NR and NMN engage the target, heterogeneous and often null effects on anything a person would notice [8].

Injected and IV NAD+

This is the form most often sold alongside peptides, and it has the least evidence. The 2026 systematic review found no eligible outcome trials of intravenous or intramuscular NAD+ for anti-ageing or wellness [8].

What does exist is small. In a 2019 pilot, eight men received 750 mg of NAD+ over six hours and three received saline [9]. Surprisingly, blood NAD+ did not rise at all for the first two hours; the body cleared it as fast as it arrived. Levels peaked at about four times baseline only at six hours, and much of the NAD+ was broken down into nicotinamide-related products or excreted in urine [9].

A 2026 retrospective review from a commercial IV clinic compared 14 clients given four daily 500 mg infusions: six received NAD+ and eight received NR [10]. People on NAD+ drips reported moderate to severe stomach symptoms, a faster heart rate and chest pressure during infusions, which stretched infusion times to about 97 minutes versus 37 for NR. Symptoms stopped when infusions ended [10]. The authors were employed by the clinic chain, and the study had no placebo and no benefit endpoint worth the name.

What the animal work suggests

In mice, raising NAD+ with precursors has improved muscle, metabolic and inflammatory measures across many models, though effects vary by model and lifespan gains have been inconsistent [8]. The gap between those results and the human trials above is the main story of this field.

Safety and side effects reported

In the trials here, oral NR at 1,000 mg/day for six weeks and NMN at up to 900 mg/day for 60 days were well tolerated, with no safety signals in blood tests [1,3]. In the NR crossover, two participants in one group dropped out citing side effects [1]. Trials have been short and small, so rare or long-term effects would not show up. IV NAD+ produced frequent, unpleasant infusion reactions in the one comparative report [10].

Legal and regulatory status

No form of NAD+ or its precursors is approved as a medicine for ageing in the UK, US or EU. In the US, the FDA said in November 2022 that NMN could not be sold as a dietary supplement because it had been authorised for drug investigation; in September 2025 it reversed that position after evidence that NMN was on sale as a supplement from 2017 [13]. In Great Britain, NR chloride holds a novel food authorisation for supplements, with a maximum of 300 mg a day for adults (230 mg in pregnancy and breastfeeding) [14]. We could find no equivalent GB authorisation for NMN. Injected NAD+ is a medicinal product rather than a food, and its use in wellness clinics sits outside any licensed indication. NAD+, NMN and NR are not on the WADA Prohibited List.

What is still open

• Whether raising blood NAD+ changes anything that matters over years: disease, disability, lifespan.
• Why NMN improved muscle insulin sensitivity in one trial while NR did not in another. Dose, compound, population and tissue all differed.
• What IV or injected NAD+ does beyond causing a brief, well-cleared spike, and whether that is worth the side effects.
• How oral precursors are actually absorbed in humans, given the mouse evidence that gut bacteria convert most of them first.
• Long-term safety beyond six months.

What this means

We think the oral precursors earn a modest place on the shelf and the injected route does not, and both halves of that position come from the same body of trial evidence.

On oral NR and NMN, our position is that this is a well-tolerated supplement with a reproducible effect on a blood marker and one small, genuine effect on health in a specific population. The biochemistry moves in every trial that measured it. Health measures move sometimes, and the clearest move — 25% better muscle insulin sensitivity in 13 women with prediabetes — is worth taking seriously rather than filing away as noise. For a healthy middle-aged adult taking NR or NMN hoping to feel different, these trials offer little to point at. For someone who is insulin-resistant, the NMN result is the most concrete thing in the field, and it is the result we would build the next trial around.

On IV and injected NAD+, our position is simpler: the evidence does not support clinic prices for a six-hour drip. The one comparative report shows frequent unpleasant infusion reactions and no benefit endpoint, and the systematic reviews between them find no run of outcome data to build on. A brief spike in blood NAD+ that the body clears as fast as it arrives is not a health outcome, and we would rather see that money spent on the interventions the trials do support.

We also think the field's honesty problem deserves naming. Several of the strongest-looking trials were run with manufacturer involvement or by clinic-affiliated authors, and the pooled analyses rate the studies as carrying some or high risk of bias. That does not make the results wrong; it makes independent replication the thing to watch. So HPR's stated position is this: treat NR and NMN as a reasonable, low-risk experiment in the right population, and treat injected NAD+ as a premium product still waiting on outcome trials rather than pilot infusions to justify itself.

What readers report

First-hand accounts, shared with permission. They are self-reported, not evidence, and sit alongside the research above rather than in place of it.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.