Qaytona is free while it's in beta — Get started

Sign in

Research ·

NMN, NR or NAD+ IV: the trials side by side

Qaytona Team · 8 min read

Green leaves with one dry, curled brown leaf among them, in soft autumn light.

NMN and NR, taken by mouth, both about double the NAD+ in blood at 1 g a day, and an IV drip of NAD+ raises it too. None of the three has shown a clinical benefit in a large trial. The largest, published in JAMA on October 8, 2026, gave 410 people with early Parkinson's disease 1 g of NR a day for a year, and they did slightly worse than those on placebo. The trials of injected NAD+ that measured a clinical outcome used 5 to 50 mg a day, a fraction of what US clinics give.

Side by side

NMNNRNAD+ by IV
Blood NAD+About doubles at 1 g a dayAbout doubles at 1 g a dayFlat for 2 hours of a slow drip, about 5 times baseline at 6 hours
Muscle NAD+Didn't riseDidn't riseNot measured
Largest trial108 older adults, 12 weeks: no effect on sleep or fatigue410 adults with Parkinson's, 1 year: slightly worse than placebo180 adults with heart failure, 10 mg a day for 7 days
Longest trial24 weeks52 weeks7 days of doses
Pooled trialsDiastolic pressure about 2 mmHg lower, low certainty. No change in glucose, lipids or muscleNo improvement in fatigue, sleep, endurance or strengthNone pooled. No trials of aging, energy or addiction
Sold asSupplement, again since September 2025SupplementCompounded injection, no approved product

The only trial to compare the two oral forms found them close: after two weeks at 1 g a day, NMN raised whole-blood NAD+ by 43 µM more than placebo and NR by 49 µM.

Higher blood levels haven't carried through to muscle. In each trial that measured it, muscle NAD+ stayed where it was.

NMN

Most NMN trials are small, last 4 to 12 weeks and are paid for by companies that make it. In the largest, 108 adults over 65 in Japan took 250 mg a day or placebo for 12 weeks, and NMN made no difference to sleep quality or fatigue.

Pooled across 15 trials, NMN lowered diastolic blood pressure by 2.4 mmHg, on low-certainty evidence, and didn't change glucose, HbA1c, lipids or weight.

In 2022 the Food and Drug Administration (FDA) said NMN couldn't be sold as a supplement, because it had already been authorized for testing as a drug. In September 2025 it reversed course, having found NMN on sale as a supplement since at least 2017. Each maker still has to notify the FDA, and the notifications accepted since cover lower doses than many trials used: Effepharm's covers 300 mg a day long term, or 600 mg for up to 60 days.

NR

NR has the larger and longer trials, and the newest one matters most. In NOPARK, 410 adults in Norway with Parkinson's diagnosed in the previous two years took 500 mg of NR twice a day or placebo for a year, on top of their usual treatment. On the standard rating scale they ended 2.7 points worse on NR than on placebo, and their non-motor symptoms 4.2 points worse. Serious adverse events were less common on NR, 8.3% against 14.1%.

Elsewhere NR has mostly done nothing beyond raising NAD+. A 2026 analysis of 27 trials found that no NAD+ booster improved fatigue or sleep, and NR improved none of the outcomes it pooled, including endurance and strength.

The trials give 1 to 3 g a day. ChromaDex, which sells NR as Niagen, told the FDA in 2018 that its product was for no more than 300 mg a day, and not for anyone under 18.

NAD+ by IV

The randomized trials with a clinical outcome are three, all in China and all supported by the maker of a Chinese NAD+ injection, at 5 to 50 mg a day for about a week. The largest gave 180 adults with ischemic heart failure 10 mg a day for seven days. A month later their ejection fraction was 45.4% against 42.4% on placebo; heart failure admissions over six months didn't differ significantly.

US clinics report giving 250 to 750 mg per drip, where the only randomized data are blood levels. In 8 healthy men given 750 mg over six hours, plasma NAD+ didn't change for two hours and was about five times baseline by the end. No one had side effects.

At a US wellness chain, six clients given 500 mg at a rate they set themselves, averaging about an hour and a half, all had cramping, nausea, vomiting, a racing heart and chest pressure until the drip ended. No study has compared infusion rates directly.

There is no FDA-approved NAD+ injection, so clinics use compounded product. In 2024 the FDA warned that some was being made from food-grade powder, after reports of chills, shaking and vomiting consistent with bacterial endotoxin. In January 2026 it cited a compounder whose NAD+ sent three patients from one vial to the emergency room.

Before recommending one

  • Parkinson's disease. The one large trial found NR left patients slightly worse after a year, and its authors concluded the results don't support it as a treatment.
  • What the patient expects. The consistent effect is more NAD+ in blood. Trials of fatigue and sleep have been negative, and none has tested injected NAD+ for aging, energy or addiction.
  • Training. In 11 untrained young men, 1.2 g of NMN a day for a week blocked the gain in muscle mitochondria that exercise produced on placebo.
  • Heart disease. NMN and NR raise 4PY, a breakdown product linked to heart attack and stroke in two cohorts. No trial has tested whether the supplements change that risk.
  • Pregnancy, breastfeeding and children. Neither oral form has been studied in them. Effepharm's NMN notification excludes all three, and ChromaDex's excludes anyone under 18.
  • The product. Of 70 NMN products that state an amount, the median directs 250 mg a day, the lowest dose the trials used. 36% direct more than Effepharm's 300 mg long-term amount, none more than 1 g, and 11% add trimethylglycine (TMG). Every label was entered before the FDA's reversal.

Our NAD+ monograph isn't live yet. The ingredient library lists the ones that are.

Sources

  1. 1Christen et al., 2026PubMed CentralOpen-label randomized trial, 67 adults aged 18 to 50, 14 days. Run by Nestlé Research.
  2. 2Elhassan et al., 2019PubMed Central12 men aged 70 to 80, 1 g of NR a day for 21 days. Blood NAD+ more than doubled; muscle NAD+ was 210 against 197 pmol/mg. NR supplied by ChromaDex.
  3. 3Kim et al., 2022PubMed Central108 community-dwelling adults aged 65 and over. NAD+ wasn't measured. Funded by Mitsubishi Corporation Life Sciences, which supplied the NMN.
  4. 4Yang W. et al., 2026PubMed CentralMeta-analysis of 15 randomized trials at 250 to 2,000 mg a day for 2 to 24 weeks. Adverse events were no more common than on placebo.
  5. 5FDA, 2025regulations.govResponse to the Natural Products Association's citizen petition, September 29, 2025. NMN remains a new dietary ingredient.
  6. 6Brakedal et al., 2026PubMedPhase 3 trial at 11 Norwegian centers, published October 8, 2026. The gap was smaller than the 5.2 points the trial counted as a clinically important worsening.
  7. 7Gao et al., 2026PubMed CentralNetwork meta-analysis of 27 studies and 1,599 people, 12 of the studies on NR and 7 on NMN.
  8. 8FDA, 2018regulations.govAcknowledgment of ChromaDex's new dietary ingredient notification for NR chloride, March 7, 2018. Its 2015 notification set 180 mg a day.
  9. 9Yu et al., 2026PubMed Central180 adults with an ejection fraction of 45% or less, at one center. Blood NAD+ wasn't measured. Grant and drug supply from Kaifeng Knature Pharmaceutical.
  10. 10Grant et al., 2019PubMed Central8 men on NAD+ and 3 on saline. Funded partly by NAD+ Research Inc., whose director runs a clinic that offers IV NAD+.
  11. 11Reyna et al., 2026PubMed CentralRetrospective review of 6 clients on NAD+ and 8 on IV NR at Restore Hyper Wellness, which funded it.
  12. 12FDA warning letter, 2026FDAGenoGenix, an outsourcing facility in Boca Raton, January 20, 2026. An unopened vial from the lot measured 3,360 endotoxin units per mL.
  13. 13Yang D. et al., 2026PubMed CentralDouble-blind crossover with muscle biopsies. Mitochondrial content rose 171% with exercise on placebo. Funded by Taiwan's science council.
  14. 14Ferrell et al., 2024PubMed CentralObservational cohorts of 2,331 people in the US and 832 in Europe. Adjusted hazard ratios for major cardiovascular events over three years were 1.89 and 1.99.
  15. 15NIH Dietary Supplement Label DatabaseNIH label databaseSearched October 9, 2026: 92 on-market labels with NMN, reduced to 72 products from 45 brands, newest label each. The newest was entered in August 2025.