NAD+ side effects by evidence tier: the 2019 infusion study, rate-dependent flushing, chest tightness and nausea in clinic infusions, oral NR and NMN trials, and why precursor data is not NAD+ data.
NAD+ — nicotinamide adenine dinucleotide — is a coenzyme, not a peptide. It contains no amino acids and sits on this site only because it is supplied the same way: a lyophilised vial reconstituted for research. That matters for safety, because "NAD+ side effects" covers three different things — the infused coenzyme, injected forms with no data, and oral precursors with quite a lot — and the tiers do not transfer to one another. It is a research reference beside the NAD+ profile, not advice.
The only published pharmacokinetic study of infused NAD+ in humans is the 2019 pilot by Grant and colleagues in Frontiers in Aging Neuroscience, which infused 750 mg of NAD+ intravenously over six hours in a small group of healthy men and sampled the plasma and urine metabolome throughout. Two findings bear on tolerability. First, plasma NAD+ did not rise measurably for the first two hours — the coenzyme was being cleared as fast as it arrived — before accumulating later, alongside rising urinary nicotinamide metabolites. Second, at that slow rate the authors reported no adverse events.
That is the entire controlled human record for the injected molecule. It was a pharmacokinetic study in a handful of people, not a safety trial, and it tested one dose at one rate by one route. The NAD+ research overview explains why extracellular NAD+ is broken down by CD38 and related enzymes before cells rebuild it, which is the mechanism behind the two-hour lag and, plausibly, behind the effects described next.
Intravenous NAD+ has been used in clinics since the 1960s, originally for alcohol and opioid withdrawal, without a controlled trial. That uncontrolled experience is large and consistent on one point: adverse effects are rate-dependent. Chest pressure or tightness, flushing, nausea, abdominal cramping, headache and a sense of anxiety are described during rapid infusion, and they typically resolve when the drip is slowed — which is why clinic infusions run over two to six hours. Whether these reflect the coenzyme itself, the nicotinamide released by its breakdown, or the ADP-ribose released by CD38 has not been established.
Because this is case-series and practice experience rather than trial data, it carries no incidence figure, no comparator and no verification of dose: enough to describe a pattern, not to quantify it. The NAD+ dosage notes set out what the literature and clinic practice administered by each route.
Subcutaneous and intramuscular NAD+ injections of tens of milligrams circulate in the research community with no published pharmacokinetic or safety data of any kind. What is reported informally is injection-site pain and swelling — plausible for a millilitre-scale volume of a mildly acidic, hyperosmolar solution — and the same flushing and nausea seen with infusion. These reports are uncontrolled and unverified, and no regulator has reviewed the injected route.
Randomised trials of nicotinamide riboside are the best-quality safety data in the field. The 2018 crossover trial in Nature Communications gave 1,000 mg of NR daily for six weeks to healthy middle-aged and older adults and reported it well tolerated with a side-effect profile similar to placebo; trials at up to 2,000 mg per day reported no serious adverse events. The 2021 trial in Science gave 250 mg of NMN daily for ten weeks to prediabetic women and reported the same. The caveat attached to all of them is nicotinamide: high exposure consumes methyl groups and has been associated with altered methylation markers, and nicotinamide at high doses can affect liver enzymes.
NR and NMN are precursors that cells convert to NAD+, taken orally at known doses in blinded trials. Injected NAD+ is the finished coenzyme, largely broken down outside cells into those same precursors and rebuilt inside them — but by a route, at a rate and with a plasma exposure that the oral trials never produced. The precursor safety record supports precursor products. It does not support the NAD+ vial sold on this page.
Beyond the methylation question above, the main one is malignancy: NAD+ is a substrate for cancer-cell metabolism as much as for healthy cells, and the effect of raising it in people with an existing cancer is unstudied — a caution the precursor literature itself raises. Because PARP enzymes consume NAD+ during DNA repair, PARP inhibitors used in oncology alter NAD+ dynamics, and no interaction study exists. Neither concern has been demonstrated; neither has been ruled out.
NAD+ is less stable in solution than a peptide. The pyrophosphate and nicotinamide-ribose bonds hydrolyse at neutral-to-alkaline pH and at room temperature, yielding NMN, AMP and free nicotinamide, and a degraded solution may show nothing more than a yellow tint. A vial reconstituted for weeks is therefore a partly different chemical from a fresh one, and some share of reactions in unregulated use plausibly traces to degradation products, endotoxin or the hyperosmolar volume rather than to the coenzyme. Reading the certificate of analysis and respecting the reconstitution date are the only ways to separate compound effects from product effects — see lab testing and COAs and the cold-chain and shelf-life guide. The NAD+ sold here is third-party tested by lot.
Injectable NAD+ has no Health Canada market authorisation and no Drug Identification Number, and research-grade material is sold in Canada for laboratory use only. That is distinct from the oral market, where nicotinamide riboside has been licensed as a natural health product; those licences do not extend to the injected coenzyme. The U.S. FDA did not evaluate NAD+ in its September 2023 section 503A determinations, so there is no category to cite, though it has acted against clinics making therapeutic claims for NAD+ infusions. NAD+ is not on the WADA Prohibited List; intravenous infusions above 100 mL in a 12-hour period are prohibited under M2 regardless of the substance infused.
Does NAD+ have a documented side-effect profile? Only in pieces. One six-hour infusion study reported no adverse events; clinic infusions describe rate-dependent chest tightness, flushing, nausea, cramping and headache; oral precursor trials report a profile similar to placebo. Injected NAD+ has no published safety data.
Why do clinic infusions cause flushing and chest tightness? The effects are rate-dependent and resolve when the infusion is slowed. Whether the coenzyme, its nicotinamide breakdown product or the ADP-ribose released by CD38 is responsible has not been established.
Does the NR and NMN safety data apply to NAD+ injections? No. Those trials gave oral precursors at known doses under blinding. Injected NAD+ is a different molecule by a different route with a different exposure, and it has no controlled trial of its own.
Is NAD+ a peptide? No. It is a dinucleotide coenzyme with no amino acids, listed here because it is supplied as a lyophilised vial like the peptides.
Is NAD+ prohibited in sport? It is not on the WADA Prohibited List. Intravenous infusions over 100 mL in 12 hours are prohibited under M2 whatever they contain.
Is this medical advice? No. This page summarises published research for laboratory reference. It is not medical advice, not a safety assurance, and not a dosing recommendation. NAD+ is sold here for research use only.
