NAD+ Dosage: What a Dose Would Require

Ask for an NAD+ dosage and figures arrive quickly. Some are milligram quantities per capsule, some are gram quantities per infusion bag, and they differ from one another by orders of magnitude without anyone remarking on it. That spread is not noise in the data. It is the signature of numbers that came from different interventions and were then filed under one heading.

NAD+ dosage: the conversion nobody performs

Most quantities in circulation originate in trials of oral precursors, principally nicotinamide riboside and nicotinamide mononucleotide. Those trials administered a specific mass of a specific compound to human participants and reported what happened.

Carrying such a number across to injected NAD+ requires three substitutions at once. A different molecule is substituted for the one tested. A different route replaces the one used. And a different pharmacokinetic profile is assumed to follow, without being measured. Each substitution invalidates the number on its own. Performed together, they leave a figure with the appearance of a citation and none of the content of one.

Milligrams of a precursor are not milligrams of the coenzyme

The most basic problem is arithmetic rather than clinical. Nicotinamide riboside and NAD+ are different molecules with different molecular masses. A given mass of the precursor does not contain, and cannot yield, an equal mass of the coenzyme.

Nor does conversion run to completion. A swallowed precursor is subject to absorption across the gut wall, first pass processing in the liver, distribution, competing metabolic fates and excretion, and only a fraction of the administered mass is converted into the coenzyme in any tissue. The size of that fraction differs between the two common precursors, between tissues and between individuals, and it is not a constant that a reader could apply even if it were published.

So the mass on a precursor label and the mass in an infusion bag are not comparable quantities in any respect. They are two different compounds, delivered by two different routes, and the numbers were never on the same scale.

An endogenous pool is a moving baseline

Every other compound covered on this site is absent from the body until it is administered. That makes the administered quantity the whole of the exposure. NAD+ inverts this: a substantial pool already exists, is synthesised continuously, is consumed continuously, and is under regulatory control.

Adding to such a pool is not the same operation as introducing a foreign molecule. The relevant question is not how much was given but how much the pool changed and for how long, in the tissue where an effect is proposed. That is a measurement, not a calculation, and it requires knowing the baseline, the turnover rate and the regulatory response of the salvage enzymes to a sudden supply. None of those quantities is available to someone reading a vendor page.

Route changes what arrives at the cell

There is a further complication specific to the injected form. NAD+ is a large charged molecule, and extracellular enzymes including CD38 and surface nucleotidases degrade it outside cells into smaller fragments. Those fragments are themselves salvage substrates.

If a meaningful proportion of an infused dose is broken down before uptake, then the quantity in the bag and the quantity of intact coenzyme reaching an intracellular compartment differ by an unknown factor. The published literature has not settled how large that factor is in humans. Until it does, an infusion volume describes what left the bag rather than what reached a cell, and stating it as a dose implies a precision the pharmacology does not support.

Concentration, volume and delivered quantity are three measurements

Infusion figures are quoted as a single number, which hides the fact that three separate values determine what a person receives.

Concentration is a property of the solution and belongs to whoever prepared it. Volume is the amount of that solution placed in the bag. Delivered quantity is the product of the two, less whatever remains in the line, and it is the only one of the three that describes the person rather than the equipment. A page that states a quantity without stating concentration has not given enough information to reconstruct any of them.

For a compounded sterile product there is a fourth value underneath all of these, which is whether the concentration in the container matches the concentration on it. That is an analytical question about a specific batch, answerable only by testing that batch, and it is unrelated to any clinical reasoning about how much a person should receive.

The specifications a stated quantity has to carry

A number becomes a dose only when the following are attached to it. The table below records which of them are present for the figures that circulate on this topic.

The pattern is that the left column belongs to a research programme and the right column belongs to a service. A service can be competently run and still produce no quantity that means anything outside its own chair, because it was never designed to establish one.

What is actually being copied when a figure circulates

Trace an infusion volume back through the pages repeating it and the trail usually ends at another commercial page rather than at a study. The figure is stable across sources because it is being copied, and copying produces agreement without producing evidence.

Agreement between sources is the weakest form of corroboration there is, and it is the one most likely to persuade, because a reader who checks three pages and finds the same number treats the repetition as independent confirmation. Independence is the property that makes corroboration worth anything, and it is exactly what copying removes.

A quick test separates a copied figure from a derived one. Ask what the number would have to be measured against to be wrong. A quantity established by dose ranging can be wrong, because a study could show that a different quantity produced the outcome better or that neither did. A quantity that arrived by copying cannot be wrong in that sense, because no outcome was ever attached to it. Figures that cannot be wrong are not conclusions, and treating them as conclusions is the mistake the uniformity encourages.

What a defensible number would require

Establishing a dose for injected NAD+ would mean running the work that has not been run: pharmacokinetic studies in humans characterising what happens to the infused molecule, dose ranging against a prespecified outcome rather than against a satisfaction question, and controlled comparison against placebo infusion. That programme would produce a number attached to a route, a population and an endpoint.

Nothing shorter substitutes for it. Not a longer list of clinics using the same volume, not a mechanism argument, and not a precursor trial, however well conducted, since a well conducted trial of a different molecule remains a trial of a different molecule.