The NAD+ research, read straight: mechanism, decline, and the trials that moved the needle.

In plain English

NAD+ research splits into two questions. First, what does the molecule do? It carries electrons to make energy, and it is the fuel for maintenance enzymes that repair DNA and tune metabolism. Second, can you raise it usefully? In people, swallowing the precursors NMN or NR reliably lifts the NAD+ measured in blood — that part is settled. What is not settled is whether that lift makes anyone healthier or live longer; the strongest 'reversal' data are still in mice. IV NAD+ has the least evidence of all. Everything below is what specific studies found, not advice.

Mechanism: a redox carrier and a consumed substrate

NAD+ does two jobs at once. As a redox couple it cycles between NAD+ and NADH, accepting electrons during the breakdown of food and donating them into the mitochondrial electron transport chain to drive ATP synthesis ^[5]. As a signaling substrate it is consumed — physically used up — by three enzyme families. Sirtuins (SIRT1-7) are NAD+-dependent deacylases that regulate metabolism, stress resistance and, in model organisms, lifespan; their activity is rate-limited by how much NAD+ is available ^[8]. PARP1 (poly(ADP-ribose) polymerase 1) is a DNA-repair enzyme that burns large amounts of NAD+ when it responds to DNA damage ^[10]. CD38 is an NAD-consuming ectoenzyme that rises with age ^[2]. Because all three draw on the same pool, NAD+ is the shared currency linking energy metabolism to DNA repair and inflammation ^[7].

The supply side is the salvage pathway: nicotinamide is recycled to NMN by the rate-limiting enzyme NAMPT, then to NAD+ by NMNAT ^[5]. NR enters one step earlier, converted to NMN by NRK kinases ^[5]. This is why precursors work where oral NAD+ does not — they feed routes the cell already uses to build NAD+.

Age-related NAD+ decline

Tissue NAD+ declines with age across organisms and tissues ^[5]. In human skin and brain the fall is on the order of ~50% by middle to late life, tracking with declining SIRT1/PGC-1alpha activity and a state of disrupted nuclear-mitochondrial communication the authors termed pseudohypoxia ^[6]. The decline is driven in part by rising consumption: CD38 is the principal NAD+-consuming enzyme that increases with age, and deleting it in mice preserves NAD+ and SIRT3 activity and improves mitochondrial and metabolic health with age ^[2]. Cellular senescence feeds the same loop — senescent cells secrete factors that activate CD38+ macrophages, and clearing senescent cells partly restored tissue NAD+ in mice ^[9]. A parallel axis runs through DNA repair: when PARP1 over-activates on DNA damage it depletes NAD+ and, by substrate competition, inhibits SIRT1; aged mice show this signature, and raising NAD+ reverses parts of it in those models ^[10].

What outcomes the studies actually measured

Framed as endpoints rather than 'NAD+ benefits', the human record is specific. Oral NR at 100/300/1000 mg/day for 8 weeks raised whole-blood NAD+ by 22%, 51% and 142%, without elevating LDL cholesterol or disrupting one-carbon metabolism ^[4]. Oral NMN at 300/600/900 mg/day for 60 days raised blood NAD+ across all doses and improved six-minute walking distance versus placebo, with 600 mg/day identified as the optimal dose in that trial ^[3]. In prediabetic, postmenopausal women, 250 mg/day of NMN for 10 weeks increased muscle insulin sensitivity measured by hyperinsulinemic-euglycemic clamp, with no change in body composition or HbA1c ^[11]. In aged mice, long-term NMN at 100-300 mg/kg/day for 12 months suppressed age-associated weight gain and improved energy metabolism, insulin sensitivity, eye function and bone density ^[13]. The pattern: blood-NAD+ elevation is reproducible; functional endpoints are real but mixed; the largest 'anti-aging' effects remain in rodents ^[14].

IV NAD+ therapy in the research literature

NAD IV therapy — intravenous infusion of NAD+ in wellness and clinical settings — has the weakest controlled evidence of any route. It is an unapproved compounded therapy, marketed aggressively, on a thin base of pilot and retrospective data; reviews call for rigorous randomized trials before any clinical claim ^[14]. There is also a documented safety signal: a compounded injectable NAD+ product was subject to an FDA Class I recall for elevated bacterial endotoxin, and rapid infusion can cause chest pressure, abdominal discomfort, flushing and nausea. None of this establishes a clinical benefit; it describes an unapproved therapy with real quality risks.

Injectable and IV NAD+: pharmacokinetics and tolerability

The pharmacokinetics explain the skepticism. NAD+ itself is not freely taken up intact by most cells; infused IV NAD+ is rapidly cleared from plasma, with near-complete plasma removal within roughly the first two hours of an infusion — a pilot study found infused NAD+ was extensively metabolized extracellularly before plasma NAD+ rose ^[14]. That is the opposite of the oral-precursor profile, where blood NAD+ climbs over days to weeks and stays elevated through chronic dosing ^[4]. Subcutaneous and intramuscular NAD+ injection is compounded with minimal peer-reviewed pharmacokinetic data, and sublingual, intranasal and transdermal-patch products carry little controlled evidence at all. Where infusion-related symptoms occur they track infusion rate and resolve on completion.

NAD+ in inflammation and DNA repair

Beyond aging metabolism, NAD+ sits at the center of the inflammatory and DNA-repair literature — almost entirely preclinical. In mouse macrophages, pharmacological NAD+ depletion (via the NAMPT inhibitor FK866) primed NLRP3 inflammasome assembly and caspase-1 activation, an effect reversed by restoring NAD+ with NMN ^[12]. In LPS-activated macrophages, mitochondrial-ROS-mediated DNA damage activated PARP and consumed NAD+, making the cells dependent on NAMPT-driven salvage to sustain their metabolism ^[15]. As a review frames it, NAD+ is the limiting substrate that couples PARP1-mediated DNA repair to energy depletion and, when PARP over-activates, to cell death ^[10]. These are mechanistic findings in cells and mice, tagged PRECLINICAL — they map the biology, not a human treatment.