Blood NAD tracks heart disease, not age, in 1,068 adults

NAD has been marketed for years as a molecule whose decline drives human ageing, but one of the largest population measurements yet conducted suggests the story is more complicated. In a cross-sectional study of 1,068 adults in Lithuania, researchers found that whole-blood NAD levels do not fall with age in healthy people. Instead, the levels remain stable in men across the age range studied and actually increase in women, while the only clear reductions appeared in people living with chronic disease, specifically cardiovascular disease and hypertension. The work, published in Nature Metabolism, indicates that whole-blood NAD tracks cardiovascular disease rather than chronological age.

The team, led by Diana Romenskaja-Avgul and colleagues with senior authorship shared by Marius Dagys and Dalius Ratautas, carried out 1,340 NAD measurements in 1,068 adults recruited across a range of ages, sexes and demographic backgrounds between 2023 and 2025. All procedures were approved by the Vilnius Regional Biomedical Research Ethics Committee, and participants gave informed consent after completing a structured questionnaire covering supplement use, chronic disease history, pregnancy status and lifestyle factors. The study was explicitly not designed to test causality, and the authors state that the data do not provide direct evidence that NAD supplementation could treat cardiovascular disease or hypertension.

Participants collected their own blood samples using finger-prick dried blood spot kits containing a lancet, alcohol tissue, plaster and blood collection card. To standardize conditions, they were instructed to sample in the morning after an overnight fast, before consuming any food or beverages, after washing and drying their hands and letting the disinfectant dry completely. Once a sufficiently large drop of blood formed, it was applied to the collection paper without excessive pressure, air-dried for at least three hours, sealed in a foil bag and mailed to the laboratory. Samples were stored at −20 °C, where NAD on collection cards remains stable for at least 30 days without observable loss in concentration.

Analysis took place within one week of receipt in a licensed clinical laboratory operating under an ISO 15189-evaluated quality management system. Three-millimetre discs were punched from the dried blood spot cards into 96-well plates, extracted in buffer for 30 minutes at 25 °C with shaking, and then subjected to an enzymatic cycling assay in which absorbance at 450 nm was measured before and after a four-hour incubation in the dark. NAD concentration was calculated from the change in absorbance against a calibration curve. In a parallel validation workflow, heat-treated aliquots were used to isolate NADH, allowing NAD+ to be calculated by subtraction from the total NAD pool.

The measurement platform itself was validated across multiple sites and technologies. Paired liquid whole-blood and dried blood spot samples from 31 individuals showed a strong correlation, with a Pearson’s r of 0.93. In a further cross-format test, paired samples from five individuals were analysed independently at three sites by orthogonal methods: the dried blood spot enzymatic cycling assay in Lithuania, a liquid whole-blood colourimetric assay at Vilnius University, and frozen whole-blood liquid chromatography–mass spectrometry at the University of Helsinki in Finland. These comparisons yielded Pearson’s correlations of 0.87 to 0.91, supporting robustness across analytical technologies, laboratories and countries. In-house quality controls using NADase-treated blood, NAD-spiked samples and pure NAD standards confirmed expected assay responsiveness across the measurement range, and all testing sites were blinded to sample values.

A defining feature of dried blood spots is that the NAD+/NADH ratio collapses to equilibrium within the first hour of drying, while the total NAD pool is conserved. In a 30-day storage time course at −20 °C involving two donors with three biological replicates per timepoint, 54 independent samples in total, total NAD remained stable, with means of 24.5 µM and 21.7 µM and coefficients of variation of 6.4% and 7.1%. In liquid whole capillary blood from healthy donors, NAD+ is the dominant species at roughly nine to one in the researchers’ measurements, so total NAD closely approximates NAD+ in that context. The authors nonetheless report total NAD as the primary quantity and note that they do not assume this approximation holds where the liquid NAD+/NADH ratio may differ, such as in disease contexts.

The healthy population was defined by strict inclusion criteria: participants had to report no use of vitamin B3 or NAD-boosting supplements, no diagnosis or suspicion of chronic disease, and no medications for chronic disease. Age stratification used five-year intervals, and the youngest [18,20) age groups differed significantly from the rest of the subgroups, so they were excluded from all analyses, which were restricted to participants older than 19 years. Because the male NAD distribution deviated significantly from normality, the researchers used non-parametric methods, reporting medians with 95% confidence intervals and applying two-sided Mann–Whitney U-tests with significance defined at P < 0.05.

For the chronic disease analysis, participants were classified only when a reported diagnosis was supported by medication compatible with that diagnosis, a criterion met by 243 participants. After excluding 46 people whose medication was uninterpretable, non-disease-specific or inconsistent with the reported diagnosis, 197 participants, 116 females and 81 males, remained available for stratification. Disease categories included cardiovascular disease with subgroups of dyslipidaemia, hypertension and other cardiovascular conditions, type 2 diabetes, thyroid disease, asthma, immune-mediated inflammatory disease, oncological disease, other diseases and multimorbidity, defined as two or more chronic diseases. Categories were non-exclusive by design to reflect the comorbidity structure of the ageing population, and all classifications were performed blinded to whole-blood NAD values. Disease groups with fewer than 15 participants were excluded from statistical testing, and because most disease groups were small, age was not binned; instead, NAD was analysed against age directly by linear regression, and female disease groups were compared against an age-matched healthy reference standardized using the healthy female age-slope.

Against this backdrop, the central finding stands out: in healthy individuals, whole-blood NAD does not decline with age, remaining stable in males and increasing in females, while levels are reduced in chronic disease, specifically cardiovascular disease and hypertension. P values from the disease-versus-healthy comparisons were adjusted for multiple testing using the Holm–Bonferroni procedure, with both original and adjusted values reported in the paper’s extended data tables. The researchers also examined lifestyle factors, including smoking habits, physical activity level and dietary patterns, by filtering datasets according to sex, age range, supplement use and chronic disease status, then calculating medians with 95% confidence intervals and performing pairwise Mann–Whitney tests; one such analysis restricted the population to omnivores to control dietary variation.

The study also probed the temporal dynamics of NAD under self-directed supplementation. In one exploratory arm, two individuals were followed over 19 days of continuous sampling, covering five days before and 14 days during voluntary oral use of a commercially available pure NMN at 1,000 mg per day, self-purchased and self-consumed independently of the investigators, with the manufacturer not disclosed to avoid commercial affiliation. Morning samples collected before food or beverage intake enabled high-resolution characterization of whole-blood NAD concentrations under these conditions, and parameters including saturation half-time, baseline and peak NAD were estimated by fitting the data to a previously derived first-order equation. Two other sub-studies assessed variability: one tracked a single male and a single female participant for one year to assess long-term individual variability, and another sampled nine participants three times daily for seven consecutive days, capturing short-term intra-individual variability under standardized morning, post-lunch and bedtime sampling conditions. The complete analysis pipeline, including Python scripts and synthetic data for functional assessment, is openly available on Zenodo, while the anonymized metadata are deposited under restricted access and available to qualified researchers upon reasonable request. The authors’ conclusion is measured: whole-blood NAD levels track cardiovascular disease rather than chronological age, and because the research was not designed to investigate causality, it does not provide direct evidence that NAD supplementation may serve as a treatment for cardiovascular disease or hypertension.

Subject of Research: Whole-blood NAD levels in ageing and chronic disease

Article Title: Whole-blood NAD levels track cardiovascular disease but not chronological age

Article References: Romenskaja-Avgul, D., Stakelytė, D., Laurynėnas, A., Ramonas, E., Zabitaitė, R., Cibulskė, S., Danilevičienė, S., Danilevičius, R., Ratautė, K., Dagys, M., & Ratautas, D. (2026). Whole-blood NAD levels track cardiovascular disease but not chronological age. Nature Metabolism. https://doi.org/10.1038/s42255-026-01660-3

Image Credits: AI Generated

DOI: 10.1038/s42255-026-01660-3

Keywords: NAD, NAD+, cardiovascular disease, hypertension, ageing, dried blood spots, NMN supplementation, cross-sectional study, biomarkers, metabolism, chronic disease, Lithuania