Nicotinamide Riboside in Heart Failure: What the LVAD, HFrEF, and Atrial Fibrillation Trials Show

Nicotinamide riboside in heart failure is one of the most disciplined translational stories in the NAD+ field — three published clinical artifacts (a pilot LVAD trial, a placebo-controlled HFrEF safety trial, and the HF-AF ENERGY protocol), backed by mechanism work tying PARP1 activation and NAD+ depletion to dilated cardiomyopathy and atrial fibrillation. None of it yet proves NR improves clinical heart failure outcomes. All of it tells you why the question is being asked.

What this article covers

• Why the heart is an NAD+ tissue of interest
• The Wang 2022 NR-HF safety trial (HFrEF)
• PilotNR-LVAD: directly measuring failing myocardium
• HF-AF ENERGY: NR for atrial fibrillation burden
• HFpEF, ischemic cardiomyopathy, and adjacent trials
• Mechanism: why NAD+ depletion damages cardiomyocytes
• All five trials side by side
• Safety profile across the cardiac trials
• What the evidence does not prove
• Bottom line for clinicians and patients

Why the heart is an NAD+ tissue of interest

The adult human myocardium turns over its entire ATP pool roughly every ten seconds at rest and faster under load. Cardiomyocytes are mitochondria-dense by design: mitochondrial volume occupies 25-35% of cardiomyocyte cytoplasm, the highest fraction of any non-brown-fat tissue in the body (Schaper et al., 1985). The oxidative machinery that runs that turnover is NAD+-dependent at every step — Complex I of the electron transport chain accepts electrons from NADH, the TCA cycle consumes NAD+ in three reactions per turn, and beta-oxidation of fatty acids relies on NAD+ as the principal cofactor.

When NAD+ falls, mitochondrial flux falls with it. Failing myocardium shows measurably reduced NAD+ and NADH concentrations versus healthy controls, with the most consistent reductions in patients with end-stage HFrEF. Hershberger et al. (Circ Heart Fail 2017, PMID: 29242350) measured NAD+ in left ventricular tissue from explanted failing hearts and found 30-50% reductions in NAD+/NADH ratios compared with non-failing donor controls. The change predates and likely contributes to the contractile failure rather than being a downstream consequence — a point the preclinical models support.

This is the biology that motivated the NR cardiac program. If the failing heart is NAD+-deficient and the deficiency tracks with contractile dysfunction, an oral compound that doubles blood NAD+ and is well-tolerated is a defensible clinical experiment. Whether that experiment translates to outcomes is exactly what the published trials were designed to start answering — and where they fall short, by design.

The Wang 2022 NR-HF safety trial: what 2 g/day for 12 weeks actually showed

The first placebo-controlled trial of oral NR in heart failure was published by Wang, Airhart, and colleagues in JACC: Basic to Translational Science in December 2022 (PMID: 36644285; ClinicalTrials.gov NCT03423342). It enrolled 30 ambulatory stage C HFrEF patients on guideline-directed medical therapy, randomized 2:1 to NR (1,000 mg twice daily, total 2 g/day) or matching placebo for 12 weeks.

The trial was explicitly designed as a safety and tolerability study, not an efficacy trial. The primary endpoint was the adverse event rate. Secondary endpoints included laboratory parameters and whole blood NAD+ concentration. The full population was intentionally small; the goal was to clear the safety threshold for a larger Phase 2 program in a vulnerable population.

Pharmacological result

Whole-blood NAD+ rose by 30.0 ± 20.0 µM in the NR group versus -0.3 ± 2.0 µM in placebo (P < 0.001), roughly doubling baseline. This is consistent with the magnitude of NAD+ elevation reported in healthy cohorts on similar NR doses — see Martens et al. (2018) for the canonical healthy-adult reference. Stage C HFrEF patients with diuretics, beta-blockers, and ACE/ARB therapy on board still mounted the expected NAD+ response.

Safety result

There were no significant differences in adverse event rates, serious adverse events, or treatment-emergent laboratory abnormalities between the NR and placebo arms. Hematocrit and hemoglobin — closely watched because of prior signals in healthy-volunteer NR trials — showed only small numerical differences (NR -0.8% ± 0.6 vs placebo -0.4% ± 0.8 for hematocrit; not statistically significant). One NR participant discontinued for GI intolerance but completed follow-up. Out-of-range labs were rare (one creatinine elevation, one ALT elevation).

Exploratory biomarker signals

The exploratory biomarker work is where the trial got most interesting scientifically. PBMC (peripheral blood mononuclear cell) mitochondrial respiration improved with NR, and the magnitude of improvement correlated with the magnitude of NAD+ rise:

• Basal respiration improvement vs NAD+ change: R² = 0.413, P = 0.003
• Maximal respiration improvement vs NAD+ change: R² = 0.434, P = 0.002

NLRP3 inflammasome expression in PBMCs also dropped with NAD+ rise (R² = 0.330, P = 0.020), with directionally similar but non-significant trends for IL-1β, IL-6, IL-18, and TNF-α. PBMCs are a circulating readout, not myocardial tissue, but they are the practical compromise when you cannot biopsy a living heart. The signal that doubling blood NAD+ correlates with both improved respiration and reduced inflammatory priming in immune cells is the kind of mechanistic hook a Phase 2 trial would want to test directly.

PilotNR-LVAD: the first study to measure NR-exposed failing myocardium

Patients undergoing left ventricular assist device (LVAD) implantation for end-stage HFrEF provide one of the few opportunities to obtain fresh, viable failing-human-myocardium tissue. During LVAD insertion, surgeons excise an apical core of the left ventricle to create the inflow conduit — tissue that would otherwise be discarded. PilotNR-LVAD (NCT03727646) exploited this surgical window to measure whether NR actually changes NAD+ metabolism in the failing human heart.

Five participants scheduled for LVAD received NR, up-titrated over three days to 1,000 mg twice daily, with the final dose taken before surgery. Apical tissue removed during implantation was processed in the operating room for NAD+ and NADH quantification, mitochondrial function assays, and epigenetic profiling. The comparison group was historical controls — failing myocardium from LVAD patients who had not received NR.

The pilot was designed to provide proof-of-principle, not powered outcomes. With N = 5, it is hypothesis-generating by definition. The importance is methodological: this is the first trial to directly sample NR-exposed human cardiac tissue rather than inferring tissue effects from blood. Equivalent work in skeletal muscle was done by Elhassan et al. (2019) with conventional biopsy; the LVAD population was the only viable path to the analogous cardiac question.

As of 2026, peer-reviewed results from PilotNR-LVAD have not been published. The mechanistic follow-on study NCT04528004 ("Mechanistic Studies of Nicotinamide Riboside in Human Heart Failure") extends the same group's program with a larger sample and includes pre-and-post NR cardiac MRI alongside circulating NAD+ pharmacokinetics. The design documents are public on ClinicalTrials.gov; publication is pending.

HF-AF ENERGY: testing NR against atrial fibrillation burden

The HF-AF ENERGY trial protocol was published by van Marion, Wiersma, Brundel, and colleagues in Cardiovascular Drugs and Therapy in October 2022 (PMID: 36227441). It is the first registered human trial of NR with atrial fibrillation burden as a clinical secondary endpoint, and it sits at the intersection of two mechanism stories: NAD+ deficiency in the failing heart and NAD+ depletion as a direct driver of atrial electrical remodeling.

Trial design

HF-AF ENERGY is a 20-patient prospective intervention study enrolling ambulatory adults aged 18-80 with ischemic cardiomyopathy, paroxysmal or persistent AF, and an implantable cardioverter-defibrillator (ICD) with atrial sensing capability. The design uses each patient as their own control across two sequential 4-month periods: a baseline observation phase followed by a 4-month NR intervention up-titrated to 2 g/day (eight 250 mg capsules daily).

Endpoints

Primary endpoints are blood-based energy metabolite concentrations (NAD+/NADH and downstream nucleotides) plus mitochondrial function markers in PBMCs and circulating exosomes. AF burden — measured continuously by the ICD via remote rhythm monitoring across the observation and intervention periods — is the primary clinical secondary endpoint. Secondary biomarkers include circulating cardiomyocyte stress markers and structural remodeling indices.

Why ICD-based monitoring is the right readout

Conventional AF endpoints in drug trials suffer from sampling bias. Holter monitors capture 24-72 hours; event monitors capture symptomatic episodes; clinic ECGs catch whatever happens to be present. ICD atrial-lead monitoring is continuous and lossless across the full study period, making it the gold-standard readout for AF burden as a dependent variable. The trade-off is sample size: the population of HF patients with both ICDs and paroxysmal/persistent AF is small.

As of 2026, no published results are available from HF-AF ENERGY. The protocol documents the registered design and prespecified analysis plan. Result publication is pending. The trial is registered in the Netherlands Trial Register (Onderzoek met Mensen reference 52394).

HFpEF and ischemic cardiomyopathy: the adjacent evidence

Two further bodies of work bracket the HFrEF and AF trials and matter for any honest read of where NAD+ therapeutics stand in cardiac medicine: the HFpEF mechanism story and the Diao IV NAD+ ischemic cardiomyopathy trial.

HFpEF: nicotinamide in the mouse, NR in the wings

Abdellatif et al. (Sci Transl Med 2021, PMID: 33568522) used three independent HFpEF rodent models — aged C57BL/6J mice, Dahl salt-sensitive rats, and ZSF1 obese rats (a cardiometabolic syndrome model) — and showed oral nicotinamide supplementation improved diastolic function in all three. The mechanism traced to titin and SERCA2a deacetylation: NAD+ supplementation restored sirtuin activity, which deacetylated those two proteins, reducing passive cardiomyocyte stiffness and improving calcium-dependent active relaxation.

Tong et al. (Circ Res 2021, PMID: 33614947) extended this with NR specifically, showing that NR supplementation reversed HFpEF features in a two-hit mouse model (aged plus high-fat diet plus L-NAME-induced hypertension) and that human HFpEF myocardium showed downregulated NAD+ biosynthesis genes versus non-failing donor controls. The combined Abdellatif and Tong story made HFpEF — long considered a therapeutic wasteland — a credible NAD+ therapeutics target.

Human HFpEF NR trials have not yet been published. The mechanism case is sufficient that Phase 2 HFpEF trials of NR or related compounds are likely; absence of published human data is not absence of biological rationale.

Diao 2025: IV NAD+ in ischemic cardiomyopathy

Diao et al. (Am J Cardiovasc Drugs 2025, PMID: 40954388) published a 180-patient randomized placebo-controlled trial of intravenous NAD+ in adults with ischemic cardiomyopathy and LVEF ≤ 45% (NYHA II-III). Patients received either IV NAD+ at 10 mg/day or 5% glucose placebo for seven days, with transthoracic echocardiography at baseline and one-month follow-up, plus six-month clinical follow-up.

The trial reported significant echocardiographic improvements in the NAD+ arm versus placebo at one month. The interpretation requires caution on three fronts:

1. The dose was unusually low (10 mg/day IV) — well below the doses used in oral NR cardiac trials (2 g/day) and below most clinic IV NAD+ protocols (500-1,000 mg per session)
2. IV NAD+ pharmacokinetics are dominated by extracellular CD38 hydrolysis — see our IV NAD+ therapy evidence review for the mechanistic detail on why most infused NAD+ does not enter cells intact
3. The mechanistic pathway from a one-week IV course to a one-month echocardiographic change is not well-characterized

The trial is positive evidence that NAD+ supplementation in some delivery form may produce measurable cardiac effects in some populations. It is not interchangeable evidence for oral NR.

Mechanism: why NAD+ depletion damages cardiomyocytes

The cardiac NAD+ mechanism case rests on three convergent lines: PARP1 consumption, CD38 hydrolysis, and sirtuin substrate scarcity. Each is druggable in principle; each has been tested preclinically; the clinical translation is what the trials above are exploring.

PARP1 activation depletes cardiomyocyte NAD+

Zhang/Wiersma/Brundel (Nat Commun 2019, PMID: 30872574) provided the clearest mechanistic link between NAD+ depletion and atrial fibrillation pathophysiology. Tachypacing of HL-1 atrial cardiomyocytes and Drosophila hearts induced oxidative DNA damage, which activated PARP1 — the NAD+-consuming DNA damage response enzyme. PARP1 hyperactivation depleted cellular NAD+ pools and produced contractile dysfunction that mirrors clinical AF remodeling. Critically, both NAD+ replenishment and PARP1 inhibition prevented the contractile loss, identifying NAD+ repletion as a mechanistically plausible AF intervention.

The same pattern appears in HFrEF. Diguet et al. (Circulation 2018, PMID: 29217642) used a serum-response-factor-deletion mouse model of dilated cardiomyopathy and showed dietary NR supplementation reduced left ventricular contractile dysfunction and chamber dilation. The proposed mechanism: PARP1 chronic activation in the failing ventricular myocardium drains NAD+; NR restores supply; mitochondrial function recovers; contractile force improves. The DCM model is narrow, but the principle generalizes — chronic DNA damage in any long-lived post-mitotic tissue drives PARP1 activation, which drains NAD+.

CD38 hydrolysis in cardiac tissue

CD38 is the dominant NAD+-degrading ectoenzyme in most tissues and a principal driver of age-related NAD+ decline — see our CD38 inhibitors overview and the CD38 mechanism page for the full picture. In cardiac tissue, CD38 is upregulated in failing myocardium and in inflammatory states. This means oral NR is fighting a rising-degradation-rate problem: supply goes up, but so does consumption. The implication, not yet tested clinically in heart failure, is that CD38 inhibition might complement NR supplementation — plug the leak while pouring water into the top.

Sirtuin substrate scarcity

Sirtuins (particularly SIRT1, SIRT3, and SIRT6) are NAD+-dependent deacetylases that regulate mitochondrial biogenesis, fatty acid oxidation, and DNA repair. When NAD+ falls, sirtuin activity falls, and the proteins they normally keep deacetylated accumulate acetyl marks that impair function. The Abdellatif HFpEF work specifically identified titin (cardiomyocyte structural protein governing passive stiffness) and SERCA2a (calcium pump governing active relaxation) as sirtuin substrates whose hyperacetylation contributes to diastolic dysfunction. NAD+ restoration via nicotinamide reversed both.

For the broader sirtuin biology and why NAD+ is the rate-limiting cofactor, see our sirtuin mechanism page.

All five trials side by side

Synthesizing the cardiac NR evidence requires putting the trials in the same frame. The table below is the working comparison the editorial team uses internally; each citation is verifiable via PubMed.

Read across the rows, a pattern emerges. The two trials with published clinical results — Wang 2022 and Diao 2025 — used different compounds (NR vs IV NAD+), different doses (2 g/day oral vs 10 mg/day IV), different populations (HFrEF vs ischemic cardiomyopathy), and different endpoints (safety vs echocardiographic LVEF). Neither is the trial you would design if you wanted to settle "does oral NR improve clinical heart failure outcomes." That trial has not yet been run.

Safety profile across the cardiac trials

The aggregated safety signal from the published NR cardiac work is favorable, with three caveats worth naming.

Adverse events overview

In the Wang 2022 NR-HF trial, NR and placebo arms had statistically indistinguishable rates of all AE categories: total AEs, serious AEs, and treatment-emergent lab abnormalities. The most common AE in the NR arm was GI complaint (4 events in 19 NR patients vs 3 events in 11 placebo patients), consistent with the GI tolerability profile in healthy-volunteer NR trials.

Hematocrit signal

Earlier healthy-adult NR work raised a small but consistent hematocrit-reduction signal — most notably Conze et al. (2019) in the 8-week 1,000 mg/day Niagen trial. The Wang HFrEF trial did not detect a significant hematocrit effect at 2 g/day for 12 weeks (-0.8% ± 0.6 NR vs -0.4% ± 0.8 placebo), though the trial was underpowered to rule out a small effect in this direction. For our broader treatment of NR long-term safety, see long-term NAD+ precursor safety: what we know and don't.

Arrhythmia signal

No published NR trial in any population has reported a pro-arrhythmic signal attributable to NR. The HF-AF ENERGY trial is specifically designed to detect changes in AF burden in either direction — its readout could in principle reveal a pro-arrhythmic signal if one exists, though the mechanism case predicts the opposite.

Drug interactions

Heart failure patients are typically on multi-drug regimens — diuretics, beta-blockers, ACE inhibitors or ARBs, mineralocorticoid receptor antagonists, SGLT2 inhibitors, anticoagulants. No clinically significant NR-drug interactions have been published. See our NAD+ precursor drug interactions evidence review for the broader pharmacology, and discuss any NR addition with the managing cardiologist before initiation.

What the evidence does not prove

The cardiac NR evidence is the kind of mechanistically coherent, safety-cleared, biomarker-supported early-phase work that justifies Phase 2 outcome trials. It does not yet prove the questions that matter most to patients and clinicians:

• Does oral NR improve left ventricular ejection fraction? No published trial of oral NR has reported a significant LVEF change.
• Does oral NR reduce HF hospitalizations or mortality? No trial has been powered for these endpoints.
• Does NR reduce atrial fibrillation burden in humans? The mechanism case is strong; HF-AF ENERGY is the first registered trial designed to answer this, and results are not yet published.
• Does oral NR raise myocardial NAD+ specifically? PilotNR-LVAD was designed to answer this for failing myocardium; results pending.
• How does NR compare to optimized guideline-directed medical therapy? No head-to-head trial exists, and none would be the right design — NR would be added on top of GDMT, not substituted for it.

Evidence grading per our methodology page: NR safety in HFrEF is graded emerging (one placebo-controlled trial, small N, consistent signal). NR efficacy on any cardiac clinical endpoint is graded preclinical-dominant — robust animal and mechanism data, no positive human outcome trial yet.

Bottom line for clinicians and patients

The honest summary of nicotinamide riboside in heart failure as of 2026:

1. NR at 2 g/day appears safe in stable HFrEF for 12 weeks (Wang 2022, N = 30). Larger and longer trials are needed; the existing data is consistent with the broader healthy-adult NR safety record.
2. NR doubles whole-blood NAD+ in HFrEF patients on guideline-directed therapy, consistent with the pharmacological response in healthy adults.
3. Mechanistic exploratory signals are favorable — improved PBMC mitochondrial respiration, reduced NLRP3 expression — but PBMCs are a circulating readout, not myocardium.
4. The HF-AF ENERGY trial is the first attempt to test whether NR translates to a clinical AF outcome. Results are pending.
5. No published oral NR trial has demonstrated improved LVEF, hospitalization, or mortality in heart failure. Trials adequately powered for those endpoints do not yet exist.
6. Any cardiac patient considering NR should coordinate with their cardiologist. The safety signal is favorable but the interaction profile with the full HF medication stack has not been comprehensively tested.

For the broader landscape of NR pharmacokinetics and how it compares to NMN as a precursor, see NR vs NMN: what head-to-head research actually shows and the dose-response detail in NAD+ precursor dosing in human trials. For the mechanism layer underlying age-related NAD+ decline that motivates the cardiac translational program, see NAD+ decline with age: why tissue levels drop 50% by 70 and the supply-side picture in NAMPT, the rate-limiting enzyme behind NAD+ decline.

Heart failure is a serious chronic condition with high-quality evidence-based therapies that demonstrably reduce mortality and hospitalization. NR is not a substitute for any of them. Whether it becomes a meaningful add-on therapy is the question the next decade of properly powered trials will answer — and the question the currently published cardiac trials, taken together, justify asking.