Dietary Nitrate Provides Sustained Blood Pressure Lowering in Hypertensive Patients: A Randomized, Phase 2, Double-Blind, Placebo-Controlled Study

Systemic hypertension remains the largest attributable risk factor for mortality worldwide.^1,2 Worryingly, the scale of the problem is increasing, with the proportion of adults with hypertension predicted to increase to almost 1 in 3 (1.57 billion) by 2025.³ Despite >60 years of innovation in the pharmacotherapy of hypertension,⁴ only ≈half of hypertensives are treated for their blood pressure (BP) and of those only ≈half are controlled to guideline-driven targets.^5,6 Thus, novel therapeutic strategies including dietary approaches are of great interest.

An approach that has been explored in the treatment of hypertension is the delivery of the potent vasodilator nitric oxide (NO). Endothelial NO generation, achieved through the conventional l-arginine/NO synthase pathway, plays a critical role in sustaining vascular health. However, in most cardiovascular diseases (CVDs), including hypertension, the levels of endothelial NO are diminished.^7,8 Accordingly, therapeutic strategies restoring NO levels in hypertension have been explored. However, supplementation with substrate (l-arginine) and other essential cofactors required for healthy NO generation from NO synthase have yielded equivocal results.⁹ In addition, NO donors (organic nitrates), such as nitroglycerin, have also been tested but having problems of induced endothelial dysfunction¹⁰ and tachphylaxis¹¹ that have limited their clinical utility.¹²

However, NO production from the chemical reduction of inorganic nitrite (NO₂⁻), a phenomenon previously thought to occur only in extreme acidosis^13–15 has emerged as a potential pathway that might be exploited as a method for delivery of NO to the blood vessel. Evidence in healthy volunteers suggests nitrite reduction occurs readily within the circulation after elevation of plasma nitrite levels, by provision of dietary or oral inorganic nitrate salts.^16,17 It is now accepted that a significant proportion of orally ingested inorganic nitrate once absorbed across the upper intestine is extracted from the blood via the salivary glands and secreted into the oral cavity where it comes into contact with symbiotic bacteria that reduce inorganic nitrate (NO₃⁻) to nitrite.¹⁸ On swallowing the saliva, the nitrite then enters the circulation,^16,17,19 where it meets mammalian nitrite reductases that convert it to NO resulting in vasodilation^20,21 and significant BP lowering.^16,17,19 Exploitation of this NO₃⁻-NO₂⁻-NO (alternative) pathway by supplementation with dietary sources of NO₃⁻ (eg, beetroot juice) is associated with elevations in both plasma nitrite and cyclic guanosine monophosphate (cGMP, an exquisitely sensitive marker of bioactive NO production)²² and significant BP reductions over 24 hours in healthy volunteers and drug-naive stage 1 patients with hypertension.^17,23 We explored in this phase 2 clinical study whether a once daily dietary nitrate supplementation for 4 weeks would confer sustained BP reduction in both drug-naive and treated patients with hypertension.

Of the 151 patients screened for the study, 68 patients were enrolled, of whom 64 completed the study protocol and had complete data. Four patients withdrew after randomization but before the first study visit and returned no analyzable data (Figure S2). Both dietary nitrate and placebo interventions were well tolerated. No serious adverse events were reported. Common, expected findings were beeturia and fecal discoloration. All patients completed the dietary interventions for the duration of the study. The nitrate content of the active treatment juice was 25.7±5·3 mmol/L, giving ≈6·4 mmol nitrate in a 250 mL daily dose. The nitrate content of the placebo juice was 0.028±0.008 mmol/L, giving ≈0.007 mmol nitrate daily. The nitrite content was below the limits of detection in both interventions (ie, <50 nmol/L). It has been shown previously that there are no significant differences in major cationic components (ie, sodium, potassium) of the active and placebo juices.²⁵ Demographics and baseline screening characteristics were similar in both treatment allocation groups (Table). All patients were confirmed to have significant hypertension by ABP monitoring at trial inception.²⁴

Dietary nitrate ingestion was associated with elevations in both nitrate and nitrite concentrations in all biological compartments assessed. These levels returned to baseline after a 2-week washout period (Figure 1; Figure S3). Importantly, plasma nitrite concentrations were also elevated from baseline by ≈2.7-fold with a mean Δ0.52 μmol/L (95% CI, 0.39–0.65) after consumption of dietary nitrate with no change in the placebo limb (P<0.001; Figure 1A). The change in circulating nitrite and nitrate levels in the dietary nitrate limb was also associated with an ≈1.4-fold increase in plasma cGMP concentrations with no change in the placebo limb (P<0.001, Figure 1D). There were no changes in transcutaneous arterial methemoglobin concentrations after either intervention (Figure S3D) and no changes from baseline or between groups in glycemic, serum biochemical, or hematological indices (Table S1).

The 2 treatment allocation groups were well matched for baseline (preintervention) BP and heart rate measured by the methods utilized in this study (Table S2). Consumption of dietary nitrate was associated with decreases in clinic BP (Figure 2). Clinic systolic BP (SBP) and diastolic BP (DBP) decreased compared with baseline by 7.7 mm Hg (95% CI, 3.5–11.8; P<0.001) and 2.4 mm Hg (95% CI, 0.0–4.9; P=0.050), respectively; changes not evident in the placebo group (Figure 2A and 2C). Similarly, 24-hour ABP measurements exhibited a similar pattern with a mean decrease in SBP and DBP compared with baseline of 7.7 mm Hg (95% CI, 4.1–11.2, P<0.001) and 5.2 mm Hg (95% CI, 2.7–7.7, P<0.001), respectively (Figure 2B and 2D). Observation of the hourly profile of the change in 24-hour ABP after intervention revealed that consumption of dietary nitrate was associated with reduction in BP over the entire 24-hour period for both SBP and DBP (Figure S4) compared with placebo. Splitting 24-hour ABP into day-time (07:00–23:00) and night-time (23:00–07:00) periods, dietary nitrate consumption was associated with decreases in BP in both time periods (Figure S5).

Home BP was reduced within 1 week of consumption of dietary nitrate, but not placebo, for both SBP and DBP and reduced over the entire 4-week intervention period (Figure 3A and 3B). Peak decreases in BP occurred at week 6 (ie, last week of dietary nitrate intervention), with decreases in SBP compared with placebo of 8.1 mm Hg (95% CI, 3.8–12.4; P<0.001) and in DBP of 3.8 mm Hg (95% CI, 0.7–6.8; P<0.01; Figure 3A and 3B). After washout, both SBP and DBP started to return to baseline (Figure 3). There were no changes relative to baseline or compared with placebo in heart rate after dietary nitrate intervention by any method used (Figures 2E, 2F, and 3C; Figure S5E and S5F). Change in plasma nitrite from baseline in the intervention arm correlated inversely with change in SBP measured in clinic and at home, but not with ABP (Figure S6A–S6C).

Dietary nitrate consumption was also associated with improvements in vascular function (Figure 4). Pulse wave velocity (PWV) was reduced after dietary nitrate consumption by 0.59 m/s (95% CI, 0.24–0.93; P<0.01) compared with baseline and 0.58 m/s (95% CI, 0.05–1.10; P<0.05) compared with placebo. Augmentation index was also reduced after dietary nitrate consumption by 5.2% (95% CI, 2.9–7.5; P<0.001) compared with baseline and 6.1% (95% CI, 3.0–9.1; P<0.01) compared with placebo. Baseline brachial artery diameter and time to peak dilatation after reactive hyperemia were similar across all treatment groups and visits (Table S3). However, dietary nitrate consumption was associated with an increase in peak flow-mediated dilatation (FMD) of 1.0% (95% CI, 0.3–1.5; P<0.001) not evident in the placebo limb.

A priori subgroup analyses were conducted by stratification for baseline hypertension treatment status (drug-naive or treated). We recruited n=34 into each subgroup. Four patients dropped out before receiving intervention: all 4 were in the treated hypertension arm, split equally between intervention and placebo arms. There were no differences in baseline demographics, characteristics and BPs of the subgroups (Table S4), and between the placebo and dietary nitrate subgroups.

Dietary nitrate consumption produced similar rises in plasma, urine, saliva nitrite and nitrate, and plasma cGMP in both stratified limbs (Table S5). Importantly, BP was reduced compared with placebo by all methods used in both subgroups (Figures S7 and S8). PWV velocity was not altered in drug-naive patients after dietary nitrate consumption, although endothelial function trended to improvement (P=0.06) though this failed to reach conventional levels of statistical significance (Figure S7C and S7D). In the treated hypertensives, PWV velocity was reduced after dietary nitrate (0.82±0.87 m/s; P<0.01), and endothelial function was improved (change in FMD compared with placebo: 1.1%; 95% CI, 0.1–2.0; P=0.04; Figure S8C and S8D).

Previously the oxidation of endogenously generated NO to nitrite, and nitrate was viewed as a 1-way, linear termination of NO activity. However, the discovery of the NO₃⁻-NO₂⁻-NO reductive pathway, dependent on the enterosalivary circuit, has led to a radical revision of our understanding of the pathways that govern endogenous NO generation and NO metabolism.²⁶ This novel paradigm reveals the nitrogen oxides to be in an NO cycle that can be augmented through the provision of inorganic nitrate, given either by dietary or inorganic supplementary route. By capitalizing on this cycle, herein we show that a once daily dietary nitrate intervention augments NO generation through this pathway in patients with hypertensionto lower BP. We have demonstrated that the intervention is well tolerated, safe, and is associated with robust BP reductions measured in and out of clinic.

Dietary nitrate supplementation providing ≈6 mmol nitrate daily for 4 weeks, caused substantial increases in plasma nitrate concentrations (≈5.5-fold), and is similar to the peak fold increases in previous studies with similar doses.²³ Plasma nitrite concentration was elevated by ≈2.7-fold from baseline; the reduced fold increase compared with plasma nitrate concentrations reflecting use of the enterosalivary circulation in the bioconversion of ingested nitrate to nitrite that is critically dependent on the symbiotic relationship between oral bacteria and host.²⁷ In this study, dietary nitrate consumption elevated both saliva nitrate and nitrite concentrations with a ratio of saliva nitrate:nitrite of ≈3. This is similar to the ratios identified in healthy volunteers after single dose nitrate supplementation,^28,29 suggesting that the enterosalivary circulation is intact and functioning normally in patients with hypertension. In these processes, dietary (or inorganic) nitrate can be thought of as a prodrug for the generation of plasma nitrite. The half-life of nitrite in the circulation after a single oral dosing has been estimated to be ≈30 minutes.³⁰ In contrast, nitrate has a half-life of ≈6 hours after oral dosing of inorganic nitrate.³¹ The recirculation of nitrate, however, also extends the apparent half-life of nitrate-derived nitrite, with plasma nitrite peaking at ≈3 hours after inorganic nitrate ingestion and remaining elevated for at least a further 3 hours in both healthy and hypertensive subjects.^17,19,23

In healthy volunteers, it has been shown that at least some of the nitrite in the swallowed saliva enters the circulation where it is chemically reduced by the action of ≥1 mammalian nitrite reductases to generate vasodilator NO.³² That this occurs in patients with hypertension also is demonstrated biochemically in this study by increases in the downstream secondary messenger cGMP, that is elevated between 3–24 hours after dietary (or inorganic) nitrate ingestion,^19,23 confirming production of bioactive NO. Peak BP reductions coincide with these peak plasma nitrite elevation and plasma cGMP elevation, at ≈3 hours following dietary (or inorganic) nitrate ingestion,^17,19,23 and in this study, the increases in plasma nitrite concentration after dietary nitrate were associated with BP reductions measured both in- and out-of-clinic, providing further evidence that nitrite reduction to NO underlies the BP-lowering effects seen with dietary nitrate consumption.

Out-of-clinic BP measurements (ie, home and ABP) are recognized to be more predictive of target organ damage and mortality in population cohort studies^33,34 and underlies the use of all 3 methods in this study. Interestingly, the magnitude of BP reduction after dietary nitrate consumption was similar across all 3 methods of measurement with clinic BP reduced by 7.7/2.4 mm Hg, 24-hour ABP by 7.7/5.2 mm Hg, and home BP by 8.1/3.8 mm Hg. Irrespective of the method of measurement, the magnitude of BP reduction is of clinical significance because it resembles the average BP reduction achieved with a single antihypertensive medication at standard dose (9.1/5.5 mm Hg).³⁵ Most pharmacological treatments for raised BP give larger maximal BP reductions dependent on higher baseline BP values,³⁵ including following acute, single inorganic, and dietary nitrate dosing in healthy and drug-naive stage 1 patients with hypertension.^17,19,23 Thus, one may postulate that the BP-lowering effects seen in these patients with mild hypertensive phenotypes may be greater in patients with more severe hypertension. However, this has not been tested in our study and is not known at this time and there is the possibility that in patients on multiple antihypertensive agents with established vascular damage from long-standing uncontrolled hypertension, the effects could be attenuated rather than amplified.

It is noteworthy that the home BP measurements over the 4 weeks demonstrate, if anything, an increasing magnitude of the effect with time with a reversal occurring only on washout. These findings are in agreement with primate studies demonstrating no tachyphylaxis to repeated and continuous systemic nitrite administration over 2 weeks,³⁶ confirming an absence of the development of tolerance: a characteristic that has profoundly limited the clinical utility of the organic nitrates in CVD.³⁷

Dietary nitrate consumption also improved indices of vascular function including aortic PWV, augmentation index, and FMD. PWV is the gold standard measure of arterial stiffness and is recognized as a powerful predictor of cardiovascular events.³⁸ The exact underlying mechanism for these changes in arterial stiffness is uncertain, but preclinical studies of age-induced arterial stiffening in mice suggest that systemic nitrite therapy reduces oxidative stress and advanced glycation end products that are associated with arterial stiffening.³⁹ However, it is also possible that the improvement in endothelial function measured by FMD plays a role in the improved PWV. A recent report in healthy, elderly subjects supplemented with 0.1 mmol/kg inorganic nitrate (≈6–8 mmol/d) for 4 weeks demonstrated a modest increase in FMD of 0.5%, compared with no change in the placebo group.⁴⁰ We have previously demonstrated in young healthy individuals that while a single acute dose of 8 mmol of nitrate does not alter FMD (in healthy individuals the FMD response ranges from between 7% and 14%),²⁸ that single acute doses ranging between 6 and 24 mmol do protect against transient endothelial dysfunction induced by forearm ischemia–reperfusion injury.^17,19

Finally, in all of the analyses, there were no significant changes in any functional parameter measured after consumption of the nitrate-free placebo intervention. This substantiates the proposal that inorganic nitrate is responsible for the beneficial effects seen with beetroot juice. The average intake of nitrate from regular food sources (predominantly vegetables) is 1.5 to 2 mmol daily,⁴¹ and the acceptable daily intake is set by the World Health Organization at 3.7 mg/kg per day⁴² (≈4.2 mmol daily for a 70 kg person) because of concerns over methemoglobinemia and carcinogenesis. In this study, no suggestion of significant methemoglobinemia was evident, and previous studies that were associated with micromolar plasma nitrite concentration have not demonstrated clinically significant methemoglobinemia.^20,43 Although there are well established links between preformed nitrosoamines and carcinogenesis,⁴⁴ recent US National Toxicology Program reports on 2-year rodent feeding studies with nitrite⁴⁵ and a comprehensive World Health Organization report summarizing epidemiological cohort studies in humans evaluating the risk of cancer with nitrate intake have largely assuaged concerns on carcinogenesis.⁴² Importantly, vegetable intake is associated with small reductions in cancer incidence, rather than increases.⁴⁶ Patients predisposed to oxalate renal stones may need to avoid certain high-nitrate vegetables that also contain oxalate, such as spinach and beetroot. Although study participants reported no adverse effects apart from expected discoloration of urine (beeturia) and feces from the purple betacyanin pigments in beetroot, we cannot be certain that prolonged intake of beetroot juice is reliably acceptable as a therapeutic source of dietary nitrate. There is batch-to-batch variation in nitrate content of vegetables and their juices and clinical studies using beetroot juice with varying concentrations delivering between ≈4 and 24 mmol. This natural variation could be controlled by use of anion exchange resin to provide fixed nitrate concentrations.²⁵ Our previous studies have suggested that 4 mmol is a threshold dose for BP lowering healthy volunteers, although it is not clear from our data whether this is also true for hypertensives or what dose of dietary nitrate might provide maximal BP lowering is not known.

Overall, our results presented in this study demonstrate that dietary nitrate provision to patients with hypertension provides robust BP lowering that is dependent on the conversion of nitrate to nitrite and thence NO, with no suggestion of tachyphylaxis over a 4-week period.