Beetroot Juice for Runners: Dose, Timing, and Who It Helps

Nitrate is nitrate. Beetroot dominates this research area not because it is the richest source but because, as one of the labs that tested these products put it, beets "are readily juiced." A 2026 review of nitrate-rich vegetables says it directly: "many commonly consumed vegetables, particularly leafy and stem varieties, contain comparable or higher levels of nitrates."

The umbrella review is Poon et al., Sports Medicine, 2025. It covered 20 systematic reviews, 180 primary studies and 2,672 participants. It found improved time to exhaustion (SMD 0.33; 95% CI 0.19-0.47), "with subgroup analyses indicating more pronounced improvements when a minimum dose of 6 mmoL/day (372 mg/day) and chronic (> 3 days) supplementation protocol was implemented."

The same review found effects on total distance covered (SMD 0.42), muscular endurance (SMD 0.48), peak power output (SMD 0.25) and time to peak power (SMD -0.76). Every other performance measure it looked at showed no significant improvement.

Read the quality rating alongside the result. The authors grade their own evidence base: "The AMSTAR-2 ratings of most included reviews ranged from low to critically low." That is not a critic saying it. That is the review team.

On the doses actually used: 8.4 mmol per day in the two recreational-runner studies, 8.2 mmol in the sprint study, and roughly 7 mmol in the well-trained trial that came back null.

Three limits on that table. The ranges come from a single narrative review, and other published figures for beetroot are lower, so treat them as one source's range rather than a settled value. They apply to raw, fresh produce only — cooking, storage and juicing all move the number. And they cannot be converted into powder figures by multiplying for water loss, because nobody has published the nitrate loss rate for dehydration.

The width of each range is not sloppiness. It is the real uncertainty created by soil nitrogen and storage conditions, which is also why the same product differs between batches.

Overall effect in the 80-study analysis (Senefeld et al., Medicine and Science in Sports and Exercise, 2020): d = 0.174 (95% CI 0.120-0.229, P<0.001), described by the authors as objectively small.

The effect was "not observed in studies with only women" (n = 6; d = 0.116; 95% CI, -0.126 to 0.358; P = 0.347), and "not observed in well-trained endurance athletes" — those at 65 mL/kg/min or above (n = 26; d = 0.021; 95% CI, -0.103 to 0.144; P = 0.745).

In the single trial, ten well-trained men (VO2max 66 ± 7 mL/kg/min, 10K personal bests of 36 ± 2 minutes) took roughly 7 mmol 2.5 hours before testing. Plasma nitrite rose from 61 to 473 nmol/L (P<0.001). Time to exhaustion showed no difference (P=0.5). The 10 km time trial showed no difference (P=0.6). The 11-second gap that did appear was smaller than the 51-second minimum important difference the researchers had set in advance, and only 3 of the 10 athletes showed a possibly beneficial change. The point estimates leaned slightly toward nitrate, they just meant nothing.

One caveat on that study and one release from it. It was run in simulated altitude, and its conclusion sentence ends "in normobaric hypoxia", so on its own it says nothing about sea level. But the 80-study analysis found the effect was not modulated by oxygen availability, and its 65 mL/kg/min threshold lands exactly on these athletes' 66. Taken together, the finding travels down to sea level.

The 118-study analysis of nitrate-rich and polyphenol-rich foods reached its conclusion from its own data: well-trained males — those at 65 ml/kg/min or above — "exhibited small, significant benefits following polyphenol, but not nitrate consumption." On the other null subgroup it states that "no effect of nitrate or polyphenol consumption was found in females."

Above and below those subgroup lines, individual response still varies. The IOC consensus states it outright: responses "may vary widely between individuals because of factors that include genetics, the microbiome, and habitual diet." That is why the race-week list above ends with a rehearsal step rather than a race-morning debut.

Both used 8.4 mmol per day for 3 days in recreational runners, with the final dose 2 hours before testing, and both were published the same year by the same group.

The physiology study (n=13) found real improvements: VO2max 46.6 ± 6.4 versus 45.1 ± 5.8 mL/kg/min (P=0.022), velocity at VO2max 14.5 ± 0.8 versus 13.9 ± 1.0 km/h (P=0.024), and peak velocity 15.5 ± 1.1 versus 15.2 ± 1.2 km/h (P=0.038).

The performance study (n=14) put the same protocol into a 10 km time trial. Result: 50.1 ± 5.3 minutes on beetroot versus 51.0 ± 5.1 on placebo, P=0.391. Not significant. The first 5 km were significantly faster (P=0.027), and 10 of the 14 runners finished quicker, but the finishing times as a group did not separate.

What each row rests on: the sports shot claims 400 mg (6.45 mmol) and was independently measured at 6.41 ± 0.60 mmol; the richest bulk juice measured 18.77 ± 1.59 mmol per 500 mL and the weakest 3.35 mmol per 500 mL; the powder measured 1.03 mmol per serving with no nitrate figure published by the brand; the gel is labelled 100 mg and sold as an in-race top-up, not the main dose; and raw beetroot rests on one review's range of 250-400 mg per 100 g fresh weight.

The structural problem behind all of this: there is currently no agreed unit of measure that lets a shopper compare one product against another. On the beetroot side the cause is the raw material — nitrate content tracks soil nitrogen, growing conditions and storage, so batches genuinely differ and a printed number can drift away from what is in the bottle. On the tart cherry side of the shelf the cause is different and harder: the measurement method itself is not standardised, on top of the same raw-material variation.

A 2019 study bought 45 lots of 24 different beetroot products and measured them by HPLC. Between batches of the same product, the mean coefficient of variation was 30% ± 26% (range 2-83%).

A 2025 round of independent testing by a consumer testing lab found the same shape. One serving of a beetroot product could deliver "as much as 495.7 mg to as little as 4.3 mg of nitrate - more than a 100-fold difference." Juices ran about 200 to 300 mg per 8 fl oz cup; one concentrated shot came in near 500 mg. That lab also notes that clinical trials have typically used "around 300 mg to 600 mg daily", which brackets the 372 mg floor.

Three limits on the servings table. The measured values come from samples bought between September 2017 and June 2018 by a single laboratory, so they describe those lots, not the bottle in your hand. Given a between-batch coefficient of variation that reached 83% in that dataset, no single figure here should be read as a guarantee. And no comparable retest has been published since, so there is no way to say whether the industry has improved.

The 2026 market analysis behind the three-item check covered 18 nitrate and beetroot supplements sold in Europe: 27.8% state no dose at all, only 38.9% match protocols established in the scientific literature, and 61.1% carried 300-500 mg per serving.

On the label forensics: this is not a story about brands being untrustworthy. In that 2019 dataset one product lines up exactly as it should — label claim 400 mg, independent measurement 6.41 ± 0.60 mmol, which is about 397 mg. Claim and assay agree. Two footnotes keep that clean. The lab tested it under an earlier product name, and the identification rests on the arithmetic matching (400 divided by 62 equals 6.45 mmol, the exact claimed value in the paper) rather than on any announcement from the company. And the 400 mg lives in the product description, not in the regulated nutrition panel — nitrate does not appear on the panel at all.

The opposite case is not fraud either, which is exactly what makes it instructive. HumanN does not publish a nitrate figure for SuperBeets anywhere on its product page; what occupies that space is "1 scoop = 40X the functional benefits of fresh beets." There is no milligram claim to be wrong about, any more than there is in "made from N beets". The problem is that you cannot verify what you took. That is not one brand’s quirk: measured contents across the six powders ran from 0.43 to 2.56 mmol per serving, so the weakest of them would take fourteen scoops or more to reach the research dose. The form that most often omits the number is also the form least likely to contain enough of it.

On cost, only the structural picture is stable enough to be worth printing. Powders were often the most cost-effective route to a given amount of nitrate in the 2025 testing round, despite being the form most likely to omit the number. Among supermarket juices, the volume needed for one dose spans that 5.6-fold range, and the cost gap runs with it.

One honest caveat on the yardstick itself: nitrate is the component that can be measured and compared across products, but whether it accounts for the whole of what beetroot juice does is not settled.

In a 7-day study, antibacterial chlorhexidine mouthwash cut oral nitrite production by 90% (p<0.001) and plasma nitrite by 25% (p=0.001), and pushed systolic and diastolic blood pressure up by 2 to 3.5 mmHg. The effect appeared within one day of disrupting the oral bacteria. The authors concluded that "the recycling of endogenous nitrate by oral bacteria plays an important role in determination of plasma nitrite levels."

Countermovement jump and maximal voluntary contraction both dropped after the race, with no difference between groups (P>0.05). Muscle soreness: no difference (P=0.694). Interleukin-6: no interaction (P=0.213). Leukocytes, creatine kinase, AST and hs-CRP: no differences.

One limit on how far to push that. The trial answered a specific question, clearly: inflammation and muscle damage after a marathon. A 2024 review comparing beetroot juice against nitrate salts raises the separate possibility that the juice carries advantages the salts do not, recovery among them, attributed to the wider spectrum of compounds it contains — while stating that "significant limitations to published studies directly comparing BRJ and NIT make it quite challenging to draw any firm conclusions". A null result for post-marathon recovery is not a proven absence of any recovery role.

Put the other way round, 6 mmol only equals the ADI at a body weight of about 100 kg. And the trials cited earlier went higher than the protocol band: the 8.4 mmol used in the two recreational-runner studies is about 521 mg, roughly twice the ADI for a 70 kg adult.

The limit was set by the SCF in 1997 and JECFA in 2002. JECFA derived it from a no-observed-effect level of 370 mg/kg body weight per day in rats, divided by a safety factor of 100. The hazards behind it are methaemoglobinaemia in the short term and endogenous N-nitroso compound formation in the long term. It is a toxicology figure for an additive, not a dosing guideline for vegetables or for sport.

The committee that set it says as much. Data on the putative health benefits of nitrate "were not assessed, as this is not a safety issue and is therefore not within the purview" of the committee. EFSA, re-evaluating the additive in 2017, found that "it was not possible to derive a single value of the ADI from the available data", while also concluding there was "insufficient evidence to withdraw this ADI". Neither sentence is an endorsement and neither is a retraction.

Then the line that changes how the 1.4-to-1.9 multiple reads. The same EFSA opinion states that "if all sources of exposure to dietary nitrate are considered (food additive, natural presence and contamination), the ADI would be exceeded for all age groups at the mean and the highest exposure." Being over the ADI is not a place beetroot juice takes you on your own. It is roughly where people who eat vegetables already are.

A unit warning, because this is where the arithmetic usually breaks. The 3.7 mg/kg figure is expressed as nitrate ion, the same basis as the millimole conversion used throughout this page, so the two can be divided directly. The same committee also expressed the limit as 5 mg/kg on a sodium nitrate basis. Those are two ways of writing one limit, not two limits, and only the nitrate-ion version can be set against a dose worked out as millimoles times 62.

What is genuinely unresolved is the long run, and both sides of it are on the record. A systematic review on the benefits and risks of beetroot juice warns that drinking it "may easily increase nitrate intake above the acceptable daily intake, which is known to stimulate the endogenous formation of N-nitroso compounds", then adds that "compared to studies on the beneficial effects, the amount of data and literature on the negative effects of BRJ is rather limited, and should be increased in order to perform a balanced risk assessment."

Against that, JECFA's own review of the epidemiology did not find the association you would expect if dietary nitrate were driving cancer risk. A Dutch cohort followed for 6 years "found no significant association between the incidence of gastric cancer and intake of nitrate from food or drinking-water", and across the case-control studies of dietary nitrate "no association was found with oral, oesophageal, gastric or testicular cancer." The Australian Institute of Sport calls the older safety concerns "in any case disputed", while stating in the same document that "chronic use of nitrate supplements has not been well studied."

For scale in the other direction: the Australian Institute of Sport estimates average dietary nitrate intake among adults in the US, Europe and Australia at 1-2 mmol per day, about 60-120 mg, with vegetables supplying roughly 80% of it. Measured against that estimate, the research dose is roughly 3 to 8 times an ordinary day's intake.

Those two official statements do not fit together, and both are official. The European regulator, above, says average intake already exceeds the ADI once every source is counted; a national sports institute puts typical adult intake well below it. We have no basis for deciding between them, so both are given here as published — do not read them as confirming each other.

Beeturia appears in clinical references precisely because it sits on the differential for red urine. And "beet burps" is community vocabulary with no counterpart in the literature.

The official wording on both effects, from the Australian Institute of Sport: "The consumption of beetroot/juice may cause a temporary pink coloration of urine and stools. This is a harmless side-effect." And: "Beetroot juice, particularly in concentrated form and larger doses, sometimes causes mild gastrointestinal discomfort. Athletes who are inclined to use nitrate supplementation pre-competition are advised to first practice in training."

The research concluded that beeturia "does not arise from deficiencies in hepatic metabolism or renal excretion" of the pigments, and that the variation between people is driven largely by stomach acidity and gastric emptying rate, with absorption of the pigment in the colon doing the rest. It is reported more often in people who are iron deficient or have malabsorption. The proportions usually quoted are repeated from older literature rather than measured, so they are not worth attaching a number to.

Searches for reported rates of belching, bloating or gastrointestinal distress with beetroot or dietary nitrate come back empty. The only quantified signal found came from a four-arm trial in which gastrointestinal distress was greater with beetroot juice plus caffeine than with placebo (by 2.4 ± 3.6 a.u.; p < 0.05). That arm combined the two, so it says nothing about juice on its own.

If you juice raw beetroot yourself, handle it like any other raw produce. Finland recorded a cluster of gastrointestinal illness outbreaks in 2010 linked to raw grated beetroot served in institutional canteens, where no common foodborne pathogen or toxin was ever identified and all tested samples were of poor microbiological quality. That was catering-scale raw beetroot with an undetermined cause, not juice and not supplements. The transferable part is only about freshness and handling.

In a trial of 18 men with normal blood pressure, doses of 100, 250 and 500 g lowered systolic (P < 0.01) and diastolic (P < 0.001) pressure across 24 hours in a "near dose-dependent manner".

The label for sildenafil contraindicates it with "nitric oxide donors, such as organic nitrates or organic nitrites in any form", because it "was shown to potentiate the hypotensive effect of nitrates".

On kidney disease, the one relevant dataset is eight haemodialysis patients given a single 70 mL, 400 mg dose: plasma nitrate exposure, both area under the curve and peak concentration, was significantly higher than in healthy volunteers, while blood pressure and plasma potassium were unaffected and no adverse events were reported. A single dose in eight people tracked for 44 hours says nothing about repeated use.

On the nitrate-versus-nitrite warning, the AIS also notes that nitrate salts used as supplements carry a greater risk of misjudging dosages than food-form juice does.

On infants, EFSA's contaminants panel advised against feeding spinach to children with bacterial gastrointestinal infections, and notes that badly stored cooked vegetables can convert nitrate to nitrite in situ.

Each of the cautions above rests on a documented absence rather than a question nobody thought to ask: no published trials of dietary nitrate with antihypertensives, none with PDE-5 inhibitors or organic nitrate drugs, and no long-term safety data on nitrate supplementation.