Drinking hydrogen water enhances endurance and relieves psychometric fatigue: a randomized, double-blind, placebo-controlled study

Experiment 1: untrained adults aged 40–70. Experiment 2: fitness-club members aged 20–69, including men and women.

Experiment 1: 500 mL water containing 0.8 mg/L (0.8 ppm) H₂, 30 minutes before approximately 10 minutes of submaximal cycling; all received placebo at the initial visit, followed by hydrogen or placebo two weeks later. Experiment 2: 500 mL containing 1.0 mg/L (1.0 ppm), 10 minutes before 11 minutes of cycling; the second randomized visit was one week after an initial placebo visit.

Experiment 1: 0.8 mg/L (0.8 ppm). Experiment 2: 1.0 mg/L (1.0 ppm), supported by the preparation description. A protocol sentence prints 1.0 mg/mL instead, an unresolved source unit typo.

The same water without H₂ in identical packages; each experiment used its own parallel comparison.

Estimated VO₂max from a submaximal ergometer heart-rate algorithm, not direct oxygen-consumption measurement; Borg perceived exertion and VAS psychometric fatigue.

Experiment 1: estimated VO₂max did not differ significantly. VAS fatigue change favored hydrogen (P = 0.046), as did heart-rate change (P = 0.015); overall Borg difference was nonsignificant. Higher-baseline-fatigue subgroup analyses were exploratory. Experiment 2: estimated VO₂max change favored hydrogen (P = 0.015; relative change P = 0.019), and Borg RPE change P = 0.00078; age-decade subgroup differences were nonsignificant. Exact between-arm mean differences were not printed in the checked narrative/captions and were not reconstructed from graphs. In the separate 4-volunteer breath substudy, 1.2 ppm water at 0.012 mg H₂/kg produced a breath peak at 10–20 minutes and above-baseline readings at 30–40 minutes; this does not establish universal whole-body retention.

The available main full text was reviewed for methods, results, limitations and declarations. Uncertainties, supplementary-source limits and conflicting findings are stated in this record.

Systematic adverse-event assessment was not reported; absence of a harms statement does not establish absence of harm.

The checked source is author revision R1, not the final typeset article. Two heterogeneous cohorts/protocols are not a pooled 159-person replication. Outcomes were estimated aerobic capacity and subjective fatigue, with exploratory subgroup/multiple comparisons. Five randomized Experiment 1 participants were not analyzed. A unit typo and paired-test wording for group comparisons remain concerns; formal statistical appraisal was not completed. Commercial support and patent interests were declared.

Short-term responses in the two reported adult exercise cohorts. Neither direct VO₂max, lasting training benefit nor clinical efficacy was established.

Experiment 1 was sponsored by a hydrogen-water consortium: ITO EN, Melodian, Kenkoukazoku, ABIST H&F and Medisol. Medisol and Blue Mercury donated water.

K. Tano and S. Ohta were inventors on a patent application; F. Ohta was a hydrogen-related consultant. Other authors declared no conflict.

PMID: 31251888 · DOI: 10.1139/cjpp-2019-0059

AI-assisted editorial review of the available main full text, 29 October 2026. No independent clinical or statistical assessment has been performed.

Author manuscript Methods, pages 4–8; Results 9–10; Discussion 10–13; conflicts/funding 14; Table 1, page 21; figure captions 22–23 and all figure pages.

AI-assisted editorial review of the available main full text, 29 October 2026. Complete 28-page author manuscript, revision R1 submitted 3 May 2019, in the University of Toronto repository; marked Draft, not the final typeset article. The supplementary figure is included in this manuscript.

UMIN000029062 is article-reported; prospective registration timing was not independently confirmed. A 2021 discussion (DOI 10.1139/cjpp-2021-0031) raises interpretation concerns; its full text was not independently reviewed.