Hydrogenology editorial study record

Effects of Molecular Hydrogen Assessed by an Animal Model and a Randomized Clinical Study on Mild Cognitive Impairment

Nishimaki K et al. · Curr Alzheimer Res. 2018;15(5):482–492.

HumanH₂-rich waterPublished 2018

Study record, not medical adviceThis page reports one source and Hydrogenology’s source-text extraction. It does not establish that molecular hydrogen is effective, safe or appropriate for any person. Human, animal and laboratory evidence must not be treated as equivalent.

Source and classification

What kind of evidence is this?

Evidence stream

Human

Reported design

Randomized double-blind placebo-controlled parallel clinical study with separate animal experiments

Research topic

Other neurological conditions

Administration classification

H₂-rich water

Study result signal

Not reported in this record.

Outcome type

Not reported in this record.

Reported in the source

Methods at a glance

Population or model

Older adults with mild cognitive impairment

Sample

81 randomized; 73 analyzed: H₂ water n=35 and placebo n=38

Duration

1 year

Intervention

One 500 mL pouch of hydrogen-rich water daily

Hydrogen form

H₂ dissolved in water — H₂ only, not Brown’s gas.

H₂ specification

More than 0.6 mM dissolved H₂ when supplied.

H₂ flow

Not applicable — oral water, not gas inhalation.

O₂ delivered with H₂

No O₂ was co-delivered as part of the intervention.

Comparator

Identically packaged placebo water, 500 mL/day

Reported, not endorsed

Outcomes and reported result

Outcomes measured

ADAS-cog and cognitive subscales, including exploratory APOE4-carrier analyses

Reported result

Overall ADAS-cog did not differ significantly between groups. Favorable findings were reported in the small APOE4-carrier subgroup; average adherence was about 64%.

Results-extraction completeness

Human dose, methods, full results, funding and conflicts checked in the open-access full text; independent appraisal remains incomplete.

A reported association, difference or mechanism is not automatically a clinical benefit. Null findings, outcome type, study design and precision all matter.

Interpretation limits

Cautions and applicability

Design and reporting cautions

Eight randomized participants were excluded after randomization, adherence averaged 64%, and the positive interpretation relies on a very small genotype subgroup despite a null overall cognitive result. Water was supplied by Blue Mercury; one author had previously advised the company.

Applies directly to

Older adults with mild cognitive impairment; the animal component is separate from the human trial.

Preliminary appraisal framework

Not reported in this record.

Preliminary risk-of-bias status

Not reported in this record.

Appraisal rationale

Not reported in this record.

No single-study GRADE certainty rating is assigned. Certainty is assessed by important outcome across a complete eligible evidence set, not by attaching a final grade to one source.

Provenance

Sources and record status

Identifiers

PMID: 29110615 · DOI: 10.2174/1567205014666171106145017

Publisher access

Open-access full text is available through PMC.

Extraction basis

Open-access PMC full text and PubMed bibliographic record.

Record revision

2026-08-10

Open bibliographic source ↗Open lawful free full text ↗Report a discrepancy