Benefits
Cholesterol and blood sugar markers (small effects)
In a 24-week randomized trial in 60 adults with metabolic syndrome, part-funded by the tablet maker, hydrogen-rich water lowered total cholesterol about 18 mg/dL, triglycerides about 47 mg/dL, fasting glucose and HbA1c versus placebo, but not LDL; HDL fell slightly. A 3-month trial in 181 people found no overall difference, and a pooled analysis of 13 randomized trials in 757 overweight or obese adults found only a 6.7 mg/dL cholesterol drop and no triglyceride change, which its authors judged too small to matter for cardiovascular risk.
Selective radical scavenging (shown in cells and rats)
In cultured cells, hydrogen reduced the hydroxyl radical, the most damaging reactive oxygen species, and did not react with hydrogen peroxide, superoxide or nitric oxide, which have signalling roles. Inhaled hydrogen then reduced brain injury in a rat stroke model. This selectivity has not been demonstrated in people drinking hydrogen water.
Blood vessel function (one 2-week trial)
In a randomized trial in 68 healthy adults, drinking 500 mL of water containing 7 ppm of hydrogen (3.5 mg) improved finger-artery endothelial function on peripheral arterial tonometry, while the placebo group did not change. The hydrogen group's own reading rose about 22 percent at 24 hours and 25 percent after 2 weeks; the difference between the two groups reached significance at both points (p=0.02). That is one small, short trial of a surrogate measure, not of heart attacks or strokes.
Inflammation markers (one trial, not replicated)
In the 24-week metabolic syndrome trial, hydrogen-rich water lowered TNF-alpha, IL-6 and C-reactive protein versus placebo. That is the main human inflammation result, and it comes from a single manufacturer-funded trial in 60 people. Work on ulcerative colitis and COPD is either in animals or uses inhaled hydrogen gas in hospital, not hydrogen water by mouth.
Exercise: less perceived effort and lactate, no endurance gain
A pooled analysis of 27 publications in 597 healthy adults found no effect of hydrogen on VO2max, endurance exercise, 30-second anaerobic capacity or muscular strength. It did find small effects on lower-limb explosive power, on rating of perceived exertion and on blood lactate. A separate randomized trial in trained runners found no change in time to exhaustion.
Cognition: the one-year trial was negative
In a one-year randomized double-blind trial, 73 adults with mild cognitive impairment drank about 300 mL of hydrogen water or placebo daily. There was no difference between the groups on the ADAS-cog score, the trial's main measure. Only a subgroup carrying the APOE4 gene improved, which is an after-the-fact finding that needs its own trial.
Mechanism of action
Selective scavenging of hydroxyl radicals and peroxynitrite
In cultured cells, H2 reacted with •OH (hydroxyl radical, the most reactive ROS) and ONOO- (peroxynitrite) but not with H2O2 or O2•- (superoxide). If that selectivity holds in people it would reduce the most damaging ROS while preserving physiological ROS signalling, unlike typical antioxidants that may blunt adaptive responses. It has not been shown directly in humans.
Nrf2 pathway activation
In animal studies, hydrogen's protective effects have been linked to activation of Nrf2 (nuclear factor erythroid 2-related factor 2), a transcription factor that switches on the body's own antioxidant defences such as heme oxygenase-1 (HO-1). This is a proposed route to antioxidant effects beyond direct radical scavenging and has not been shown in people.
NF-κB inhibition and inflammasome modulation
In animal studies, hydrogen reduced NF-κB activity and pro-inflammatory cytokines (TNF-α, IL-6, IL-1β), and one rat study suggested the NLRP3 inflammasome is involved. This is the proposed basis for anti-inflammatory effects; in people, lower inflammation markers rest mainly on one 60-person trial of hydrogen water.
Mitochondrial function support
Hydrogen passes easily through cell membranes into mitochondria, and in a mouse liver-injury study, hydrogen-rich saline protected against mitochondrial dysfunction, with higher ATP levels and respiratory function. In people with mitochondrial myopathy, two small trials of hydrogen water (5 and 12 patients, the second placebo-controlled) improved blood lactate measures, indirect markers of mitochondrial function, with no objective clinical change; energy production itself has not been measured directly in people.
Cell growth signalling (Ras-ERK1/2 and Akt pathways)
In cultured vascular smooth muscle cells, hydrogen blocked activation of Ras, MEK1/2, ERK1/2 and Akt, which slowed cell growth and raised the rate of cell death; in rats, injected hydrogen-rich saline reduced artery thickening after balloon injury. That is growth suppression rather than cell protection, and it has not been studied in people.
Gut microbiome modulation
Gut bacteria make hydrogen themselves by fermentation, and in mice, hydrogen released by colonized gut bacteria eased an experimentally induced hepatitis. Whether drinking hydrogen water changes the gut microbiome in a way that matters for health has not been established.
Clinical trials
Randomized double-blinded placebo-controlled trial (LeBaron TW, Singh RB, Fatima G, Kartikey K, Sharma JP, Ostojic SM, Gvozdjakova A, Kura B, Noda M, Mojto V, Niaz MA, Slezak J, Diabetes Metab Syndr Obes 2020;13:889-896, doi:10.2147/DMSO.S240122).
60 adults of Indian ethnicity (30 men, 30 women, mean age 43) with metabolic syndrome, recruited in Moradabad, India, randomized to hydrogen-rich water or placebo for 24 weeks. Dose was one hydrogen-generating tablet in 250 mL of water three times daily, more than 5.5 mmol (about 11 mg) of H2 per day. Endpoints: body composition, blood lipids, glucose, inflammation and redox markers.
Versus placebo, hydrogen water lowered total cholesterol (about 18 mg/dL), triglycerides (about 47 mg/dL), fasting glucose, HbA1c, TNF-alpha, IL-6 and CRP (P<0.05). LDL did not reach significance (P=0.06), HDL fell slightly in the hydrogen group while it rose in placebo (P=0.01), and TBARS did not change (P=0.309). The manufacturer, HRW Natural Health Products, donated the tablets and part-funded the study. A later 3-month trial in 181 people found no overall difference, and the cholesterol drop here is nearly three times the pooled average of 13 randomized trials, which found no triglyceride change.
Laboratory study in cultured cells and in rats, not a trial in people (Ohsawa I, Ishikawa M, Takahashi K, Watanabe M, Nishimaki K, Yamagata K, Katsura K, Katayama Y, Asoh S, Ohta S. Hydrogen acts as a therapeutic antioxidant by selectively reducing cytotoxic oxygen radicals. Nat Med 2007;13(6):688-694, doi:10.1038/nm1577).
Cultured cells and rats. No human participants. The animal work used inhaled hydrogen gas, not hydrogen water by mouth.
In cultured cells, H2 reduced the hydroxyl radical (•OH) and peroxynitrite (ONOO-) but not H2O2 or superoxide (O2•-), which have signalling roles. In rats, inhaling 2 to 4 percent H2 reduced brain infarct volume after artery occlusion. This is the paper the whole selective-antioxidant idea rests on, and it is preclinical: nothing in it was measured in people, and nothing in it was given by mouth.
Narrative review article, not a study in people (Ge L, Yang M, Yang NN, Yin XX, Song WG. Molecular hydrogen: a preventive and therapeutic medical gas for various diseases. Oncotarget 2017;8(60):102653-102673, doi:10.18632/oncotarget.21130).
No participants. This is an author summary of other people's animal and human studies, most of them animal, and most using routes other than drinking water.
The review collects reports of hydrogen effects in animal models of ischemia-reperfusion injury, inflammation and metabolic disturbance, and describes proposed mechanisms including selective radical scavenging, Nrf2 activation and NF-kB inhibition. Much of the material uses inhaled gas, injected hydrogen-rich saline or hydrogen baths rather than hydrogen water by mouth. A review pools other people's results and cannot itself show that hydrogen water works.