Hyperbaric oxygen · human studies first, animal studies apart
Plate 41
Pressure record
What a trial ran
73
Sources on the page
18
Goals in people
On this page. Counts, not a grade and not a ranking.
5 goals carry the stronger record (decompression sickness and gas embolism, carbon monoxide poisoning, diabetic foot ulcers, late radiation injury, sudden sensorineural hearing loss). The rule is printed above the goals, and the count it read is printed under each one.
Read this first
Educational use only. This page is not medical advice, not a referral, and not a treatment plan. Pressures and session counts are what a trial ran, not a protocol for you. The indications listed by medical societies are decided by a hyperbaric physician for a named patient. Aaron Cuha is a coach, not a doctor.
Educational use only. This database summarizes published research and is not medical advice. Where a dose appears it is a record of what a trial or clinic ran, or what people describe doing, with its source attached. Nothing here is a recommendation for personal use.
Hyperbaric oxygen therapy means breathing close to 100 percent oxygen at a pressure above sea level, in a chamber. Pressure is counted in atmospheres absolute (ATA), where 1.0 is the air around you. The Undersea and Hyperbaric Medical Society says approved treatments are usually delivered at 1.9 to 3.0 ATA. The clinical trials on this page ran from 1.5 to 2.5 ATA, and the mild-chamber studies from 1.3 to 1.4.
It is used as standard care in hospitals for a defined list. The society's list includes decompression sickness, carbon monoxide poisoning, diabetic wounds, delayed radiation injury, necrotizing fasciitis, gas gangrene and refractory osteomyelitis. A European consensus graded each indication by evidence and in 2016 added explicit 'not indicated' recommendations for the first time.
Outside that list, chambers are sold for ageing, recovery, brain injury, long COVID and wellness, often as soft or 'mild' chambers at 1.3 to 1.5 ATA. The society calls treatment below 1.5 ATA unproven, and says most people in those chambers breathe well under 95 percent oxygen through masks that do not seal.
The clinical rationale is dissolved oxygen: pressure raises the oxygen carried in plasma on top of what haemoglobin carries, which the society's indications page sets out as 0.003 ml per ml of plasma for each mmHg of oxygen. The newer longevity story is different, and it is a hypothesis. The human outcomes are in the goals section; this section says what the mechanism papers propose and where they disagree.
The longevity hypothesis
The hyperoxic-hypoxic paradox is the idea that cells read a swing in oxygen, not its level, so repeated high-oxygen sessions with air breaks switch on responses normally caused by low oxygen: new vessels and stem cell activity. It comes from the research group behind the healthy-ageing trials, whose authors declare employment with or shares in AVIV Scientific. It is a review, not a measurement in people.
Stem cells in the blood
In people, one 2-hour session at 2.0 ATA doubled circulating CD34+ progenitor cells, and 20 treatments raised them eightfold, without a rise in total white cells. Cells in the bloodstream are a marker; the paper did not measure what those cells went on to repair.
Dose and stem cells
The 20-patient study found post-treatment CD34+ CD45-dim counts were always about twice the pre-treatment value, and higher at 2.5 than at 2.0 ATA after the 10th and 20th sessions. It was funded by the US NIH and the Office of Naval Research.
Where the mechanism story is weak
In 10 patients treated for radiation injury, a Danish group found a weak, non-significant rise after one session, a minor rise after 15, and none after 30 in the six who finished. The authors say the study was too small to settle it either way.
Blinded trials give the control arm air at a light pressure so people cannot tell which they had. A French validation trial found 1.3 ATA with added cues made 86.5 percent of volunteers believe they had been at 2.0 ATA. That is the pressure a soft chamber runs at, which is one reason sham arms in brain injury trials could improve as much as treatment arms.
Every number in this section is a record of what a trial ran, with that trial named beside it. None of it is a session plan for a reader, and pressure is the variable that separates a hospital treatment from a wellness chamber.
Each row is a record from a paper. It is not a protocol for you. Where an abstract does not state the pressure, the row does not guess one.
| Goal | Who | What the trial ran | Source |
|---|---|---|---|
| Carbon monoxide poisoning | People with symptomatic poisoning, 76 per arm when the trial stopped | Three chamber sessions within 24 hours. The abstract does not state the pressure. | PMID 12362006 |
| Diabetic foot ulcer | 94 patients with ulcers present over three months | 40 sessions of 85 minutes, five days a week for eight weeks, against hyperbaric air. | PMID 20427683 |
| Radiation cystitis | 79 analysed, pelvic radiotherapy at least six months before | 30 to 40 sessions of 100 percent oxygen at 240 to 250 kPa, 80 to 90 minutes daily. | PMID 31537473 |
| Post-concussion symptoms | 72 service members, at least four months after injury | 40 sessions at 1.5 ATA. Sham: room air at 1.2 ATA. Third arm: no chamber. | PMID 25401463 |
| Combat PTSD | 63 male veterans randomised, 56 completed | 60 daily sessions of 90 minutes at 2 ATA, 5-minute air breaks every 20 minutes. Sham: air at 1.02 ATA. | PMID 39566051 |
| Fitness in older adults | 63 sedentary adults over 64 | 60 daily sessions over 12 weeks, 100 percent oxygen at 2 ATA for 90 minutes with 5-minute air breaks every 20 minutes. Control: no intervention. | PMID 38961397 |
| Long COVID | 80 patients aged 18 to 60 | 10 sessions at 2.4 bar for 90 minutes. Placebo: air at 1.34 to 1.2 bar. | PMID 40228859 |
| Autism, mild chamber | 62 children aged 2 to 7 | 40 one-hour sessions at 1.3 atm and 24 percent oxygen. Control: 1.03 atm air. | PMID 19284641 |
Stronger records first. Nulls and marketing-ahead goals stay on the same page. Every card is people only; animal work is in its own section below. Nothing here is ranked against the others, and no clinic or device is recommended.
How the tier is decided, so you can check it against the counts printed under each one. Stronger: a clinical guideline or a medical society's indication list recommends it in a named setting, or a pooled analysis of randomised trials favours hyperbaric oxygen and the source rated the certainty moderate. Thin: the pooled result was null or went against hyperbaric oxygen for the goal as named, the source rated it very low, or nothing was pooled at all. Mixed: everything else, including a result that moved one measure and not another.
Stronger human record
a guideline or society indication list names it in a named setting; nothing pooled; certainty not graded by the source; no pooled result
Who was studied
Divers and others with bubbles in tissue or blood after a pressure change. The society lists decompression sickness and air or gas embolism, and the European consensus graded its indications in 2016. Both are professional bodies; neither is a trial.
The society calls hyperbaric oxygen a standard of care for decompression sickness. No randomised trial on this page compares it with no recompression. The European consensus has an evidence level for exactly this case: conditions that do not allow proper randomised trials but have ample international expert consensus.
Setting in the evidence
Hospital multiplace or monoplace chamber, under a hyperbaric physician.
What this does not show
Anything about chambers outside a medical facility, or about people without decompression illness. This is the clearest indication on the page and the least relevant to wellness use.
Stronger human record
a guideline or society indication list names it in a named setting; 6 randomised trials pooled, 1,361 people; certainty not graded by the source; result null
Who was studied
People poisoned by carbon monoxide; the French trials enrolled poisonings at home. The Cochrane review pooled six trials and 1,361 participants, and one of its authors formerly directed a hyperbaric unit, as declared. The positive 2002 double-blind trial, two French trials and an Australian sham trial are cited directly.
In the 2002 double-blind trial, three sessions within 24 hours cut cognitive problems at six weeks to 25.0 percent against 46.1 percent. The Australian sham-controlled trial found the opposite: more extra treatments and worse learning scores with hyperbaric oxygen. In the French trials, one session added nothing after brief loss of consciousness (58 against 61 percent recovered), and two sessions did worse than one after coma (47 against 68 percent). Pooled, the odds ratio for neurological deficits was 0.78 (0.54 to 1.12), not significant.
Setting in the evidence
Hospital chamber, starting within hours of poisoning.
What this does not show
That the Cochrane question is settled. Its authors write that existing trials do not establish whether hyperbaric oxygen reduces neurological harm, and that both positive trials were stopped early for benefit. The society listing reflects clinical practice, not a pooled trial win.
Stronger human record
a guideline or society indication list names it in a named setting; 10 randomised trials pooled, 531 people; certainty not graded by the source; result mixed
Who was studied
People with diabetes and chronic foot ulcers on top of standard wound care. The Cochrane review covered ten trials and 531 people with these ulcers. The diabetic foot guideline group gives conditional support where standard care has failed. One positive Swedish trial and two null trials are cited directly.
Pooled, healing at six weeks was more likely (risk ratio 2.35) but the difference was not evident at one year, and major amputation did not differ significantly. The Swedish double-blind trial healed 52 percent against 29 percent at a year. The Canadian double-blind trial found no difference in amputation criteria (11 of 49 against 13 of 54) or healing (20 against 22 percent), and the Dutch ischaemic-ulcer trial found no significant difference in limb salvage or healing. A US cohort of 6,259 found treated patients healed less often, though the treated were chosen, not randomised.
Setting in the evidence
Hospital or wound-centre chamber as an add-on to wound care.
What this does not show
That a given ulcer will respond. The Cochrane authors were not confident in the results, the trials disagree, and the guideline support is conditional on standard care having failed and resources being available.
Stronger human record
a guideline or society indication list names it in a named setting; 18 randomised trials pooled, 1,071 people; certainty low; result mixed
Who was studied
People months or years after radiotherapy, with injury to the bladder, bowel, jaw or other tissue. The 2023 Cochrane update covered 18 studies and 1,071 participants, and several of its authors are hyperbaric physicians or oncologists who refer for it. The trials cited directly were funded by cancer charities and public health research funds.
Pooled, complete or significant improvement was more likely (risk ratio 1.39, low certainty) and head and neck wound breakdown less likely (0.24, low certainty). In the Nordic bladder trial, urinary symptoms improved 10.1 points more than with standard care. The UK sham-controlled bowel trial found no difference at 12 months. The UK jaw-extraction trial found osteoradionecrosis in 6.4 against 5.7 percent, and two Danish and Dutch trials combined found 70 against 51 percent healing, not significant.
Setting in the evidence
Hospital chamber, typically 30 to 40 sessions at 2.4 ATA or 240 to 250 kPa.
What this does not show
That every radiation injury responds. The positive bladder trial had no sham and no masking; the sham-controlled bowel trial was null. The Cochrane review rates visual acuity loss with treatment at high certainty (risk ratio 4.03).
Stronger human record
a guideline or society indication list names it in a named setting; 3 randomised trials pooled, 150 people; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
Adults with sudden hearing loss of unknown cause. The 2022 meta-analysis pooled three trials, 88 people given hyperbaric oxygen and 62 controls. The Cochrane review pooled seven smaller, poorer trials with 392 participants. The US guideline lists it as an option.
With hyperbaric oxygen added to medical treatment, mean hearing gain was 10.3 dB greater and the odds of recovery 4.3 times higher. The Cochrane review found 15.6 dB more improvement and more people reaching a 25 percent gain, but not a 50 percent gain, and no benefit when people presented at six months. The guideline says clinicians may offer it with steroids within 2 weeks of onset, or as salvage within 1 month.
Setting in the evidence
Hospital chamber, combined with steroids, early.
What this does not show
Benefit for tinnitus alone, for hearing loss present for months, or as a replacement for steroids. The Cochrane authors call the clinical significance unclear.
Mixed, local, or early
4 studies pooled, 254 people; certainty not graded by the source; result mixed
Who was studied
Mostly service members and adults with symptoms months to years after concussion. The pooled analysis combined four US Department of Defense trials with 254 participants. Four individual trials are cited: two US military sham trials, an Israeli crossover with no sham, and a 2025 US sham trial.
In HOPPS, both chamber arms improved against no chamber, and hyperbaric oxygen at 1.5 ATA did not beat sham at 1.2 ATA. In the Air Force trial, 2.4 ATA did no better than sham at 1.3 ATA and both arms improved. The 2025 trial found symptom scores fell 10.6 points against 3.6 with sham (difference 7.0). The Israeli trial found gains after treatment and none in a no-treatment period. Pooled, symptoms trended better (-2.3, p=0.18) and verbal memory improved.
Setting in the evidence
Clinical chambers at 1.5 to 2.4 ATA, 30 to 40 sessions.
What this does not show
A settled answer. When the comparison is a pressurised sham, two of three trials were null. The pooled authors call for a definitive trial.
Mixed, local, or early
4 studies pooled, 254 people; certainty not graded by the source; result mixed
Who was studied
Male veterans with combat PTSD and no brain injury history in one trial, 63 randomised and 56 completed, against a sham at 1.02 ATA. The pooled data from four Department of Defense trials looked at PTSD alongside concussion in 254 participants.
With 60 sessions at 2 ATA, CAPS-5 PTSD scores fell from 42.57 to 25.8, while the sham arm's scores rose from 45.11 to 49.22 at follow-up; depression scores also improved. In the pooled military data, the PTSD checklist trended 2.7 points better, not significant (p=0.09).
Setting in the evidence
Clinical chamber, 60 sessions of 90 minutes at 2 ATA with air breaks.
What this does not show
Replication by an independent group, or effects in women or civilians, who were not in the positive trial.
Mixed, local, or early
9 studies, randomised and not, pooled, 288 people; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
Mostly women with fibromyalgia. A 2023 meta-analysis pooled nine studies and 288 patients, randomised and not. The Israeli crossover had 60 women; the Brazilian HOTFy trial had 56, and its lead author holds stock in and directs a hyperbaric centre.
Pooled, pain was lower than with control (standardised mean difference -1.56). In the Israeli trial, symptoms improved after 40 sessions at 2 ATA and not during a no-treatment period. In HOTFy, over eight weeks the questionnaire score fell 31.1 against 14.4 percent with standard care, and some gains regressed after sessions ended. Adverse events reached 23.8 percent in the pooled studies.
Setting in the evidence
Clinical chambers at 2.0 to 2.3 ATA, around 40 sessions.
What this does not show
Benefit beyond the attention of daily chamber sessions, because no trial here used a sham.
Mixed, local, or early
10 randomised trials pooled, 299 people; certainty not graded by the source; result mixed
Who was studied
Healthy volunteers and athletes after damaging exercise. A 2026 meta-analysis pooled 10 trials and 299 people; an athletes' review pooled six studies; the 2005 Cochrane review covered nine small trials and 219 people.
Markers of muscle injury recovered faster, but soreness did not (interval -0.91 to 0.48). Blood lactate after recovery was lower in athletes. In the Cochrane review, pain at 48 hours was 0.88 points higher on a 10-point scale when sessions started immediately. A 24-man trial at 2.5 ATA found no difference in strength recovery against normoxic gas at the same pressure.
Setting in the evidence
Clinical chambers, at pressures from below to above 2.0 ATA in the pooled trials.
What this does not show
Faster return to training, better performance in the next session, or anything about injuries rather than soreness.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result mixed
Who was studied
Adults with symptoms months after COVID-19, in three double-blind sham-controlled trials of 73, 80 and 101 patients. The positive trial declares two authors employed by and the senior author holding shares in AVIV Scientific. The two null trials were run in Sweden and Belgium.
The Israeli trial found cognition, energy, sleep and pain improved more than with sham after 40 sessions. The Swedish trial found no difference in physical functioning or role limitation after 10 sessions; both arms improved. The Belgian trial compared four doses of oxygen and pressure and found no difference between any of them. A 2024 BMJ living review of 24 trials found no compelling evidence for hyperbaric oxygen.
Setting in the evidence
Clinical chambers, 10 to 40 sessions; the two null trials ran 2.4 and 2.5 ATA, and the positive trial's abstract does not state its pressure.
What this does not show
That longer courses are null; the null trials ran 10 sessions against the positive trial's 40. It also does not show the positive trial was wrong, only that it has not replicated.
Thin, null, or marketing ahead
11 randomised trials pooled, 705 people; certainty not graded by the source; result null
Who was studied
Adults in the days after an ischaemic stroke. The 2014 Cochrane review covered 11 trials and 705 participants and was funded by the Chief Scientist Office.
Death at six months did not differ (risk ratio 0.97). Four of 14 disability measures improved. The authors found no good evidence that hyperbaric oxygen improves clinical outcomes in acute stroke.
Setting in the evidence
Hospital chamber in the acute phase.
What this does not show
Anything about stroke survivors treated years later, which is a different question and the next card.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
74 patients enrolled and 15 excluded, 6 to 36 months after a stroke with a lasting motor problem, in an Israeli crossover. The comparison was two months without treatment. Funded by the hospital, a family foundation, a university chair and a private donor.
Neurological function and quality of life improved after 40 sessions at 2 ATA and not during the control period, and brain imaging changes tracked the improvements.
Setting in the evidence
Clinical chamber, 40 sessions at 2 ATA.
What this does not show
That the gains exceed what attention and a daily routine produce, because there was no sham, or that an independent group can repeat them.
Thin, null, or marketing ahead
nothing pooled; certainty low; result null
Who was studied
Children with autism. The Cochrane review found one eligible sham-controlled trial of 60 children, rated low quality. A US mild-chamber trial of 62 children was run by authors who earn revenue from hyperbaric treatment, with chambers supplied by a trade association.
In the Cochrane trial, there was no improvement in social interaction, behaviour, communication or cognition, and more side effects with hyperbaric oxygen. The Thai trial at 1.5 ATA against sham found both arms improved, with no significant difference between them. The US mild-chamber trial at 1.3 atm found more clinician-rated improvement (30 against 8 percent much improved).
Setting in the evidence
Mild chambers at 1.3 atm, and clinical chambers at 1.5 ATA.
What this does not show
Benefit from the mild chambers sold for autism; the only positive trial was run by people who sell the treatment.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
63 healthy adults over 64, randomised to three months of sessions or to no intervention, with no sham. Six authors declare employment with AVIV Scientific and one holds shares. A separate imaging study of 20 healthy young adults used a placebo condition.
Global cognition improved more than in the no-intervention group, mainly attention (net effect 0.745) and processing speed (0.788), with changes in brain blood flow. In the young adults, spatial working memory and memory scores rose.
Setting in the evidence
Clinical chamber over three months. The abstract does not state the pressure; the same group's fitness trial ran 60 sessions at 2 ATA with air breaks.
What this does not show
Fewer dementia diagnoses, slower decline over years, or a result that holds against a sham or at an independent centre.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
35 healthy adults aged 64 and over for the blood study, and 13 men of about 68 from the same cohort for skin biopsies. Neither had a separate control group; the skin study compared each man before and after. Both were funded by the Sagol network, and their authors declare employment with or shares in AVIV Scientific.
Telomere length in several blood cell types rose by over 20 percent and senescent T helper cells fell 37.30 percent after 60 sessions. In skin, collagen density, elastic fibre length and blood vessels rose and senescent cells fell.
Setting in the evidence
Clinical chamber, 60 daily sessions. The abstracts do not state the pressure.
What this does not show
Longer life, slower ageing, or better health. Telomere length in isolated blood cells is a marker, not an outcome, and these are single-group studies from one group.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result favours hyperbaric oxygen
Who was studied
63 sedentary adults over 64 against no intervention, and 37 master athletes aged 40 to 50 against a sham at 1.02 ATA, of whom 16 and 15 were analysed. Both trials declare authors employed by AVIV Scientific and a shareholder and co-founder.
In the older adults, VO2max per kilogram rose 1.91 ml/kg/min and cardiac blood flow rose. In the athletes, VO2max rose more than with sham (effect size 0.989), with gains in mitochondrial respiration.
Setting in the evidence
Clinical chamber, 40 to 60 sessions at 2 ATA.
What this does not show
Race results, independent replication, or that the gain outlasts the sessions.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result null
Who was studied
One trial of 32 people with ankle sprains and one of a knee ligament injury, both against sham, within the 2005 Cochrane review. No trial on this page tests hyperbaric oxygen after a surgical tendon repair in people.
Time to recovery, function, pain and swelling did not differ in the ankle sprain trial, and knee function did not differ in the ligament trial. The animal tendon and ligament studies, which were mostly positive, are in their own section below.
Setting in the evidence
Clinical chamber after acute soft tissue injury.
What this does not show
That hyperbaric oxygen helps or harms a repaired tendon in a person. That trial has not been done, so the gap is an absence, not a null.
Thin, null, or marketing ahead
nothing pooled; certainty not graded by the source; result mixed
Who was studied
16 young women for one session at 1.4 ATA, 10 men after hard cycling at 1.3 ATA, and 60 young men living at 4300 m for five days. The society defines mild hyperbaric treatment as below 1.5 ATA.
Natural killer cells and parasympathetic activity rose after one session; mood scores improved after exercise; sleep quality and working memory improved at altitude. None measured a health outcome. The society states mild hyperbaric treatment is unproven.
Setting in the evidence
Soft or mild chambers at 1.3 to 1.4 ATA, 31 to 40 percent oxygen where stated.
What this does not show
That a soft chamber delivers what a clinical chamber does. The pressure and oxygen fraction are both lower, and the trials of clinical treatment do not transfer.
Animal results are kept here, away from the cards about people, because they are where most of the positive tendon, ageing and brain findings live. They show what hyperbaric oxygen can do in a rat or a rabbit with a controlled injury. They do not show it in a person.
Tendon and ligament, review of 37 studies
A 2026 systematic review of 37 animal studies, 2141 animals, found histology improved more often than biomechanics, with tendon and ligament reconstruction models among the more promising. Risk of bias was high or unclear in most studies, and protocols ranged from 1 to 56 sessions.
Rat ligament healing
Removing a 2 mm segment of a rat's knee ligament and treating at 2.5 ATA for 2 hours, 5 days a week, gave more scar tissue, more type I procollagen gene expression, and greater load and stiffness at 14 days than room air.
Rat ligament, three exposures
In 44 rats, 1.5 ATA for 30 minutes, 2 ATA for 30 minutes and 2 ATA for 60 minutes all helped ligament healing at 14 days; 2 ATA for 60 minutes gave the most collagen synthesis.
Rat Achilles repair
In 56 rats with a sutured Achilles tendon, hyperbaric groups had more fibrosis and differed in force and stiffness on day 11. The abstract does not give the direction of each measure; the authors concluded the effect was positive.
Rabbit ACL graft
After ACL reconstruction, 2.5 ATA for 2 hours a day for the first 5 days gave better graft maturation, smaller bone tunnels, and greater load to failure and stiffness at 12 weeks.
Rabbit rotator cuff
After a full-thickness tear, 2 ATA for 2 hours a day for 5 days gave more vessels and collagen at the tendon-bone interface and higher tensile strength, but no significant effect on biomechanical stability.
Aged rats, brain
In rats aged chemically with D-galactose, with or without a high-fat diet, 14 days at 2 ATA attenuated insulin resistance, cognitive impairment and hippocampal changes against sham.
Aged rats, bone
In chemically aged rats, 14 days of sessions fully restored bone microarchitecture in lean rats and partly in obese rats, and shifted the senescence genes p16 and p21.
Alzheimer model mice
In 5xFAD mice, cognitive flexibility and associative learning improved in females but not males; balance improved, while gait and anxiety-like behaviour did not.
Aged female mice, fertility
Ten days of hyperbaric oxygen improved serum AMH, oocyte maturation, fertilisation and blastocyst formation against untreated aged mice.
Animal stroke, 51 studies
Pooled across 51 animal stroke studies, infarcts were 32 percent smaller and mortality 18.4 against 26.7 percent, largest when given immediately at 2.0 ATA. The human Cochrane review of acute stroke, above, found no good evidence of benefit.
Mouse stem cell mobilisation
In mice, sessions raised circulating stem cell factor by 50 percent and stem-cell-marked cells 3.4-fold, and the rise did not occur in mice lacking endothelial nitric oxide synthase.
Three kinds of evidence sit under one name. Society indications and European consensus for a defined list of conditions; randomised trials, some against a pressurised sham and some against nothing; and single-group studies of markers in healthy people. They are not interchangeable, and a clinic that cites the first to sell the third is borrowing.
The sham matters more here than almost anywhere. When trials gave the control arm a light pressurisation, the chamber itself often helped: in HOPPS both chamber arms beat no chamber and did not differ, in the Air Force trial both arms improved, and in the Belgian long COVID trial nothing separated four doses. A trial against no intervention cannot tell oxygen from the routine and attention of daily sessions.
Disputes stay on the page as disputes. Carbon monoxide has a positive double-blind trial and a negative sham trial. Diabetic foot ulcers have a positive Swedish trial and null Canadian and Dutch trials. Long COVID has one positive trial and two null ones. None is deleted to make a cleaner story either way.
Money is named both directions. The healthy-ageing, VO2max, telomere, skin and the positive long COVID trial all declare authors employed by or holding shares in AVIV Scientific LTD. The positive autism trial was run by authors who earn revenue from the treatment. The null military brain injury trials were government funded, and the HOT2 and HOPON trials were funded by Cancer Research UK. Declared is not disqualified, and undeclared would be worse.
The blocks above are the studies. The block below is a judgment about what to do with them. It has to say where it goes past the measurements.
Aaron Cuha’s position
This is opinion, not a finding. Everything else on this page is the evidence.
Hyperbaric oxygen is real medicine for a short list of conditions and an unproven product everywhere else. The healthy-ageing record comes from one research group with a declared commercial stake, and only one of its trials used a sham.
The reasoning
Where this goes beyond the evidence
Calling it unproven for longevity is a judgment about absence. No sham-controlled trial at an independent centre has counted years of life, new disease or lasting function in healthy people, and absence of a trial is not proof of no effect. The intermittent-hyperoxia mechanism is plausible and the master-athlete trial did use a sham. If an independent sham-controlled replication confirms the older-adult results, this position was early. It also goes past the evidence the other way for the accepted indications, because a listing does not promise that a given patient will respond, as the diabetic foot trials show.
Who this is not for
Anyone with uncontrolled epilepsy, the one group the 62,614-session series excluded from its safety conclusion; anyone who struggles to equalise their ears, the group with the most barotrauma in the 5,962-patient series; anyone pregnant, a group none of these trials studied; and anyone who wants this page to replace a hyperbaric physician's assessment.
Aaron Cuha, who built this archive and reads the studies behind the pages. He is a coach, not a doctor, and no medical credential is claimed for him anywhere on this site. More about who writes this
Instead of an emergency
Decompression illness and carbon monoxide poisoning are emergencies treated in hospital chambers. A wellness chamber is not that service.
Uncontrolled epilepsy
The 62,614-session series found oxygen-toxicity seizures extremely rare and still made uncontrolled epilepsy its one exception.
Ears and sinuses that do not equalise
In 5,962 patients, middle ear barotrauma was recorded at a rate of 9.2 percent and was more common in people with rhinitis or difficulty equalising.
Pregnancy
None of the trials on this page studied pregnant women, so this page has no evidence to offer for them.
Anyone wanting a personal protocol
Pressures, oxygen fractions and session counts on this page are records of what a trial ran. They are not instructions for your body.
At one Israeli centre, seizures occurred in 0.011 percent of 62,614 sessions (1 in 8,945), and one was clearly from oxygen toxicity (1 in 62,614). An Australasian audit of 96,670 treatments found 2.4 seizures per 10,000 at 243 kPa, which is 1 in 222 patients over a course. Per-patient risk is higher than per-session risk.
Middle ear barotrauma is the commonest side effect. In 5,962 patients it was recorded at a rate of 9.2 percent, mostly minor, and stopped therapy for 1.49 percent of all patients. In the Cochrane radiation review, barotrauma was 9.08 times more likely when the control arm had no sham pressurisation.
Short-sightedness can develop during a course. After 21 sessions at 240 kPa the shift was about -0.55 dioptre by mask and -1.10 by hood, back to baseline within 6 to 10 weeks. The Cochrane radiation review rates reduced visual acuity with treatment at high certainty (risk ratio 4.03), usually temporary.
The society warns that chambers outside medical facilities, often in commercial premises, frequently operate without fire safety and chamber construction standards, putting users at risk of serious injury and death.
In the French carbon monoxide trial, comatose patients given two sessions recovered less often than those given one (47 against 68 percent). The Australian trial found worse learning scores with hyperbaric oxygen. Side effects were 23.8 percent in pooled fibromyalgia studies.
A 2017 review written for informed consent catalogues the side effects and notes that protocols vary widely between institutions. In two military trials of 4,245 sessions, chamber-related adverse events were 1.1 and 2.2 percent, mostly minor barotrauma.
This site teaches a blood panel, protein and training before gadgets. Hyperbaric oxygen does not move that order. For a listed medical condition it is a hospital decision; for longevity it is an unproven add-on.
Disclosed both ways: the owner of this archive uses chamber sessions in his own rehabilitation after a tendon repair, and publishes that plan, with its own caveats, at /my-protocols. That is a personal choice recorded there. It is not a finding on this page, and the tendon card above says plainly that no human trial covers it.
Nothing on this page is for sale. No clinic or chamber maker wrote it, and none is named or ranked.
Four setups sold under one name. No brand or clinic is named, and none is ranked.
A rigid room-sized chamber pressurised with air, where patients breathe oxygen by mask or hood. The Swedish ulcer trial and the Australian carbon monoxide trial were run in multiplace chambers.
Evidence match. The setting for the society-listed conditions, at the 1.9 to 3.0 ATA the society describes.
Mismatch. Not a wellness service. Access usually follows a referral for a listed condition.
A rigid single-person tube, often filled with oxygen. Used in hospitals and in some private clinics.
Evidence match. Can be run at clinical pressures, so it can match a trial row if the pressure, oxygen and supervision do.
Mismatch. A hard chamber in a clinic is only clinical if the pressure, oxygen and supervision are.
An inflatable chamber at about 1.3 to 1.5 ATA, often with an oxygen concentrator feeding a loose mask. Sold for home and wellness use.
Evidence match. The setting of the mild-chamber studies on this page, at 1.3 to 1.4 ATA and 24 to 40 percent oxygen.
Mismatch. The society calls treatment below 1.5 ATA unproven. Clinical trial results at 2.0 to 2.5 ATA do not transfer to it.
A course of around 60 sessions at 2 ATA with scheduled air breaks, sold for ageing, cognition or performance.
Evidence match. The protocol used in the healthy-ageing trials on this page.
Mismatch. Those trials come from one group with a declared commercial stake, mostly without a sham, and none measured disease or lifespan.
Floor habits and the neighbouring pages. None of these is a cart.
Oxygen
Extra oxygen at normal pressure, during training. A thinner record.
Neighbour
Another recovery device, held to the same rule.
Floor
What has been measured about strength. A chamber does not replace it.
Floor
A marker, or a panel you still need. Not a diagnosis.
Disclosure
Where chamber sessions appear in a personal plan, labelled as one.
Order
Measure first. Floor habits before gadgets.
Map
Every pillar, this one among them, held to the same rule.
Knowledge stays free. No seller wrote this page. Every source below opens the record that was checked before the sentence was written, with what it found and who funded it.
Review arguing that fluctuations in oxygen, rather than its absolute level, are read by cells as a lack of oxygen, so repeated intermittent hyperoxia can switch on responses usually caused by hypoxia. Hypothesis literature. One author works for AVIV Scientific LTD and the other is a shareholder, as the paper declares.
In people, circulating CD34+ cells doubled after one 2-hour exposure at 2.0 ATA and rose eightfold over 20 treatments without a rise in total white cells. The same paper's mouse work is reported separately below. Funded by the US NIH.
20 consecutive patients at two centres. CD34+ CD45-dim cells were always 2-fold higher after treatment than before, at both 2.0 and 2.5 ATA, and higher at 2.5 ATA after the 10th and 20th sessions. Funded by the US NIH and the Office of Naval Research.
10 patients with radiation injury. A weak, non-significant rise after one session, a minor rise after 15 sessions, and no effect after 30 sessions in the 6 who finished. Too few patients to prove or disprove the effect, the authors say. No conflict statement in the PubMed record.
60 volunteers, then 25 more. A sham compression to 132 kPa (1.3 ATA) with forced ventilation, heating and a five-minute compression made 86.5 percent believe they had been at 203 kPa. Conflict of interest and funding: nil.
Double-blind randomised trial stopped at the third of four interim analyses, 76 per group. Three hyperbaric sessions within 24 hours against one normobaric oxygen session and two room-air sessions. Cognitive sequelae at six weeks in 25.0 percent (19 of 76) against 46.1 percent (35 of 76), adjusted odds ratio 0.45. Cerebellar dysfunction before treatment was more common in the normobaric group (15 against 4 percent). PubMed tags it as non-government funded; no conflict statement in the record.
HODFU: single-centre, double-blind, placebo-controlled, 94 patients with Wagner grade 2 to 4 ulcers present over three months. 40 sessions of 85 minutes against hyperbaric air. Index ulcer healed at one year in 52 percent (25 of 48) against 29 percent (12 of 42). Funded by unrestricted grants from two Swedish foundations and Lund University; no conflicts reported.
Open randomised phase 2 to 3 at five Nordic hospitals, 87 enrolled and 79 analysed. 30 to 40 sessions at 240 to 250 kPa against standard care, no sham and no masking. Urinary symptom score improved 10.1 points more (2.2 to 18.1). 41 percent had transient grade 1 to 2 sight or hearing events. Funded by Swedish regional research and health technology funds and a cancer charity.
HOPPS: double-blind, 72 service members with symptoms at least four months after mild brain injury. 40 sessions at 1.5 ATA, 40 sham sessions of air at 1.2 ATA, or no chamber. Both chamber arms improved against no chamber; hyperbaric oxygen did not differ from sham (P = .70). PubMed tags it as US government funded; no conflict statement in the record.
Randomised, sham-controlled, 63 male veterans with combat PTSD and no brain injury history, 56 completed. 60 sessions at 2 ATA against air at 1.02 ATA. CAPS-5 fell from 42.57 to 25.8 with hyperbaric oxygen and rose from 45.11 to 49.22 at follow-up with sham. No conflict statement in the PubMed record.
63 sedentary adults over 64, 60 sessions at 2 ATA for 90 minutes with air breaks, against no intervention; no sham. VO2max per kilogram rose 1.91 ml/kg/min (net effect 0.455); cardiac blood flow rose. Four authors work for AVIV Scientific LTD and one is a shareholder and co-founder, as the paper declares.
Double-blind phase II, 80 patients, 10 sessions at 2.4 bar against air at 1.34 to 1.2 bar. No difference in physical functioning (0.63) or role limitation (2.35) at 13 weeks; both arms improved. Funded by the Swedish Research Council, Region Stockholm, the Swedish Heart-Lung Foundation and Oura Health Oy.
62 children aged 2 to 7, 40 one-hour sessions at 1.3 atm and 24 percent oxygen against 1.03 atm air. Clinician-rated overall functioning improved more; 30 against 8 percent rated much improved. Several authors derive revenue from hyperbaric treatment, and the International Hyperbarics Association supplied chambers and technician salaries, as the paper declares.
The society's indications page. Calls hyperbaric oxygen a standard of care for conditions including decompression sickness, carbon monoxide poisoning, diabetic wounds, delayed radiation injury, necrotizing fasciitis, gas gangrene and refractory osteomyelitis, with approved treatments usually at 1.9 to 3.0 ATA. States that treatment below 1.5 ATA, often called mild hyperbaric oxygen, is unproven, and that chambers outside medical facilities often operate without fire safety and construction standards. A society's page, not a study, so there is no funding to disclose.
European Committee for Hyperbaric Medicine consensus from a 2016 meeting, using GRADE and Delphi. Sorts indications into three types by strength of evidence and, for the first time, issued negative recommendations where evidence shows hyperbaric oxygen is not indicated. No conflict or funding statement in the PubMed record.
Double-blind randomised trial with sham sessions, all grades of poisoning, Melbourne. Both arms breathed high-flow oxygen between sessions. More hyperbaric patients needed extra treatments (28 against 15 percent), they did worse on a learning test, and delayed neurological sequelae occurred only in the hyperbaric group. The authors could not recommend it. No funding or conflict statement in the PubMed record.
Two randomised trials, 385 patients, Garches, France. After transient loss of consciousness (179 people), complete recovery at one month was 58 percent with one hyperbaric session and 61 percent with normobaric oxygen alone. After initial coma (206 people), it was 47 percent with two hyperbaric sessions against 68 percent with one. PubMed tags it as non-government funded; no conflict statement in the record.
Six trials, 1,361 participants. Two found fewer neurological sequelae at one month and four did not. Pooled odds ratio for neurological deficits 0.78 (0.54 to 1.12), not significant, with marked heterogeneity and design flaws in every trial; both positive trials were stopped early for benefit. Authors: existing trials do not establish whether hyperbaric oxygen reduces adverse neurological outcomes. One author formerly directed a hyperbaric unit; the other authors declare no known conflicts.
Double-blind, 107 randomised, 103 adjudicated. 30 sessions of oxygen at 244 kPa against sham air at 125 kPa. Met criteria for amputation: 11 of 49 against 13 of 54, odds ratio 0.91. Healed: 20 against 22 percent. PubMed tags it as non-government funded; no conflict statement in the record.
Multicentre randomised trial, 120 patients with diabetes and ischaemic leg ulcers, standard care with or without hyperbaric oxygen, no sham. At 12 months limb salvage in 47 against 53 patients and healed wounds 28 against 30; neither difference significant. 35 percent of the hyperbaric arm did not complete the protocol. No funding or conflict statement in the PubMed record.
Twelve trials, 577 participants; ten trials and 531 people with diabetic foot ulcers. Healing at six weeks risk ratio 2.35 (1.19 to 4.62; five trials, 205 people), not evident at one year. Major amputation risk ratio 0.36 (0.11 to 1.18), not significant. Authors not confident in the results. Funded by the Department of Health, as PubMed lists the grant; one author received consultancy payments from drug companies, the others none.
Observational cohort of 6,259 people with diabetic foot ulcers in a wound-care network, propensity-adjusted. Those given hyperbaric oxygen were less likely to heal (hazard ratio 0.68) and more likely to have an amputation (2.37). Observational, so who got treated is the main risk to the result. Funded by the US AHRQ and NIDDK; two authors consulted for the wound-care companies whose database was used, as the paper declares.
International Working Group on the Diabetic Foot guideline, GRADE method, 29 recommendations. Conditional support for hyperbaric oxygen where best standard of care has not healed the wound and resources are available. No conflict statement in the PubMed record.
18 studies, 1,071 participants. May give complete resolution or significant improvement (risk ratio 1.39, 1.02 to 1.89; 5 trials, 468 people; low certainty) and fewer wound breakdowns after head and neck surgery (0.24; low certainty); unlikely to change one-year death (0.93). Higher risk of reduced visual acuity (4.03; high certainty) and of ear barotrauma when the control arm had no sham pressurisation (9.08). Several authors are hyperbaric physicians or referring oncologists, as their conflict statements declare.
Double-blind, sham-controlled phase 3, 84 patients with bowel symptoms a year or more after pelvic radiotherapy. 40 sessions at 2.4 ATA against air at 1.3 ATA. No difference in the bowel score at 12 months (median change 4 in both arms) or in rectal bleeding. Eye refractive changes in 30 against 11 percent. Funded by Cancer Research UK and the NHS; the funder had no role.
Phase 3, 144 randomised and 100 analysed, extractions or implants in a jaw irradiated above 50 Gy. 30 sessions at 2.4 ATA against no hyperbaric oxygen. Osteoradionecrosis at six months in 6.4 against 5.7 percent. Dropout was higher in the hyperbaric arm. Funded by Cancer Research UK.
Two randomised trials combined, 65 analysed, surgery for jaw osteoradionecrosis with or without 30 sessions before and 10 after at 243 kPa. Healed at one year 70 against 51 percent, not significant; the authors say the study was underpowered. PubMed tags it as non-government funded; no conflict statement in the record.
Three randomised trials, 88 adults given hyperbaric oxygen and 62 controls. Mean hearing gain 10.3 dB greater (6.5 to 14.1) and odds of recovery 4.3 (1.6 to 11.7), with hyperbaric oxygen as part of combination treatment. No conflicts reported.
Seven trials, 392 participants, small and of generally poor quality. More people reached a 25 percent hearing improvement (risk ratio 1.39) and the average threshold improved 15.6 dB more, but a 50 percent improvement was not significant, and there was no benefit when people presented at six months. Authors: clinical significance unclear. The authors declare no known conflicts.
American Academy of Otolaryngology clinical practice guideline update. As options: clinicians may offer, or refer for, hyperbaric oxygen combined with steroids within 2 weeks of onset, or as salvage therapy within 1 month. PubMed tags it as non-government funded; no conflict statement in the record.
US Air Force single-centre, double-blind, 50 service members after combat mild brain injury. 30 sessions at 2.4 ATA against sham air at 1.3 ATA. No difference on symptom or PTSD checklist scores; both arms improved. PubMed tags it as US government funded; no conflict statement in the record.
Double-blind, 49 enrolled and 47 analysed, persistent symptoms after non-stroke brain injury. 40 hyperbaric sessions against 40 sham sessions over 12 weeks. Symptom inventory fell 10.6 against 3.6 points at 13 weeks (difference 7.0, 1.7 to 12.3). Only 15 of the sham arm completed the later open series. No competing interests declared.
Randomised crossover, 56 patients one to five years after mild brain injury. 40 sessions at 1.5 ATA; the comparison was two months without treatment, not a sham. Cognition and quality of life improved after hyperbaric oxygen and not after the control period. Funded by the hospital research fund, a family foundation and a university chair; no competing interests declared.
Pooled individual data from four US Department of Defense trials, 254 participants. Trends favouring hyperbaric oxygen on post-concussion symptoms (Rivermead -2.3, p=0.18) and on the PTSD checklist (-2.7, p=0.09); verbal memory improved (3.8). Authors call for a definitive trial. No conflicts declared.
Double-blind, sham-controlled, 73 patients, 40 daily sessions. Global cognition, attention, executive function, energy, sleep, psychiatric symptoms and pain interference improved more than with sham (effect sizes 0.463 to 0.737). Two authors work for AVIV Scientific LTD and the senior author is a shareholder, as the paper declares.
Double-blind, 101 patients, four arms of 10 chamber sessions from 100 percent oxygen at 253 kPa (2.5 ATA) down to air at 101.3 kPa (1 ATA). No differences in symptoms, function or cognition between any arms. Authors suggest earlier modest improvements may have been placebo. No conflicts declared.
Living review, 24 trials, 3,695 patients. Moderate certainty that online cognitive behavioural therapy and combined physical and mental rehabilitation probably help. No compelling evidence for hyperbaric oxygen, listed among the other interventions. Funded by Long COVID Web and the Canadian Institutes of Health Research; no relevant financial relationships.
Nine studies, 288 patients, randomised and non-randomised. Pain lower than control (standardised mean difference -1.56). Adverse events in 44 of 185 (23.8 percent); 12 withdrew because of them. No competing interests declared.
Randomised crossover, 60 women with fibromyalgia. 40 sessions at 2 ATA against a period of no treatment, not a sham. Symptoms and quality of life improved after hyperbaric oxygen and not during the control period. Funded by the hospital research fund, a family fund and a university chair; no competing interests declared.
Assessor-blinded randomised crossover, 56 women, 2.3 ATA for 32 to 40 sessions, early against delayed, no sham. Over eight weeks the questionnaire score fell 31.1 against 14.4 percent and pain 54.0 against 33.5 percent; some gains regressed after sessions stopped. 44 completed. The lead author holds stock in and is technical director of a hyperbaric centre, as the paper declares.
Eleven trials, 705 participants, acute ischaemic stroke. No difference in death at six months (risk ratio 0.97); four of 14 disability scales improved. Authors: no good evidence that hyperbaric oxygen improves clinical outcomes in acute stroke. Funded by the Chief Scientist Office; the lead author is a hyperbaric physician who does not routinely treat stroke.
Randomised crossover, 74 patients enrolled and 15 excluded, 6 to 36 months after a stroke. 40 sessions at 2 ATA against two months without treatment, not a sham. Neurological function and quality of life improved after hyperbaric oxygen and not during the control period. Funded by the hospital research fund, a family foundation, a university chair and a private donor.
One eligible trial, 60 children, sham-controlled, low-quality evidence. No improvement in social interaction, behaviour, communication or cognition. More side effects with hyperbaric oxygen (Peto odds ratio 3.87), mostly minor ear barotrauma. The authors declare no known conflicts.
60 Thai children, 20 one-hour sessions at 1.5 ATA oxygen against sham air at 1.15 ATA. Both arms improved; no significant difference on the autism checklist, and no overall clinically significant benefit. PubMed tags it as non-government funded; no conflict statement in the record.
63 healthy adults over 64 randomised to three months of hyperbaric oxygen or no intervention; no sham. Global cognition improved more (interaction p=0.0017), mainly attention (net effect 0.745) and processing speed (0.788). Funded by the Sagol network; six authors work for AVIV Scientific LTD and one is a shareholder, as the paper declares.
20 healthy young adults, placebo-controlled imaging study. Spatial working memory and memory quotient scores rose after hyperbaric oxygen. PubMed tags it as non-government funded; no conflict statement in the record.
Single-arm trial, 35 healthy adults aged 64 and over, 60 daily sessions. Telomere length in T helper, T cytotoxic, natural killer and B cells rose by over 20 percent; senescent T helper cells fell 37.30 percent. No control group. Funded by the Sagol network; three authors work for AVIV Scientific LTD and one is a shareholder, as the paper declares.
13 men of about 68 from the same healthy-ageing cohort, biopsies after three months without and then three months with daily sessions. Collagen density, elastic fibre length and blood vessels rose and senescent cells fell after sessions. Funded by the Sagol network; four authors work for AVIV Scientific LTD and one is a shareholder, as the paper declares.
Double-blind, sham-controlled, 37 master athletes aged 40 to 50; 16 in the hyperbaric arm and 15 in the sham arm were analysed. 40 sessions at 2 ATA against air at 1.02 ATA. VO2max rose more with hyperbaric oxygen (effect size 0.989), with gains in mitochondrial respiration. Four authors work for AVIV Scientific LTD and one is a shareholder and co-founder, as the paper declares.
10 randomised trials, 299 people. Markers of exercise muscle injury recovered faster with hyperbaric oxygen; soreness did not (interval -0.91 to 0.48). PubMed tags it as non-government funded; no conflict statement in the record.
Eleven studies, six pooled, healthy athletes. Post-recovery blood lactate lower (effect size -1.71); limited effect on short-term power or force. No conflicts declared.
Nine small trials, 219 participants. An ankle sprain trial (32 people) and a knee ligament trial showed no differences. Pooled soreness trials showed higher pain at 48 hours when hyperbaric oxygen started immediately (0.88 points on a 10-point scale). Conflict statement: the lead author is a hyperbaric physician who does not routinely treat these injuries, and several reviewers ran included trials and did not appraise their own.
24 healthy men, eccentric elbow work, seven daily sessions of 60 minutes at 2.5 ATA oxygen against 2.5 ATA normoxic gas. No difference in strength recovery, soreness or arm swelling. PubMed tags it as non-government funded; no conflict statement in the record.
Randomised crossover, 16 healthy young women, one 70-minute session at 1.4 ATA and 35 to 40 percent oxygen against room air. Natural killer cells and parasympathetic activity rose; no rise in oxidative markers. No conflicts declared.
Randomised crossover, 10 men, 60 minutes at 1.3 ATA and 31 percent oxygen after hard cycling. Fatigue, tension and total mood disturbance scores fell; no oxidative stress difference. No conflict statement in the PubMed record.
60 young men living at 4300 m, five days of mild hyperbaric oxygen against control. Sleep quality score improved more (adjusted difference -1.08) and working memory accuracy rose. No competing interests declared.
37 animal studies, 2141 animals: 17 on muscle, 12 on tendon and ligament, 6 on bone, 2 on cartilage. Histology improved more often than biomechanics. Risk of bias high or unclear in most studies; protocols ranged from 1 to 56 sessions. No competing interests declared.
76 rats with a 2 mm segment removed from the knee's medial collateral ligament. 2.5 ATA for 2 hours, 5 days a week, against room air. More scar tissue, more type I procollagen gene expression, and greater ultimate load and stiffness at 14 days. No conflict statement in the PubMed record.
44 rats, three exposures compared against chamber air: 1.5 ATA for 30 minutes, 2 ATA for 30 minutes, and 2 ATA for 60 minutes. All helped ligament healing at 14 days; 2 ATA for 60 minutes gave the most collagen synthesis. PubMed tags it as non-government funded; no conflict statement in the record.
56 rats with a sutured Achilles tenotomy, half pretreated with steroid injections. On day 11 the hyperbaric groups had more fibrosis and differed on maximum force, stiffness and stress; the abstract does not give the direction of each, and the authors concluded the effect was positive. No conflict statement in the PubMed record.
Rabbits after ACL reconstruction, 2.5 ATA for 2 hours a day for 5 days from the day after surgery, against room air. At 12 weeks better graft maturation on MRI, smaller bone tunnels, and greater load to failure and stiffness. PubMed tags it as non-government funded; no conflict statement in the record.
Rabbits with a full-thickness supraspinatus tear, 2 ATA for 2 hours a day for 5 days. More vessels and collagen regeneration at the tendon-bone interface and higher tensile strength, with no significant effect on biomechanical stability. No conflict statement in the PubMed record.
60 Wistar rats, ageing induced by D-galactose injections, with or without a high-fat diet. 14 days of 2 ATA pure oxygen for 80 minutes a day against sham. Insulin resistance, cognitive impairment and hippocampal changes were attenuated. PubMed tags it as non-government funded; no conflict statement in the record.
Male Wistar rats with chemically induced ageing, obese or lean, 14 days of sessions against sham. Bone microarchitecture fully restored in lean rats and partly in obese rats; senescence genes p16 and p21 shifted. PubMed tags it as non-government funded; no conflict statement in the record.
Alzheimer model mice. Cognitive flexibility and associative learning improved in females but not males; balance improved; gait and anxiety-like behaviour did not change. Funded by the US NIA; no competing interests declared.
Aged female mice, 10 days of hyperbaric oxygen. Serum AMH, oocyte maturation, fertilisation and blastocyst formation improved against untreated aged mice. No competing interests declared.
51 animal stroke studies. Infarct size 32 percent smaller and mortality 18.4 against 26.7 percent with hyperbaric oxygen, largest when given immediately at 2.0 ATA. Funded by the National Natural Science Foundation of China; no competing interests declared.
2,334 patients, 62,614 sessions at one Israeli centre, 2010 to 2014. Seizures in 0.011 percent of sessions (1 in 8,945); one clear oxygen-toxicity seizure (1 in 62,614). Safe for chronic neurological disorders except uncontrolled epilepsy, the authors conclude. No conflict statement in the PubMed record.
Eight units, 5,193 patients, 96,670 treatments, 2010 to 2014. At 243 kPa, 2.4 seizures per 10,000 treatments and a per-patient risk of 1 in 222 (0.45 percent). Per-patient risk is higher than per-session figures suggest, the authors note. No conflict statement in the PubMed record.
5,962 patients at a Verona unit, 2003 to 2010. Middle ear barotrauma recorded at a rate of 9.2 percent, mostly minor, more often in women over 50 and in people with rhinitis or difficulty equalising. 1.49 percent of all patients had therapy suspended because of it. No conflicts declared.
20 patients given oxygen by mask and 12 by hood, 21 daily sessions of 90 minutes at 240 kPa. A myopic shift of about -0.55 dioptre by mask and -1.10 by hood, back to baseline within 6 and 10 weeks; no change in lens transparency. PubMed tags it as non-government funded; no conflict statement in the record.
Review of the side effects of hyperbaric oxygen for informed consent: what they are, how often, and how to prevent them. Notes that protocols vary between institutions. No commercial conflicts, the authors declare.
143 participants, 4,245 chamber sessions in two US military trials. A room-air sham at 1.2 ATA kept the blind; people guessed their arm from whether they felt better. Chamber adverse events 1.1 and 2.2 percent, mostly minor barotrauma. No conflicts declared.
Educational use only. This page is not medical advice, not a referral, and not a treatment plan. Pressures and session counts are what a trial ran, not a protocol for you. The indications listed by medical societies are decided by a hyperbaric physician for a named patient. Aaron Cuha is a coach, not a doctor.
Educational use only. This database summarizes published research and is not medical advice. Where a dose appears it is a record of what a trial or clinic ran, or what people describe doing, with its source attached. Nothing here is a recommendation for personal use.