Teenage lighthouse keeper and her father beside a cracked lens casting light across a misty sea.
Back to the library

Brain

Concussion and brain injury

Studies have looked at hyperbaric oxygen for people with symptoms that lingered months or years after a head injury. Randomised trials measured memory, attention and brain blood flow before and after a course of sessions, and reported improvements in the treated groups.

Evidence so far
Moderate
Reading timeAbout 12 minutes

Start here

Why this is being studied

A concussion is a mild traumatic brain injury in which the brain is shaken inside the skull; most people recover in weeks, but some are left with months or years of poor concentration, memory problems, headaches and fatigue despite normal scans. Functional imaging suggests that in these people some brain regions are structurally intact but underperfused and metabolically sluggish. Hyperbaric oxygen is being studied because it delivers far more dissolved oxygen to such tissue and appears to stimulate small-vessel growth and cellular repair.

Where the evidence stands

Flag text

The mechanism

What is going on, and why this might help

What is going on in the body

After a concussion or more severe brain injury, the initial mechanical insult is followed by a secondary cascade that can last weeks to years. This includes neuroinflammation driven by activated microglia, disruption of the blood-brain barrier, swelling, and damage to the small vessels that supply grey and white matter. Some regions become chronically under-perfused and metabolically sluggish: cells there are alive but not functioning normally, sometimes called the ischaemic penumbra or 'stunned' tissue. Mitochondrial dysfunction reduces energy supply, axons carry signals less efficiently, and neuroplastic repair is limited. Persisting symptoms such as headache, poor concentration, mood change and fatigue are thought to reflect this combination of inflammation, hypoperfusion and energy deficit.

Where oxygen comes in

Hyperbaric oxygen raises dissolved plasma oxygen many-fold, allowing it to diffuse into poorly perfused brain tissue. In animal models of brain injury it reduces cerebral oedema by mild vasoconstriction while still delivering more oxygen, restores mitochondrial function and dampens microglial activation and cytokines including TNF-α and IL-1β. It stimulates angiogenesis through VEGF and HIF-1α-related pathways, and mobilises bone-marrow stem cells (Thom 2006), which in rodents home to injured brain. The hyperoxic-hypoxic paradox (Hadanny and Efrati 2020) offers a framework for why repeated sessions may trigger regenerative gene expression without sustained hypoxia. Human studies using SPECT and perfusion MRI have reported increased blood flow in previously under-perfused regions after a course of sessions, which is the most direct human link to the proposed mechanism.

How solid is this

Neuroprotective and anti-inflammatory mechanisms are robustly demonstrated in animal models. Human trials in persistent post-concussion symptoms have produced mixed results, partly because of debate about whether low-pressure air controls are truly inert, so this sits in the 'supported by some trials, mechanistically plausible' category rather than established. Key references: Boussi-Gross et al. 2013; Hadanny and Efrati 2020.

Explainer illustration

The studies

Every study is laid out the same way: what the researchers wanted to find out, who took part, what they actually did, what they found, why it might work and where the limits are. No jargon, and a link to the original if you want to read it yourself.

Each study is laid out the same way: what the researchers asked, who took part, what they did, what they found, why it might work, and where the limits are. The clinician layer at the end of each one carries the numbers.

Paper 1

Hyperbaric oxygen therapy can improve post concussion syndrome years after mild traumatic brain injury: randomized prospective trial

Boussi-Gross et al. · PLOS ONE · 2013 · randomised, crossover-controlled trial, n=56

Read the paper

What they asked

Most people recover from a concussion within weeks, but a minority are left with persistent problems in memory, attention and thinking speed that can last for years, even when routine CT and MRI scans look normal. This Israeli trial asked whether a course of hyperbaric oxygen given one to six years after the injury could still improve cognition and brain function.

Who took part

Fifty-six adults, average age 44 (range 21 to 66), who had suffered a mild traumatic brain injury (loss of consciousness under 30 minutes) between one and six years earlier, on average nearly three years before. All still had post-concussion symptoms, had normal structural scans, and had been cognitively stable for at least a month before enrolling. Smokers, people with lung or inner-ear problems, and anyone with claustrophobia were excluded.

What they did

Participants were randomised to start treatment immediately (32 people) or to wait two months as a control period and then receive the same treatment (24 people), a design known as a crossover. Treatment was 40 sessions of 100 per cent oxygen at 1.5 ATA for 60 minutes, five days a week over two months. Cognition was tested with a computerised battery covering processing speed, attention, memory and executive function; quality of life was measured with the EQ-5D; and brain perfusion was imaged with SPECT. Assessors were blinded to group, though participants obviously knew whether they were in the chamber.

What they found

During the two-month waiting period, the control group did not improve on any cognitive measure. After treatment, both groups improved significantly in all four cognitive domains. In the immediate-treatment group, the gains were moderate to large: processing speed rose from about 85 to 95 on a scale where 100 is the population average, with similar ten-point improvements in attention, memory and executive function. The crossover group improved by somewhat less after their own treatment. Quality of life improved in both groups. SPECT scans showed increased activity in frontal and temporal regions, the areas most vulnerable in concussion, and no change during the waiting period.

Why it might work

The authors argue that mild brain injury leaves regions of tissue that are structurally intact but metabolically dormant, with reduced blood flow and impaired energy production. SPECT showed exactly such patches. Raising dissolved oxygen in the blood many-fold may restore function in these cells, stimulate new small blood vessels and support the regrowth of connections. The correspondence between improved perfusion and improved test scores supports, but does not prove, this idea.

Limits

There was no sham chamber, so participants knew when they were being treated, and expectation could contribute to the results. The control period was untreated rather than sham-treated. Outcomes were measured one to three weeks after the course, so durability is unknown. Fifty-six people is a modest sample and baseline scores varied widely. SPECT has limited spatial resolution. Funding came from hospital and charitable sources with no industry involvement, and the authors declared no competing interests.

Paper 2

Hyperbaric oxygen therapy for mild traumatic brain injury persistent postconcussion syndrome: a randomized controlled trial

Harch et al. · Medical Gas Research · 2020 · randomised, crossover-controlled trial, n=63

Read the paper

What they asked

This US trial, funded by the US Army, asked whether 40 sessions of hyperbaric oxygen at 1.5 ATA could improve persistent post-concussion symptoms in a mixed group of military and civilian adults years after their injury, using a design in which an untreated control group later received the same treatment.

Who took part

Sixty-three adults aged 18 to 65 enrolled and 50 completed the protocol with primary outcome testing (23 treated first, 27 controls). All had mild traumatic brain injury from blast or blunt causes at least six months earlier; the average time since injury was 4.6 years. The group included both veterans and civilians. We relied on the published abstract and the investigators' own summary because the full text sits behind a paywall.

What they did

Participants were randomised either to immediate treatment or to an eight-week untreated control period followed by identical treatment (a crossover). Treatment was 100 per cent oxygen at 1.5 ATA (150 kPa) for 60 minutes without air breaks, once a day, five days a week, for 40 sessions over eight weeks. There was no sham chamber. The two co-primary outcomes were the Neurobehavioral Symptom Inventory, a standard post-concussion symptom questionnaire, and a working memory index from neuropsychological testing. Secondary outcomes covered memory, overall cognitive efficiency, depression, anxiety, sleep quality, quality of life and post-traumatic stress symptoms. Everyone was reassessed about two months after treatment ended.

What they found

Symptom scores fell by about 26 points in the treated group compared with 2.5 points in controls over the same eight weeks, a large and highly significant difference. The working memory index, the other primary outcome, did not differ significantly between groups. Eight of 14 secondary measures favoured treatment, including memory, overall cognitive efficiency, depression, anxiety, sleep quality, quality of life and post-traumatic stress symptoms. When the control group later received treatment, their improvements were almost identical to those of the first group. Headache improved in 83 per cent of the treated group. At two-month follow-up most gains were maintained or extended.

Why it might work

The rationale is similar to the Israeli work: chronic post-concussion symptoms may reflect brain tissue that is underperfused and inflamed rather than destroyed, and repeated moderate hyperoxia may drive repair of small vessels, restore cellular energy production and reduce inflammation. The authors emphasise that the natural history of persistent post-concussion syndrome is not one of spontaneous improvement years out, which makes the change harder to attribute to time alone.

Limits

No sham, and participants knew their allocation, so expectation effects are possible; the symptom questionnaire was administered by an unblinded technician, though the neuropsychologist was blinded. One of the two primary outcomes was negative. Dropouts required the sample to be enlarged, and the final groups were unequal in size. The principal investigator runs a hyperbaric consulting company and has a financial relationship with the treating facility, which is declared. Funding was from the US Army Medical Research and Materiel Command with a small foundation bridging grant.

Paper 3

Hyperbaric oxygen therapy in children with post-concussion syndrome improves cognitive and behavioral function: a randomized controlled trial

Hadanny et al. · Scientific Reports · 2022 · randomised controlled trial in children, n=25

Read the paper

What they asked

Children can also be left with persistent post-concussion symptoms, and there is very little treatment evidence for them. This was the first randomised, sham-controlled trial of hyperbaric oxygen in children, asking whether it could improve cognition, behaviour and symptoms months to years after the injury.

Who took part

Twenty-five children aged 8 to 15 (average 11.6, 80 per cent boys) whose concussion had happened between six months and ten years earlier. Fifteen received oxygen and ten received sham. The trial had aimed for 70 children but recruitment fell far short, partly because of the pandemic years during which it ran. About a third of the children were taking ADHD medication.

What they did

The treatment group breathed 100 per cent oxygen at 1.5 ATA for 60 minutes, without air breaks, five days a week, for 60 sessions over three months. The sham group breathed air at 1.03 ATA for 60 minutes, with the chamber briefly pressurised to 1.1 ATA at the start so it felt real. Blinding worked: roughly seven in ten children in each group believed they had received the real treatment. Assessment before and after included a computerised cognitive battery, standard paediatric neuropsychological tests, parent-rated behaviour scales (Conners-3, BRIEF), post-concussion symptom inventories, quality of life, balance testing and diffusion MRI.

What they found

The oxygen group's general cognitive score rose from about 90 to 96 (100 being average for age), with no change in the sham group. Memory, figural fluency and phonemic verbal fluency also improved only in the treated children. Parents reported clearer improvements in executive function on the BRIEF questionnaire, with large effects for planning, initiation and inhibition, and smaller but significant improvements in hyperactivity and symptom scores. Quality of life in psychosocial and school domains improved. Diffusion MRI showed changes consistent with improved tissue microstructure in several regions, correlating with the cognitive gains. Adverse events, mainly ear discomfort and mild barotrauma, were common in both groups and none were serious; one child in each group needed grommets.

Why it might work

The developing brain is metabolically demanding, and post-concussion symptoms may reflect regions where blood supply and energy production have not recovered. The authors propose that repeated hyperoxia stimulates new vessel growth, mitochondrial recovery and reduced neuroinflammation; the diffusion MRI changes are consistent with that but cannot show mechanism directly.

Limits

Twenty-five children is a very small sample and the trial was badly underpowered against its own target of 70. Some effect sizes on symptom scales were small even where statistically significant. ADHD medication use was not controlled. There is no long-term follow-up. Ear problems were frequent, which matters in children. Funding and conflicts were not clearly stated in the version we read; the senior authors have declared interests in a commercial hyperbaric company in related papers.