TL;DR: Hyperbaric oxygen therapy (HBOT) has been investigated for multiple sclerosis (MS) since the early 1980s, but the randomised evidence base is dominated by twelve trials conducted between 1983 and 1987. The 2004 Cochrane systematic review and its 2010 update both concluded there is no consistent clinical benefit. Recent mechanistic research (2020-2024) on human brain endothelial cells and the experimental autoimmune encephalomyelitis (EAE) mouse model shows plausible anti-inflammatory effects, but no contemporary randomised controlled trial has been published. For Canadian researchers, the gap between mechanistic plausibility and the absence of modern clinical evidence remains a defining feature of this research area.
Hyperbaric oxygen therapy for multiple sclerosis is one of the longest-running open questions in hyperbaric medicine. Multiple sclerosis is a chronic, inflammatory, demyelinating disease of the central nervous system characterised by relapsing or progressive neurological disability. Early clinical interest in HBOT was driven by the hypothesis that intermittent exposure to elevated partial pressures of oxygen might modulate the vascular and immune mechanisms involved in demyelination. Forty years later, the question of whether that hypothesis translates into measurable clinical benefit has not been resolved by adequately powered modern trials. Canada Hyperbarics is publishing this 2026 evidence summary to give Canadian researchers a current snapshot of where the literature stands and where the methodological gaps remain.
What does the randomised evidence on HBOT for multiple sclerosis show?
The most cited reference in the field remains the 2004 Cochrane systematic review by Bennett and Heard, which identified ten reports of nine randomised trials enrolling 504 participants in total. The review found that two trials produced generally positive results, while the remaining seven reported no evidence of a treatment effect. Three of twenty-one pre-specified analyses indicated some benefit. For example, the mean Expanded Disability Status Scale (EDSS) at twelve months was lower in the HBOT group, with a group mean reduction in EDSS compared to sham of 0.85 points (95% confidence interval -1.28 to -0.42, P = 0.0001), but only the two generally positive trials reported on this outcome at this time point and they contributed only 16% of total participants. The authors concluded that there was no consistent evidence of benefit and that routine use was not justified.
The 2010 follow-up by the same authors in CNS Neuroscience and Therapeutics reviewed twelve randomised studies, all conducted between 1983 and 1987 (Bennett and Heard, 2010). The great majority of trials investigated a course of twenty treatments at pressures between 1.75 ATA and 2.5 ATA daily for 60 to 120 minutes over four weeks against a placebo regimen. None tested HBOT against alternative current best practice. The meta-analysis again concluded that no plausible benefit of HBOT on the clinical course of MS was identified, although the authors acknowledged that benefit in a subgroup not clearly identified in the trials to date could not be ruled out, and that prolonged courses at modest pressures had not been adequately tested.
Why does mechanistic interest in HBOT for MS persist?
Despite the negative clinical meta-analyses, translational research has continued. The persistence is grounded in the cellular biology of demyelination. MS pathology involves blood-brain barrier dysfunction, T-cell-mediated inflammation, and oligodendrocyte injury. Each of these processes is, in principle, modifiable by oxygen tension. Two recent studies illustrate this line of inquiry.
A 2020 UK study by Moore and colleagues, published in Multiple Sclerosis and Related Disorders, combined patient and public involvement with a laboratory study of human brain microvascular endothelial cells. The investigators exposed cells in vitro to 152 mmHg of oxygen for 60 minutes with and without pressure, replicating the 20 percent oxygen exposure achieved during a typical 1.5 ATA, 60-minute session reported by individuals with MS in the United Kingdom. A significant reduction in endothelial ICAM-1 (CD54) expression was observed under oxygen treatment. ICAM-1 is implicated in inflammatory cell margination across the blood-brain barrier and is a measurable correlate of the cellular events thought to precede MS lesion formation.
A 2021 preclinical study by Chiou and colleagues from Kaohsiung Medical University evaluated HBOT in the experimental autoimmune encephalomyelitis (EAE) mouse model, which is the standard preclinical model for MS-like demyelination. Semi-therapeutic HBOT significantly attenuated EAE progression. The mechanism appeared to involve suppression of the Th17 response, downregulation of CD4 T-helper cells expressing GM-CSF or TNF-alpha, and a boost in immunomodulatory IL-4 and IL-10-expressing CD4 T-cells in central nervous system lesions. The authors framed HBOT as a possible modulator of the early-stage T-cell response that drives EAE pathology.
These two studies do not constitute clinical evidence of benefit. They establish biological plausibility at the cellular and animal-model level. The disconnect between this mechanistic plausibility and the absence of contemporary clinical trial data is the defining methodological gap in this research area.
What are the methodological limitations of the historical trial base?
The twelve randomised trials that anchor both Cochrane reviews share several limitations relevant to any future research programme.
- Trial vintage. All twelve trials were conducted between 1983 and 1987. Disease classification, diagnostic criteria, and outcome instruments have all evolved substantially since then. The McDonald criteria for MS diagnosis were introduced in 2001 and revised in 2010 and 2017, with further updates more recently. No trial in the Cochrane evidence base used modern diagnostic criteria.
- Comparator design. None of the historical trials tested HBOT against current disease-modifying therapy. All used sham as the comparator. Modern relapsing-remitting MS is treated with high-efficacy disease-modifying therapies including monoclonal antibodies and oral small molecules. Any future trial would need to address whether HBOT offers benefit as adjunctive therapy or in patients who cannot tolerate disease-modifying therapy.
- Outcome measures. The EDSS, although widely used, is insensitive to short-term change and weighted toward ambulation. Modern MS trials typically include MRI-based outcomes such as T2 lesion load, gadolinium-enhancing lesions, and brain atrophy, alongside patient-reported outcomes and composite measures such as the Multiple Sclerosis Functional Composite. None of these were used in the historical HBOT trials.
- Dosing heterogeneity. Pressure regimens varied between 1.75 ATA and 2.5 ATA, with treatment counts and session durations also differing. The 2010 Bennett and Heard update notes that prolonged courses at modest pressures, which more closely resemble the regimens self-administered by individuals with MS through patient-led therapy centres in the United Kingdom, were not adequately tested.
- Heterogeneity of disease subtype. Modern MS phenotyping distinguishes relapsing-remitting, secondary progressive, and primary progressive disease. The historical trials did not stratify by phenotype in a way that would be considered acceptable in contemporary trial design.
What does this mean for future research design?
A contemporary randomised controlled trial of HBOT in MS would need to address each of these gaps explicitly. Key design considerations include modern diagnostic criteria with pre-specified phenotype stratification, MRI-based primary or co-primary endpoints alongside EDSS or composite measures, and a clinically meaningful comparator, which in 2026 means active disease-modifying therapy rather than sham alone. Dosing regimens should be informed by mechanistic studies, including the 1.5 ATA range investigated in the Moore et al. in vitro work. Pre-specified subgroup analyses for patients with high inflammatory activity, in whom the mechanistic case is strongest, would help interpret heterogeneous response. Patient-reported outcomes and quality-of-life measures should also be included, given the gap between EDSS-based outcomes and patient-reported benefit historically described by individuals with MS who use oxygen therapy.
Where do Canadian researchers fit in?
Canada has one of the highest age-standardised MS prevalence rates in the world, with the MS Society of Canada estimating that more than 90,000 Canadians live with the condition. The country also has a mature hyperbaric medicine infrastructure including hospital-based programmes and regulated private clinics. The combination of high disease burden, established hyperbaric capacity, and strong neurological research networks is a research opportunity that has not been systematically exploited. Canada Hyperbarics maintains an evolving directory of hospitals and regulated facilities that could in principle contribute to a multi-centre clinical trial. The Canada Hyperbarics research database currently indexes over 14,000 hyperbaric publications, of which a small subset addresses MS directly.
Health Canada considers HBOT for MS an investigational indication. It is not on the approved indications list maintained by either Health Canada or the Undersea and Hyperbaric Medical Society. Researchers planning trial work in this area should review the current Canadian regulatory framework for hyperbaric chambers and investigational indications before designing protocols.
What evidence gaps remain most pressing?
The most consequential evidence gap is the absence of any randomised trial of HBOT for MS published in the last 35 years that incorporates modern diagnostic criteria, modern outcome instruments, or the contemporary disease-modifying therapy landscape. The translational research summarised above provides biological hypotheses worth testing. The case for a well-designed exploratory trial in a defined inflammatory phenotype, with MRI-based outcomes and an active comparator, is stronger than the existing clinical literature alone would suggest, but it is not strong enough to justify clinical adoption outside a research protocol.
Comparing the historical trials and current mechanistic studies
| Feature | Historical RCTs (1983-1987) | Recent translational work (2020-2024) |
|---|---|---|
| Study type | Randomised sham-controlled clinical trials | In vitro endothelial cell study; EAE mouse model |
| Pressure range | 1.75 to 2.5 ATA | Approximately 1.5 ATA |
| Primary outcome | EDSS (clinical disability) | ICAM-1 expression; Th17 and IL-10 cytokine modulation |
| Diagnostic criteria | Pre-McDonald criteria | Not clinically applicable (preclinical) |
| Result | No consistent clinical benefit | Biological plausibility for anti-inflammatory effect |
| Direct clinical translation | Trial corpus is dated | Mechanistic hypothesis untested in modern RCT |
Frequently asked questions from the research community
Is HBOT an approved treatment for multiple sclerosis in Canada?
No. HBOT for MS is not on the Health Canada approved indications list and is not among the conditions for which the Undersea and Hyperbaric Medical Society currently recommends HBOT. Any clinical use in MS is investigational and should occur within a research protocol.
How many randomised controlled trials exist on HBOT for MS?
The 2010 Bennett and Heard meta-analysis identified twelve randomised trials, all conducted between 1983 and 1987. No contemporary randomised controlled trial using modern diagnostic criteria or outcome instruments has been published since.
What did the Cochrane review conclude about HBOT for MS?
The 2004 Cochrane review concluded that there was no consistent evidence of benefit and that routine use was not justified. The 2010 update reached the same overall conclusion, while acknowledging that prolonged courses at modest pressures and subgroup-specific benefit had not been adequately tested.
What does recent mechanistic research suggest?
Recent in vitro and animal-model work suggests biological plausibility for an anti-inflammatory effect of HBOT relevant to MS pathology. Specific findings include reduced endothelial ICAM-1 expression at 1.5 ATA in human brain microvascular cells and attenuation of EAE in mice via modulation of Th17 and IL-10 responses. None of this constitutes clinical evidence of benefit.
What pressure regimens have been studied?
Historical RCTs used pressures between 1.75 ATA and 2.5 ATA for 60 to 120 minutes daily, typically over a four-week course of about twenty sessions. Recent mechanistic work has focused on lower-pressure regimens around 1.5 ATA, which more closely resemble what individuals with MS self-administer through patient-led therapy centres.
Is there a Canadian research programme on HBOT for MS?
No dedicated Canadian programme is currently active to our knowledge. Canada has the underlying capacity in MS research networks and regulated hyperbaric medicine infrastructure to host a contemporary trial, but no such trial has been registered or published.
Where to read more
Researchers interested in the underlying evidence base can review the source studies summarised on the Canada Hyperbarics research database. For the Canadian access landscape and chamber infrastructure, the directory of hospitals and regulated facilities documents which sites operate hyperbaric chambers and under what regulatory status. For protocol questions about investigational indications, the regulatory overview summarises the current Health Canada and Undersea and Hyperbaric Medical Society positions. Researchers exploring related neurological indications may also wish to review the Canada Hyperbarics conditions index, which catalogues both approved and investigational indications with linked source studies.
Conclusion
The state of the evidence for hyperbaric oxygen therapy in multiple sclerosis is, in 2026, defined as much by what is missing as by what has been shown. The randomised trial corpus is small, dated, and methodologically out of step with modern MS research. Mechanistic studies in human brain endothelial cells and the EAE mouse model suggest biological plausibility but do not establish clinical benefit. The strongest research opportunity is a contemporary, phenotype-stratified, MRI-based trial that takes mechanistic findings seriously while applying current methodological standards. Canada is well positioned to host such work, and Canadian researchers interested in the field will find the evidence map both unsatisfying and, for that reason, an unusually open territory.
Read more from Canada Hyperbarics: Explore the directory of hospitals and regulated facilities across Canada or browse the full research database of indexed hyperbaric publications.
Medical disclaimer: This content is for informational purposes only and does not constitute medical advice. Hyperbaric oxygen therapy for multiple sclerosis is an investigational indication that is not approved by Health Canada or the Undersea and Hyperbaric Medical Society. Patients with multiple sclerosis should discuss any treatment decisions with their neurologist and a hyperbaric physician.