Fibromyalgia Responds in the Chamber. The Question Is What Happens After

Illustration of a monoplace hyperbaric chamber seen from the side, its interior glow fading toward one end, with a wall grid of illuminated session markers whose final rows sit dark and unlit

For a condition that affects something close to one person in twenty, fibromyalgia has produced a surprisingly thin hyperbaric literature. The trials have been small, rarely blinded, and — this is the part that matters — almost all of them measured their outcomes while patients were still inside the treatment course. As recently as June, a survey of the field in this publication concluded that “the long-term durability of any benefits remains unstudied.”

A trial published in BMJ Open that month has now put a first real number on the question, and the answer is more interesting than either camp in this argument would prefer.

The study is called HOTFy, run out of the rheumatology department at the University Hospital of the Federal University of Juiz de Fora in Brazil. Its results appeared on 1 June. On 22 October the same group published something considerably rarer in this corner of medicine: a cost-effectiveness analysis built on the same trial.

What the Brazilian Trial Actually Did

Fifty-six women aged 18 to 70, diagnosed at least two years earlier, were randomised one-to-one. Everyone continued standardised care — education, supervised exercise, physiotherapy, and the usual drugs, from antidepressants to anticonvulsants to muscle relaxants. Half began forty sessions of hyperbaric oxygen immediately; the other half waited eight weeks and then received the identical protocol. Sessions ran at 2.3 atmospheres absolute for 90 minutes, five days a week, across eight weeks, in monoplace chambers. The assessing rheumatologist was blinded to allocation. The analysis was intention-to-treat, with all 56 patients included and missing values handled by expectation maximisation. Forty-four finished; the withdrawal rate was 21.4 percent, mostly from non-compliance or difficulty tolerating confinement.

The effects were large and they pointed one way. On the Revised Fibromyalgia Impact Questionnaire the combined-treatment group fell 31.1 percent against 14.4 percent for standard care alone. Pain on a visual analogue scale dropped 54.0 percent against 33.5 percent. The psychopathology scale improved 28.4 percent against 3.7 percent, the physical component of the SF-12 rose 39.1 percent against 14.8 percent, and the mental component 57.4 percent against 31.9 percent. Within-group effect sizes ran to Cohen’s d of 2.5 to 2.7 for pain and 1.4 to 1.7 for functional impact — figures that, in chronic pain research, are unusual enough to invite scepticism rather than applause.

The authors supply part of that scepticism themselves. There was no placebo arm, a limitation they state plainly, and no classical washout period between the two phases. Both are the standing objections to hyperbaric trials in painful conditions, where lying in a pressurised chamber for ninety minutes, five times a week, for two months is not nothing.

Then the Sessions Stopped

Here is where HOTFy earns its place in the file. Because the design crossed everyone over, the early group spent weeks eight through sixteen off treatment while the delayed group was being treated — in effect an eight-week observation window on what happens when the sessions end.

What happened was uneven, and the pattern is the interesting part. Pain, the psychopathology score and the physical component of the SF-12 held their gains for up to eight weeks after the last session. But the Fibromyalgia Impact Questionnaire — the trial’s primary functional measure — and the mental component of the SF-12 “regressed towards standard care levels,” in the paper’s own words. The measure that came back first was the one patients would notice most: how much the condition interferes with daily life.

The authors are careful about what this means, and the care is warranted. Their conclusion is that “the durability of the benefit appears to be exposure dependent, and maintenance or booster schedules merit further evaluation.” That is a hypothesis generated by an eight-week tail, not a finding about maintenance dosing. The trial measured three time points inside sixteen weeks and then stopped. There is no twelve-month data, and no data at all from the delayed group after it finished its own course.

Against that, a smaller Canadian trial points the other way. Curtis and colleagues at the Toronto General Hospital hyperbaric unit randomised eighteen patients to immediate treatment or a twelve-week wait, ran 2.0 atmospheres at 100 percent oxygen for eight weeks, and followed everyone to three months. Their improvement was sustained at that assessment. It is a feasibility study with nine patients per arm and cannot settle anything — but it does establish that an eight-week plateau is not the only possible trajectory. That group also recorded the complications this therapy reliably produces at low frequency: mild middle-ear barotrauma in three patients and new-onset myopia in four.

The pooled literature has not been able to interrogate durability properly. A systematic review and meta-analysis in Reumatologia in June pooled five randomised trials and 227 women. Pain improved by a mean difference of 3.41 points on a ten-centimetre scale (95 percent CI 6.24 to 0.59), symptom severity by 5.27 points, and quality of life not at all. The mean follow-up across the five trials was 2.4 months, ranging from two to three. Sixteen weeks is not a durability literature, and a pooled estimate drawn from three-month horizons cannot answer a question about a year.

The Word “Hyperbaric” Is Carrying a Lot of Weight

There is a second problem, and it is one this publication has tracked through altitude medicine, topical oxygen and soft chambers: the word in the title does not describe the same intervention twice.

The Spanish group that has done the most work on low-pressure treatment published its 2020 randomised trial in Therapeutic Advances in Musculoskeletal Disease — 49 women in three arms, comparing what the paper calls low-pressure hyperbaric oxygen against supervised physical exercise and against no treatment. The methods are explicit: 100 percent oxygen at 97 percent purity delivered by mask, forty sessions of 90 minutes at five per week, in a Revitalair 430 chamber, at 1.45 atmospheres absolute. The rationale is candid. The 2 to 2.5 ATA protocols used elsewhere carry middle-ear and sinus barotrauma and prodromal signs of central nervous system oxygen toxicity, so lower pressure, in the authors’ words, “should be tested.” Their results were mixed in a useful way: perceived pain and induced fatigue improved only in the oxygen arm, while pressure pain threshold, endurance and physical performance improved with exercise too — and resting motor threshold, a measure of cortical excitability, did not move in either group.

Set that beside the Israeli programme. Efrati and colleagues ran forty sessions at 2.0 ATA in a 2015 crossover trial and reported normalisation of activity in pain-processing regions on SPECT, with no improvement during the control period. Boussi-Gross and colleagues, in a 2024 trial of adults whose fibromyalgia followed childhood sexual abuse, ran sixty sessions at 2.0 ATA against pregabalin or duloxetine and reported a functional improvement with a Cohen’s d of 1.27. HOTFy ran 32 to 40 sessions at 2.3 ATA.

So the fibromyalgia literature contains at least three genuinely different doses — 1.45, 2.0 and 2.3 atmospheres absolute — with session counts from 32 to 60, and a Spanish follow-up trial of 33 women whose abstract does not state the pressure at all. The meta-analysis that pooled five of these trials reports an I² of 89 percent for pain and 91.4 percent for symptom severity, calls the heterogeneity substantial, and says that identifying its sources “warrants further research.” Dose is the most obvious candidate and the review does not test it.

None of that is grounds for dismissal. It is grounds for precision about what any single result refers to. A 1.45 ATA course in a soft-sided chamber at 97 percent oxygen is not the same physiological intervention as 2.3 ATA in a monoplace vessel, whatever heading sits above the table.

The Invoice Arrives Before the Answer Does

The October paper is the one that changes the shape of the argument. An economic evaluation has now been published alongside the trial, again in BMJ Open, from the same Juiz de Fora group. What its incremental cost-effectiveness ratio is, this publication cannot fairly report: the full text sits behind a publisher challenge we could not clear, and the indexed record carries no abstract. What can be stated is what the title asserts — that this is a cost-effectiveness analysis of adjunctive hyperbaric oxygen for fibromyalgia, built on a single-centre randomised controlled trial in Brazil.

That a group would do the arithmetic at all is the story. Fibromyalgia is nowhere in the regulatory scaffolding around this therapy. It does not appear among the fifteen indications in the Undersea and Hyperbaric Medical Society’s Indications manual, fifteenth edition, a list running from air or gas embolism to avascular necrosis. It does not appear in the covered-conditions list of Medicare’s National Coverage Determination 20.29, which this publication re-read on the CMS site and which covers, verbatim, everything from acute carbon monoxide intoxication and decompression illness through to diabetic lower-extremity wounds meeting three named criteria. And the 2026 meta-analysis notes that the 2017 European Alliance of Associations for Rheumatology recommendations do not list hyperbaric oxygen as a complementary treatment for fibromyalgia at all.

In most of the world, then, forty sessions for fibromyalgia is an off-label course paid for out of pocket. Brazil’s public system is one of the few places where “should we buy this?” could be asked in earnest rather than rhetorically. The trial supplies the effect size. The economic evaluation, if its methods hold up, supplies the price. Neither substitutes for the missing long-term data, and a favourable ratio computed on a sixteen-week horizon would inherit exactly the weakness this article has been describing.

What the Device Rules Require Either Way

One thing is not in dispute, and it applies with full force here. Hyperbaric oxygen devices are Class II medical devices cleared through the 510(k) pathway under Product Code CBF, and the FDA’s letter to health care providers of 25 August 2025 — still the current version on the agency’s site — sets out the obligations that travel with that clearance regardless of what the chamber is being used for. It references “reports of fires that occurred with HBOT devices that resulted in serious injuries and deaths,” states that the root cause is not known, and asks providers to follow the manufacturer’s instructions for use, maintain fire prevention measures, train and retain trained staff, monitor the patient for the entire duration of treatment, keep to recommended cleaning and maintenance intervals, exclude prohibited electrical or static devices, and dress patients in hyperbaric-compatible material such as cotton.

A chamber treating a fibromyalgia patient in a private clinic carries the same obligations as one treating a decompression injury. The letter carves out no exemption for unproven indications, and none for lower pressures.

What would move this field forward is not another forty-session trial with a two-month tail. It is a trial that randomises to hyperbaric oxygen, to normobaric oxygen at the same flow and schedule, and to no chamber at all — and then follows everyone for a year, with pressure stated in atmospheres absolute and oxygen fraction stated in percent. Until somebody runs it, patients will keep being told the evidence is promising, and payers will keep being told it is insufficient. Both statements are, at present, accurate.

Sources

← Back to blog