What research would actually settle the HBOT question

Why don't we know yet whether HBOT improves hair transplant outcomes?

The only published evidence — one randomised trial, one case report, one preliminary volunteer study and a letter to the editor — was never designed to answer that question, not because it was badly done. Setting out what a genuinely conclusive study would need clarifies exactly what today's evidence is missing.

After one randomised trial, one case report, one preliminary volunteer study and a letter to the editor, we do not know whether hyperbaric oxygen improves the outcome of a hair transplant. That is not because the studies were badly done. It is because none of them was designed to answer that question.

It is worth setting out what would answer it, both because it clarifies what the current evidence is missing and because it gives you a standard against which to judge any new study a clinic cites at you.

How many patients would a trial need to detect a real HBOT effect?

Several hundred patients per arm. Graft survival after competent FUE is already high — 93.8% in the Fan trial's control arm — so any adjunct benefit is a small improvement on an already-high number, and detecting a two- or three-point difference near 95% needs hundreds per arm, not the seventeen the Fan trial had.

This is the binding constraint. Graft survival after a competent FUE procedure is already high — the Fan trial reported 93.8% in its control arm. Any plausible benefit from an adjunct is therefore a small absolute improvement on an already high number.

Detecting a difference of two or three percentage points between two proportions in the mid-90s requires several hundred patients per arm. Fan and colleagues had seventeen. That is roughly two orders of magnitude short, which is why the null result on survival carries so little information.

A trial that could settle the question would need to be multicentre simply to recruit, which brings its own standardisation demands.

What outcome measure would a definitive HBOT trial need to use?

It would need hair counts in a defined, tattooed one-square-centimetre recipient zone rather than a global survival percentage, counted by an assessor blinded to allocation, assessed at twelve months rather than nine since density is still maturing, with early shedding and inflammatory complications tracked as secondary endpoints to confirm the existing findings.

• Hair counts in a defined, tattooed recipient zone, rather than a global survival percentage. A marked one-square-centimetre area photographed and counted at baseline and at follow-up is the standard that hair research uses when it wants a defensible number.
• Counted by an assessor blinded to allocation, ideally from photographs stripped of identifying context.
• At twelve months, not nine. Final density is conventionally assessed at around twelve months, and a nine-month endpoint measures a result that is still maturing.
• With early shedding and inflammatory complications as secondary endpoints, since that is where the existing trial found its effect and it deserves confirmation.

Why does a definitive HBOT trial need a sham-pressure control group?

In the unblinded trial, the HBOT arm attended a facility daily and received more attention than the control arm, which went home — a difference that could explain some of the reported comfort and shedding results. A sham control, pressurised with air instead of oxygen, gives both arms the same schedule, isolating the treatment effect.

This is the design element most obviously missing and the one that would do most for credibility.

In an unblinded trial, the HBOT arm attends a facility daily for a week, is examined daily, and receives a great deal of attention and reassurance during the most anxious phase of recovery. The control arm goes home. Some of the difference in reported itching, comfort and even observed shedding could plausibly arise from that difference in contact rather than from the oxygen.

Sham controls are used in hyperbaric research: the control arm enters the chamber and experiences a modest pressurisation with air rather than the therapeutic oxygen dose, so that both arms have the same schedule, the same enclosure and the same ear sensation. It is not a trivial design to run, but it is established, and without it the attention effect cannot be separated from the treatment effect.

What surgical variables would a definitive HBOT trial need to standardise?

It would need to control the variables Parsley and Perez-Meza's review lists as affecting survival — time out of the body, handling, storage solution, recipient site size and depth, density, and operator skill — through a fixed protocol and a few surgeons, and specify background medical therapy, which the Fan trial did not state.

Graft survival is influenced by a long list of surgical variables — time out of the body, handling technique, storage solution, recipient site size and depth, density, operator skill. Parsley and Perez-Meza's review of factors affecting follicular graft growth catalogues them.

Any trial that does not control these will produce noise larger than the effect it is looking for. That means a fixed surgical protocol, ideally a small number of surgeons, defined graft handling and a defined density in the study zone.

It also means specifying background medical therapy. Dong and Jin's criticism of the Fan trial on exactly this point — that use of hair-loss medication was not clearly explained in the exclusion criteria — is a criticism any future trial should be designed to pre-empt.

What dosing questions does the current HBOT evidence leave unanswered?

Even a successful trial wouldn't resolve dosing: published protocols differ — 2.0 ATA for 60 minutes over seven days versus 2.4 ATA for 90 minutes over six days — and nobody knows which is better, or whether fewer sessions would work as well. No dose-finding work exists for this off-label use.

Even a successful trial would leave the practical questions open, because the two published protocols differ: 2.0 ATA for 60 minutes over seven days, versus 2.4 ATA for 90 minutes over six days. Nobody knows which is better, or whether fewer sessions would do as well.

Heyboer and colleagues make this point about hyperbaric medicine generally, calling for better characterisation of the risk-benefit profile at varying pressures and durations so that protocols can be standardised across institutions. In an off-label cosmetic application, no dose-finding work has been done at all.

Why is a definitive HBOT trial unlikely to happen soon?

Nobody has a financial incentive to fund it: hyperbaric oxygen can't be patented, so there's no pharmaceutical sponsor, public funding for elective cosmetic adjuncts is scarce, and clinics can sell it without one. The expectation is that HBOT keeps being offered as a premium add-on, supported by the same four papers, for a long time.

A trial of several hundred patients across multiple centres with twelve-month follow-up and a sham control is expensive. There is no pharmaceutical sponsor, because there is no product to license — hyperbaric oxygen cannot be patented. Public research funding for elective cosmetic adjuncts is scarce. And the clinics that would host such a trial are already able to sell the treatment without one.

That is an uncomfortable but honest assessment. The most likely future is that HBOT continues to be offered as a premium add-on, supported by the same four papers, for a long time.

Which is why the practical advice does not change: judge the offer against what has actually been shown — less early shedding, less itching and folliculitis, and no demonstrated difference in the final result.

Sources

1. Fan Z, Gan Y, Qu Q, Wang J, Lunan Y, Liu B, Chen R, Hu Z, Miao Y. The effect of hyperbaric oxygen therapy combined with hair transplantation surgery for the treatment of alopecia. Journal of Cosmetic Dermatology, 2021;20(3):917-921. pubmed.ncbi.nlm.nih.gov/32770782
2. Dong X, Jin X. The effect of hyperbaric oxygen therapy combined with hair transplantation surgery for the treatment of alopecia [letter to the editor]. Journal of Cosmetic Dermatology, 2021. doi.org/10.1111/jocd.14131
3. Parsley WM, Perez-Meza D. Review of factors affecting the growth and survival of follicular grafts. Journal of Cutaneous and Aesthetic Surgery, 2010;3(2). jcasonline.com
4. Heyboer M, Sharma D, Santiago W, McCulloch N. Hyperbaric Oxygen Therapy: Side Effects Defined and Quantified. Advances in Wound Care, 2017;6(6):210-224. pubmed.ncbi.nlm.nih.gov/28616361
5. Lee HY, Lee JY, Kim SC, Lee Y. Preliminary Effects of Hyperbaric Oxygen Therapy on Hair Follicle Characteristics in Healthy Subjects. Bioengineering, 2026;13(2):240. doi.org/10.3390/bioengineering13020240
6. Giardiello F, De Medeiros Quirino L, Brigante R, Chumak M. Hyperbaric Oxygen Therapy for Enhanced Postoperative Recovery in Hair Transplantation. Cureus, 2025;17(12):e99635. doi.org/10.7759/cureus.99635

This article summarises published research and standard clinical practice. It is general educational information, not medical advice, and it does not replace the instructions your own surgical team gives you. Where their guidance differs from anything here, follow theirs.