Why some experts are skeptical of small HBOT cohorts

Does scepticism about HBOT mean the treatment doesn't work?

No. The mechanism is coherent, hyperbaric medicine is a legitimate field, and the one randomised trial found real differences on two endpoints. The scepticism is narrower: the studies that exist are too small and too few to support the claims made from them — a criticism of the evidence's size, not of the idea.

Why couldn't the HBOT trial detect a small difference in graft survival?

With only 17 patients per arm, the trial could not distinguish a small effect from chance. Nine-month survival was 96.9% with HBOT versus 93.8% without, about three points, not statistically significant — but detecting a difference that small between proportions near 95% needs hundreds of patients per arm. The trial wasn't built to answer that.

This is the central issue and it is worth understanding, because it applies to almost every adjunct treatment marketed in this field.

Fan and colleagues randomised 34 patients, 17 per arm. Nine-month survival was 96.9% in the HBOT group and 93.8% in the control group — a difference of about three percentage points — and the study did not find it statistically significant.

There are two very different things that result can mean. It can mean there is no real difference. Or it can mean there is a real three-point difference and the study was too small to distinguish it from chance. Detecting a difference that small between two proportions both near 95% requires hundreds of patients per arm, not seventeen. A trial of this size was essentially never going to produce a significant result on that endpoint, whatever the truth was.

This cuts both ways, which is what makes it interesting. The trial does not prove HBOT improves survival. It also does not prove it does not. It simply was not built to answer that question, and quoting the 96.9% figure as evidence of benefit is as wrong as quoting the null result as evidence of no benefit.

Why is the HBOT trial's shedding result more trustworthy than its survival result?

Large differences are detectable even in small samples, while a null survival result is not informative either way. Early shedding was 27.6% with HBOT versus 69.1% without, and itching or folliculitis 11.8% versus 35.3% — both large, credible differences. HBOT probably reduces early shedding and inflammation; whether that means more surviving follicles remains unknown.

The trial's most striking finding was on early shedding: 27.6% in the HBOT group against 69.1% in controls. That is a very large difference, and large differences are detectable in small samples. The same applies to itching and folliculitis at 11.8% versus 35.3%.

So the trial's positive findings are more credible than its null finding is informative. The reasonable conclusion is that HBOT probably does reduce early shedding and post-operative inflammatory symptoms, and that its effect on final graft survival is unknown.

The remaining question about the shedding finding is what it means. Postoperative shedding of the transplanted hair shaft is a normal part of the cycle; the follicle typically survives and regrows. Less visible shedding is a better patient experience. Whether it corresponds to more surviving follicles is exactly the question the survival endpoint was supposed to answer and could not.

What specific methodological objections were raised about the HBOT trial in print?

Dong and Jin's letter raised two points: the trial did not describe whether outcomes differed between male and female patients, which they called 'indispensable' given how much androgen levels differ; and it did not specify hair-loss medication use in the exclusion criteria, despite Finasteride and Minoxidil affecting shedding, the largest-effect endpoint. Neither was resolved.

Dong and Jin wrote to the Journal of Cosmetic Dermatology after the trial was published and raised two concrete points.

• Sex was not analysed. They noted that the article did not clearly describe whether the outcome difference between groups differed between male and female patients, and argued that given the substantial differences in androgen levels this was indispensable for evaluating the study.
• Medication use was not specified. They noted that the use of hair-loss medication was not clearly explained in the exclusion criteria. Since Finasteride and Minoxidil influence shedding, and shedding was the endpoint with the largest effect, this is not a minor omission.

Neither point was resolved in the published exchange.

Why doesn't the five-patient HBOT case report count as strong evidence?

It had no control group. All five patients received HBOT, so scabs clearing in three to five days, 97-99% graft integration, and 2.8-day recovery cannot be attributed to treatment without a comparison — uncomplicated FUE alone often heals just as fast. A case series like this can generate a hypothesis but cannot demonstrate an effect.

The 2025 Cureus report of five patients is often cited alongside the trial as if it were corroborating evidence. It is not, and this is worth being precise about.

All five patients received HBOT. There was no control group. Scabs cleared in three to five days, graft integration was reported at 97-99%, average recovery was 2.8 days and no complications occurred.

Without controls, none of those observations can be attributed to the treatment, because there is no comparison to attribute them against. Scabs clear in a few days, patients recover in a few days, and integration is high after uncomplicated FUE without any adjunct. A case series can generate a hypothesis and describe safety experience. It cannot demonstrate an effect.

What did the healthy-volunteer HBOT study actually find?

Lee and colleagues gave nine healthy volunteers 50 HBOT sessions over three months. They found upward trends in follicle density, hairs per follicle and hair volume, none significant, plus a significant decrease in hair shaft thickness read as early regrowth. The authors call this preliminary — reason for a larger study, not for selling treatment.

What can a fair reader honestly conclude the HBOT evidence supports?

The evidence supports saying HBOT probably reduces early shedding and post-operative itching or folliculitis, based on one small trial. It does not support claims that HBOT improves graft survival, the twelve-month result, necrosis risk or swelling, or helps smokers or diabetics — none of that is demonstrated, and some has not been measured.

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. 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
4. 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
5. Thom SR. Hyperbaric oxygen: its mechanisms and efficacy. Plastic and Reconstructive Surgery, 2011;127 Suppl 1:131S-141S. pubmed.ncbi.nlm.nih.gov/21200283

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.