Sixteen children, five chambers
The paper landed on 18 November in the Journal of Burn Care & Research, and its subject is not hyperbaric medicine at all. It is paediatric frostbite: sixteen patients aged one to seventeen treated at an American Burn Association-verified burn centre in Minneapolis between 2018 and 2026, cross-read against a national readmissions database and a federal storm-events file. The authors counted what each child received. Rapid rewarming in 43.8%. Perfusion imaging in 56.3%. Thrombolytic drugs in 18.8%. Hyperbaric oxygen therapy in 31.3%.
Five of the sixteen went into a chamber. That is the number to hold onto, because frostbite occupies a strange position in hyperbaric medicine: it is a limb-threatening acute ischemia, it is treated with oxygen under pressure at burn centres across the United States and Europe, and it has never been tested in a randomised trial of hyperbaric oxygen. Case reports of frostbite treated in a chamber were already appearing in the surgical literature in 1968. Fifty-eight years later there is still no trial.
What the sixteen charts actually show
Two of the sixteen children lost tissue. Both presented with hemorrhagic blisters, both had delayed care, and neither received thrombolytics. The authors record no thrombolytic-related complications in the cohort. Read the sequence plainly and it points where the frostbite literature has pointed for a decade: at time to care and at reperfusion therapy, not at the chamber.
The paper’s national layer is the more useful part, because it reads the same injury through two federal files and one hospital registry at the same time. The Healthcare Cost and Utilization Project readmission database yielded 2,155 paediatric cold-weather injuries for 2016 to 2021 and an estimated hospitalisation rate of 1.62 per 100,000; 79 paediatric frostbite admissions appeared in the file, 15 of them classed as severe. The NOAA storm-events database added 16 paediatric deaths and 49 injuries spread across 30 states. The institutional series is small, the authors concede, and public health data alone cannot describe the whole burden.
What they conclude is modest: early aggressive management may reduce amputation, and standardised protocols and prevention strategies are needed. Nothing about which adjunct does the work.
The 214-patient series this group published first
Minneapolis is not new to this question. In 2025 the same centre published, in Undersea and Hyperbaric Medicine, a retrospective of 214 severe frostbite patients treated between October 2013 and March 2020 — 62 of whom received hyperbaric oxygen. That is one of the larger hyperbaric frostbite cohorts in the published record.
Its findings are the reason the new paediatric paper is worth reading carefully rather than quoting. Patients who went into the chamber were more likely to receive thrombolytics, had larger areas of tissue affected, were younger, and stayed in hospital longer. A significantly larger proportion of them needed surgery, including amputation and debridement. The authors’ own reading of that is that the difference reflects the severity of the initial injury. Hyperbaric oxygen, in this cohort, was a marker of how badly a patient was hurt rather than a variable that explained the outcome, and the paper says so: the relationship is complex, and controlled trials are needed to establish effectiveness.
Two things follow. First, any claim that frostbite patients “do worse with HBOT” misreads a cohort where the chamber was offered to the sickest. Second, the senior author of that 2025 cohort — indexed as Nygaard RM — is also the senior author of the paediatric study published this week and the first author of a companion paper in the same issue. These are three arms of one programme at one institution, which matters when a signal traces back to a single centre: it is a research agenda, not independent replication.
Two comparisons, opposite signals
The clearest test of hyperbaric oxygen in frostbite is a multicentre retrospective published in the Journal of Foot and Ankle Surgery in 2023, covering January 2016 to August 2021 and matched one-to-one between treated and untreated patients. Across both cohorts the amputation rate was 5.2%. After matching, the authors found no statistical difference in amputation characteristics between patients who received hyperbaric oxygen and those who did not. What they did find was a length of stay of 22.2 days in the treated group against 6.39 days in the untreated group.
That is not proof of harm, and a matched retrospective cannot disentangle a two-hour daily treatment from the fact that clinicians keep patients in house to deliver it. It is, however, the most concrete cost figure in this literature, and it sits in an indication Medicare does not cover.
Against it stands a European number that gets quoted far more often. The SOS-Frostbite programme, run across the hyperbaric centres of Geneva, Lyon and the Mont-Blanc hospitals, prospectively enrolled 28 patients with stage 3 or 4 frostbite treated within 72 hours between 2013 and 2019, all of whom received early hyperbaric oxygen plus iloprost. They were compared with 30 historical patients treated with iloprost alone between 2000 and 2012. Preserved segments per patient came out at 13 against 6, and after adjustment for age and delay to treatment the odds ratio was 45, with a 95% confidence interval of 6 to 335.
The direction is favourable and the confidence interval is the tell. An estimate that wide means the data cannot locate the effect anywhere more precisely than “large”, and the comparison straddles two decades, during which iloprost protocols, imaging and referral pathways all changed. The same paper is a single-arm prospective design against a historical control, which is exactly the design that produced hyperbaric medicine’s other cautionary tales.
What the guidelines do and do not recommend
The 2020 Cochrane review of interventions for frostbite injuries is short reading, because the reviewers found one eligible randomised trial: 47 patients with severe frostbite, all rapidly rewarmed and given aspirin plus intravenous buflomedil, then randomised to buflomedil, to iloprost, or to iloprost plus recombinant tissue plasminogen activator. Iloprost reduced amputations against buflomedil with a risk ratio of 0.05, and iloprost plus rtPA with a risk ratio of 0.31, both on evidence the reviewers graded very low quality. Buflomedil has since been withdrawn from practice. Hyperbaric oxygen is listed among the proposed interventions whose benefits and harms remain unclear, and no trial of it was found. “There is a paucity of evidence regarding interventions for frostbite injuries,” the authors write.
The American Burn Association’s clinical practice guideline on severe frostbite, published in 2024, is built on four themes: rapid rewarming, imaging, thrombolysis and iloprost. It conditionally recommends rewarming in a 38 to 42 °C water bath and thrombolytics for fewer amputations or a more distal level of amputation, conditionally recommends giving thrombolytics within 12 hours of rewarming, and declines to recommend on vascular imaging, on intravenous versus intra-arterial thrombolysis, on iloprost, or on imaging for surgical planning. Hyperbaric oxygen does not appear among its recommendations at all.
Paediatrics has its own answer, and it is newer. A multidisciplinary consensus guideline from a 300-bed tertiary children’s hospital, published in 2025, reviewed 72 children seen between January 2012 and March 2024. Sixty-nine were managed with wound care alone, 15 were admitted, three had grade 3 injuries and none had grade 4. One grade 3 patient treated before the guideline existed did not receive tissue plasminogen activator and eventually required amputation; two treated afterwards received thrombolysis and did not. The guideline’s own account of the two decisive cases turns on thrombolysis. Hyperbaric oxygen appears nowhere in it.
Where hyperbaric oxygen does appear in the frostbite literature, it appears as a case report. A 2026 report from Tasmania describes grade 3 to 4 frostbite in a walker who presented three days after an eleven-day trek in cold, wet, snowy conditions, treated with hyperbaric oxygen alongside vasodilators and rheological agents. The language is careful and worth preserving: the treatment “seemed to reduce the penumbra of ischemia in both feet and thus possibly reduced the level of joint amputation in the toes.” More studies are needed to establish a protocol, the author adds, because the best protocol is currently unknown.
Absent from three lists
None of the fifteen indications in the UHMS’s 15th-edition list is frostbite. The nearest entry is number 04, “Crush Injury, Compartment Syndrome And Other Acute Traumatic Ischemias,” and the word frostbite does not appear anywhere on the society’s indications page. Medicare’s national coverage determination fares no better: NCD 20.29, in the version effective 3 April 2017, limits payment to fifteen named conditions, from acute carbon monoxide intoxication to diabetic wounds of the lower extremities meeting three criteria, and frostbite is not among them. The two entries a billing office might reach for are “acute traumatic peripheral ischemia” and “crush injuries and suturing of severed limbs.”
The clinicians who use it are not inventing an indication out of nothing. The UHMS’s own indications manual carries a chapter on acute traumatic ischemias, and its author, Strauss, opens by describing a constellation that ranges “from crush injuries to compartment syndromes, from burns to frostbite, and from threatened flaps to compromised re-implantations,” on the argument that all of them share a pathophysiology that oxygen under pressure can interrupt. That is a reasonable mechanistic case. It is also the reason frostbite gets treated under an umbrella rather than on a line of its own, and why no regulator has ever had to rule on it.
The regulator’s actual message about these devices is different in kind. The FDA’s provider letter of 25 August 2025 — still the live version — states that the agency “is aware of recent reports of fires that occurred with HBOT devices that resulted in serious injuries and deaths,” that “the root cause of these events is not known,” and that serious adverse events are believed to be rare. Its eight recommendations run to instructions for use, fire prevention, grounding, staff training that is “provided and maintained,” patient monitoring and supervision “for the duration of the treatment,” cleaning, prohibited items and hyperbaric-compatible clothing. The letter says nothing about frostbite, and nothing about any indication.
What the patients said
The companion paper in the same issue is the one that should shape the trial nobody has run. Nygaard and colleagues convened sixteen stakeholders — patients with prior frostbite injury, clinicians, and community partners recruited through Healthcare for the Homeless and the International Frostbite Resource Group — and reported the exercise against the GRIPP2 and SRQR standards. Reducing severe cases and amputations was the most endorsed outcome priority. Survivors named something else as having the greatest impact on their daily lives: cold sensitivity, fine motor function, grip strength, return to activities.
Three themes came out of the transcripts. Barriers to care are multilayered and mutually reinforcing. Patient-centred outcomes extend beyond amputation prevention. Important clinical questions and care needs remain unaddressed. The authors note that no published literature describes a stakeholder-informed approach to frostbite research design, which is a polite way of saying the field has been choosing its endpoints without asking.
The enrolment problem is smaller than it looks. Seventy-nine paediatric frostbite admissions in a national readmissions file covering 2016 to 2021, plus 214 severe cases at one centre over six and a half years, does not describe an orphan disease. The reason frostbite has no hyperbaric trial is not that a trial is impossible. It is that the outcome everyone measures — amputation — is uncommon, that the patients who get the chamber are the ones most likely to lose tissue anyway, and that the outcomes patients rank highest have never been in a protocol. Until one of those three things changes, the 31.3 per cent will keep being a clinical habit with a plausible mechanism behind it, in a condition that neither the society’s indication list nor Medicare’s coverage determination includes.
Hyperbaric oxygen remains a regulated medical therapy with a defined set of recognised indications, and frostbite appears on neither the society list nor the Medicare list checked for this article. Anyone offered a chamber for cold injury should hear that plainly, alongside the fact that the physicians who do offer it are working from a mechanism, a case series and a good deal of clinical conviction rather than from a randomised result.
Sources
- Lumbard DC, et al. Pediatric Frostbite and Cold-Weather Injuries in the United States: A Multi-Database Retrospective Analysis. J Burn Care Res. 2026 Nov 18. doi:10.1093/jbcr/irag190
- Nygaard RM, et al. Partner Consultation to Inform Comparative Effectiveness Research on Frostbite Treatment: A Multi-Method Community Engagement Study. J Burn Care Res. 2026 Nov 18. doi:10.1093/jbcr/irag188
- Masters T, et al. The Role of Hyperbaric Oxygen Therapy for Severe Frostbite Injury: Insights from a Retrospective Cohort at a High Volume Burn Center. Undersea Hyperb Med. 2025;52(2):149-156
- Cooperman SR, et al. Evaluating the Efficacy of Hyperbaric Oxygen Therapy in Digital Frostbite. J Foot Ankle Surg. 2023;62(6):951-957
- Magnan MA, et al. Hyperbaric Oxygen Therapy with Iloprost Improves Digit Salvage in Severe Frostbite Compared to Iloprost Alone. Medicina (Kaunas). 2021;57(11):1284
- Lorentzen AK, Davis C, Penninga L. Interventions for frostbite injuries. Cochrane Database Syst Rev. 2020;12(12):CD012980
- Wibbenmeyer L, et al. American Burn Association Clinical Practice Guidelines on the Treatment of Severe Frostbite. J Burn Care Res. 2024;45(3):541-556
- Schuh JM, et al. A Frostbite Treatment Guideline for Pediatric Patients. J Burn Care Res. 2025;46(6):1437-1443
- Stevens G. Adjunctive Hyperbaric Oxygen Treatment for Grade-Four Frostbite: A Case Study and Short Discussion From Tasmania, Australia. Wilderness Environ Med. 2026;37(1):92-98
- Strauss MB. Acute Traumatic Ischemias: Roles of Hyperbaric Oxygen. Undersea Hyperb Med. 2025;52(3):397-412
- UHMS Hyperbaric Oxygen Therapy Indications, 15th edition (indication list)
- CMS National Coverage Determination 20.29, Hyperbaric Oxygen Therapy
- FDA — Follow Instructions for Safe Use of Hyperbaric Oxygen Therapy Devices: Letter to Health Care Providers (25 August 2025)