Understanding Where Hyperbaric Oxygen Therapy Is Established, Where Evidence Is Evolving, and Why Patient Selection Matters
Hyperbaric oxygen therapy, commonly abbreviated as HBOT, has been used clinically for decades. Its physiologic effects are well established: increased ambient pressure raises oxygen partial pressure, increases dissolved oxygen in plasma, changes the behavior of gas bubbles, and creates cellular signals that may influence inflammation, vascular repair, and tissue healing.
That physiologic foundation does not mean HBOT is equally effective for every diagnosis. Clinical evidence must be evaluated one condition at a time. The treatment objective, patient population, timing, pressure, number of sessions, comparison therapy, and measured outcome differ substantially across hyperbaric indications.
Evidence is relatively strong for some conditions, supportive but mixed for others, and based mainly on observational research or expert consensus in several uncommon emergencies. An evidence-based approach therefore asks more than whether HBOT can increase tissue oxygenation. It asks whether that increase is likely to produce a meaningful clinical benefit for this patient, with this diagnosis, at this stage of treatment.
What Counts as Clinical Evidence for HBOT?
The evidence supporting a medical treatment can come from several sources:
- Randomized controlled trials
- Prospective and retrospective cohort studies
- Systematic reviews and meta-analyses
- Clinical practice guidelines
- Mechanistic and physiologic studies
- Case series and registry data
- Expert consensus
Randomized controlled trials are especially valuable because random allocation helps reduce differences between treatment groups that could distort the results. However, an RCT is not automatically reliable simply because it is randomized. Sample size, masking, patient selection, treatment completion, outcome definitions, and loss to follow-up all affect the strength of the findings.
Hyperbaric trials present additional challenges. Sham treatment may require placing control participants inside a chamber and changing the pressure enough to preserve masking. Pressure changes can produce physiologic effects of their own, making it difficult to create a truly inactive control. Treatment courses are time intensive, and patients may struggle to complete dozens of sessions.
Some accepted indications are also uncommon, rapidly progressive, or life threatening. Randomly withholding recompression from a patient with arterial gas embolism or decompression sickness may be clinically and ethically difficult. In these circumstances, physiologic evidence, historical outcomes, observational research, and specialist consensus carry greater weight than they would for a common elective treatment.
The Undersea and Hyperbaric Medical Society’s 15th edition indications manual identifies 15 clinical categories for therapeutic HBOT. These include gas embolism, carbon monoxide poisoning, decompression sickness, selected traumatic ischemias, refractory osteomyelitis, delayed radiation injury, compromised grafts and flaps, and other defined conditions. The inclusion of an indication reflects an expert review of its scientific and clinical basis, but the type and certainty of evidence are not identical across all 15 categories. (UHMS)
Clinical Recognition Is Not the Same as Insurance Coverage
Clinical guidance and reimbursement policy answer related but different questions.
A professional society evaluates whether a treatment has an acceptable scientific and clinical rationale. A payer determines whether that treatment meets its coverage standards under a specific benefit policy.
Medicare’s national coverage determination lists chamber-based HBOT coverage for defined conditions such as carbon monoxide poisoning, decompression illness, gas embolism, necrotizing fasciitis, crush injury, chronic refractory osteomyelitis, delayed radiation injury, compromised grafts, and qualifying diabetic lower-extremity wounds. Medicare also states that HBOT must remain adjunctive to standard treatment for many of these conditions. (Centers for Medicare & Medicaid Services)
Coverage should not be presented as proof that HBOT will benefit every patient with the listed diagnosis. Conversely, lack of national coverage does not automatically prove that a treatment has no biologic effect. Clinical appropriateness requires individual assessment, while reimbursement requires compliance with the applicable payer policy.
Strong Randomized Evidence in Carbon Monoxide Poisoning
Carbon monoxide poisoning illustrates both the potential value of randomized evidence and the continuing debate that can remain after a positive trial.
Carbon monoxide reduces oxygen transport by binding to hemoglobin and also initiates inflammatory, vascular, and neurologic injury that may continue after the measured carboxyhemoglobin level falls. HBOT accelerates carbon monoxide elimination while producing very high oxygen partial pressures within plasma and tissue.
A landmark double-blind randomized trial assigned symptomatic patients to either three hyperbaric oxygen sessions or a control regimen involving normobaric oxygen and chamber sessions. Cognitive sequelae at six weeks occurred in 25 percent of the HBOT group and 46.1 percent of the control group. Cognitive sequelae were also less frequent in the HBOT group at 12 months. (PubMed)
Other studies of carbon monoxide poisoning have used different protocols and produced less consistent findings. Differences in poisoning severity, treatment timing, oxygen dose, outcome testing, and trial design make direct comparison difficult.
The evidence supports considering HBOT for selected patients with serious carbon monoxide poisoning, particularly when neurologic impairment, loss of consciousness, cardiac injury, severe metabolic acidosis, pregnancy, or other high-risk features are present. Treatment decisions should not be based on the carboxyhemoglobin number alone because the measured level may not reflect the full severity of tissue injury.
Evidence for Delayed Radiation Injury
Delayed radiation injury is one of the most studied chronic applications of HBOT. Radiation may progressively damage small blood vessels, reduce capillary density, produce fibrosis, and leave tissue chronically hypoxic. Repeated hyperbaric exposures are intended to stimulate vascular and tissue remodeling rather than merely provide temporary oxygenation.
The strongest randomized evidence within this category involves chronic radiation cystitis. The multicenter RICH-ART trial found that HBOT improved patient-reported urinary symptoms compared with standard care in patients with late radiation cystitis. (PubMed)
Five-year follow-up published in 2025 found that symptom improvements remained clinically meaningful among many initial responders. Of 70 patients included in the long-term analysis, 48 met the study’s responder definition following HBOT. Some patients underwent another hyperbaric course after symptoms recurred, demonstrating that improvement can be durable without necessarily being permanent for every individual. (PubMed)
Evidence for chronic radiation proctitis is more mixed. One randomized trial found improved healing responses and clinical benefit in patients with refractory radiation proctitis. A later sham-controlled trial involving broader chronic bowel dysfunction after pelvic radiotherapy did not demonstrate a significant benefit in its primary outcomes. (PubMed)
These findings are not necessarily irreconcilable. Radiation-associated rectal bleeding, ulceration, altered motility, pelvic floor dysfunction, and nonspecific bowel symptoms may represent different clinical problems. HBOT may be more likely to help a well-characterized hypoxic radiation injury than a broad collection of gastrointestinal symptoms with multiple possible causes.
The evidence for osteoradionecrosis prevention is also less favorable than historical practice once suggested. The HOPON randomized trial studied patients undergoing high-risk dental procedures in an irradiated mandible. Osteoradionecrosis rates were similar between the HBOT and control groups, which did not support routine preventive HBOT for every post-radiation dental extraction or implant procedure. (PubMed)
The radiation literature therefore supports selective use. Evidence is stronger for chronic radiation cystitis than for routine dental prophylaxis, and patient selection remains essential for rectal, head and neck, breast, and other soft-tissue injuries.
Evidence for Diabetic Foot Ulcers
Diabetic foot ulcers are among the most frequently discussed uses of HBOT, but they also demonstrate why broad claims can be misleading.
A diabetic foot ulcer may be driven by neuropathy, repetitive pressure, ischemia, infection, edema, deformity, and impaired glucose regulation. HBOT addresses tissue oxygenation. It does not independently correct inadequate offloading, arterial obstruction, retained necrotic tissue, uncontrolled infection, or poor metabolic management.
Randomized studies and meta-analyses have reported improved healing in some populations, particularly patients with advanced or ischemic ulcers. Other controlled studies have failed to show a reduction in amputation indications or durable wound-healing benefit. (PubMed)
This inconsistency may reflect meaningful differences among the wounds studied. A superficial neuropathic ulcer with inadequate offloading is not equivalent to a deep neuro-ischemic wound that remains hypoxic despite vascular assessment, debridement, infection management, and appropriate pressure relief.
The 2023 International Working Group on the Diabetic Foot guideline conditionally recommends considering HBOT for neuro-ischemic or ischemic diabetes-related foot ulcers when standard care alone has failed and appropriate resources are available. The recommendation is conditional because the certainty of evidence remains low. (IWGDF Guidelines)
The same guideline system advises against using HBOT solely to treat diabetic foot infection. Infection requires appropriate antimicrobial therapy, drainage, debridement, and surgical source control when indicated. (IWGDF Guidelines)
Medicare takes an intentionally narrow approach, limiting coverage to diabetic lower-extremity wounds that are Wagner grade III or higher and have shown no measurable healing after at least 30 days of standard therapy. HBOT must continue alongside vascular assessment, offloading, debridement, glucose management, nutrition, wound care, and infection treatment. (Centers for Medicare & Medicaid Services)
The most defensible interpretation is not that HBOT works for all diabetic foot ulcers or that it has no value. The evidence supports careful selection of advanced wounds in which hypoxia remains a meaningful barrier after comprehensive standard care has been implemented.
Randomized Evidence in Crush Injury and Lower-Limb Trauma
Severe crush injuries and open fractures may produce edema, microvascular obstruction, tissue hypoxia, and ischemia-reperfusion injury. HBOT is intended to increase oxygen diffusion into threatened tissue while reducing edema and supporting the surviving microcirculation.
The international Hyperbaric Oxygen for Lower Limb Trauma trial randomized 120 patients with severe open tibial fractures to standard trauma care with or without 12 HBOT sessions. The combined primary outcome of necrosis or infection was numerically lower with HBOT but did not reach statistical significance.
When tissue necrosis was examined separately, it occurred in 29 percent of HBOT-assigned patients and 53 percent of controls. Patients assigned to HBOT also experienced fewer late complications and better functional outcomes at follow-up. (PubMed)
The trial supports a clinically meaningful role for early HBOT in selected severe lower-limb trauma, while also showing the importance of reporting primary and secondary outcomes accurately. It would be misleading to state that the study met its combined primary endpoint. It would be equally incomplete to ignore its reductions in necrosis and late complications.
HBOT remains an adjunct in this setting. It must not delay vascular repair, fasciotomy, fracture stabilization, debridement, hemorrhage control, or trauma resuscitation.
Evidence in Necrotizing Soft Tissue Infections
Necrotizing fasciitis and related necrotizing infections create a strong physiologic rationale for HBOT. Affected tissues are severely hypoxic, leukocyte function is impaired, edema compromises the microcirculation, and selected anaerobic organisms or toxins may be inhibited by high oxygen tensions.
Randomized trials are lacking. Most clinical evidence comes from retrospective cohorts, registries, and prospective observational studies. These studies frequently report associations between HBOT and lower mortality or amputation risk, but they cannot completely separate the effect of HBOT from differences in patient selection, surgical timing, hospital expertise, or critical-care resources.
This limitation is particularly important because patients stable enough to reach a chamber may differ substantially from those who are too unstable for transport or treatment.
The evidence supports considering HBOT at experienced centers when it can be delivered promptly after surgical source control and without delaying another operation. It does not justify transferring an unstable patient away from immediate surgery or presenting HBOT as a replacement for debridement, antibiotics, and intensive care.
Refractory Osteomyelitis, Grafts, and Flaps
Evidence for chronic refractory osteomyelitis consists largely of physiologic studies, animal research, clinical series, and observational reports. HBOT may improve oxygen-dependent leukocyte activity, support bone and soft-tissue repair, and complement antimicrobial treatment in hypoxic tissue.
No strong randomized evidence establishes HBOT as a stand-alone cure for bone infection. Its accepted role is adjunctive treatment when infection persists or recurs despite appropriate antibiotics and surgical management. Necrotic bone, abscesses, unstable hardware, and biofilm still require direct evaluation and source control.
Compromised grafts and flaps have a similarly plausible clinical rationale. HBOT can increase oxygen availability in threatened tissue while hyperoxic vasoconstriction may reduce edema. Published clinical experience suggests potential salvage of selected hypoxic reconstructions, but much of the evidence consists of case series and nonrandomized studies.
Urgent surgical evaluation remains the first priority. A thrombosed anastomosis, twisted pedicle, compressive hematoma, or tight dressing requires mechanical correction. HBOT may support tissue that remains viable after those problems have been addressed.
Why Evidence Quality Differs Across Indications
The evidence base is shaped partly by the condition being studied.
Chronic radiation cystitis can be investigated through scheduled, controlled trials using standardized symptom measures. Arterial gas embolism may require treatment within minutes or hours, making randomization difficult. Necrotizing fasciitis is uncommon, heterogeneous, and surgically urgent. Compromised flaps deteriorate quickly, and clinicians may be unwilling to assign salvageable tissue to an untreated control group.
Treatment protocols also vary. Studies may use different pressures, oxygen durations, air-break schedules, treatment frequencies, and total numbers of sessions. A negative trial using one regimen does not automatically disprove every possible hyperbaric protocol, but it should prevent clinicians from assuming that changing the dose will necessarily produce a benefit.
Outcome selection matters as well. A treatment may improve wound closure without reducing major amputation, reduce tissue necrosis without changing infection rates, or improve patient-reported function without changing an imaging result.
Evidence should therefore be interpreted according to the specific clinical outcome the treatment is intended to influence.
Evidence Does Not Support HBOT for Every Proposed Use
The physiologic effects of pressure and oxygen have led to HBOT being promoted for a much wider range of conditions than current clinical evidence supports.
Scientific plausibility is not enough. The ability to influence oxidative signaling, inflammation, mitochondrial function, or tissue oxygen levels does not prove that HBOT improves cognition, athletic performance, general wellness, aging, autism, Alzheimer’s disease, uncomplicated surgical recovery, or other broadly marketed outcomes.
UHMS distinguishes clinical hyperbaric oxygen therapy from low-pressure services that promote treatment for conditions without adequate scientific proof. Therapeutic HBOT requires an appropriate chamber, a medically meaningful pressure and oxygen dose, qualified supervision, and a recognized clinical objective. (UHMS)
Evidence-based practice requires the same scrutiny for HBOT that would be expected for a medication, operation, or medical device. Enthusiasm should not replace outcome data, and uncertainty should not be hidden from patients.
What Evidence-Based HBOT Looks Like in Practice
Clinical evidence becomes useful only when it guides patient care.
An evidence-based hyperbaric plan should identify:
- The precise diagnosis and treatment objective
- The quality of evidence supporting HBOT for that diagnosis
- The standard treatments already completed or still required
- The reason tissue oxygenation is believed to be limiting recovery
- The proposed pressure, oxygen duration, and treatment frequency
- Objective measures that will be used to evaluate response
- Criteria for modifying or stopping treatment
- Patient-specific risks, burdens, and alternatives
A treatment course should not continue automatically because a predetermined number of sessions was initially prescribed. The wound, symptoms, function, infection status, operative plan, or other relevant outcome should be reassessed throughout treatment.
The value of HBOT is greatest when it is used to solve a defined clinical problem within a coordinated care plan. It should complement surgery, wound care, vascular intervention, antimicrobial therapy, oncology follow-up, rehabilitation, and critical care rather than compete with them.
The evidence supporting hyperbaric medicine is neither uniformly definitive nor uniformly weak. It is indication specific. Randomized trials provide meaningful support in selected conditions such as carbon monoxide poisoning, chronic radiation cystitis, and severe lower-limb trauma. Diabetic foot ulcer research supports a selective approach but remains heterogeneous. Several emergency and salvage indications rely more heavily on physiologic evidence, observational outcomes, and expert consensus.
That complexity is not a weakness unique to hyperbaric medicine. It is a reason to practice it carefully. The right question is not whether HBOT is proven in a general sense. The right question is whether the available evidence supports a favorable risk-benefit balance for the individual patient and the clinical outcome being pursued.