Oxygen toxicity is the most commonly misunderstood HBOT risk. It is real but rare at standard pressures. The drivers — pressure, time, and total dose — are well-mapped from diving medicine and clinical HBOT research.
This guide explains what oxygen-related harm actually is, when it happens, and how clinics manage the risk. It is a clear-eyed safety review, not a deterrent. Most HBOT patients never experience clinically significant oxygen-related harm.
We will cover the two main forms (CNS and lung), the dose-response curves, who is at higher risk, and how protocols limit exposure.
What oxygen-related harm actually is
A working definition. At normal air pressure, the partial pressure of oxygen (PO2) in air is about 0.21 ATA. At HBOT pressures of 2.4 ATA breathing 100% oxygen, PO2 rises to about 2.4 ATA — a 10x jump.
That high PO2 drives the therapy effect (better oxygen flow to starved tissue, germ killing, new vessel growth). It also drives the harm risk. Cells exposed to high PO2 for long enough make reactive molecules that damage proteins, fats, and DNA.
The body has antioxidant defenses like catalase and glutathione. At low oxygen levels, those defenses cope easily. At high enough PO2 and long enough exposure, defenses get overwhelmed and tissue damage results.
CNS oxygen-related harm
The acute form. CNS harm shows up as seizures during HBOT sessions. Onset is usually preceded by a brief warning phase — nausea, tunnel vision, twitching, anxiety, or muscle jerks.
Risk at standard HBOT protocols (2.0 to 2.5 ATA, 90-minute sessions) is roughly 1 to 4 per 10,000 sessions (Plafki et al., Aviat Space Environ Med 2000). At higher pressures (2.8 ATA for severe CO poisoning), risk rises to 1 to 3 per 1,000. See the carbon monoxide poisoning evidence atlas for the full study-by-study evidence breakdown.
The cause involves reactive oxygen species crossing the blood-brain barrier and disrupting nerve signals. The result is a sudden, broad seizure.
CNS seizures during HBOT usually stop once the patient breathes lower-oxygen air. The standard step: the chamber is decompressed slowly while the patient moves to room air. Most patients recover without lasting damage.
Risk factors. Some conditions raise individual CNS risk. Fever, hypoglycemia, and certain drugs (insulin, some antibiotics) all raise the risk modestly.
The UHMS safety guidelines 2023 cover the full risk-factor list.
Pulmonary oxygen-related harm
The chronic form. Lung oxygen harm shows up as cough, chest pain, and reduced lung function. The cause is inflammation of lung tissue from too much oxygen.
Lung harm scales with total oxygen exposure across sessions. The Unit Pulmonary Toxicity Dose (UPTD) is the standard measure. One UPTD equals 1 minute breathing 100% oxygen at 1 ATA, so a 90-minute session at 2.4 ATA equals about 720 UPTDs.
Symptom thresholds vary widely. Generally, exposures above 1,500 UPTDs in 24 hours start to produce mild symptoms (Clark & Lambertsen, Pharmacol Rev 1971). Above 2,500 UPTDs, clinical lung harm becomes more likely.
Standard 40-session protocols at 2.0 to 2.5 ATA generate cumulative exposures well within safe limits when spaced over 8 weeks.
Patients with pre-existing lung disease are at higher baseline risk. COPD, prior chest radiation, and chemo-related pulmonary damage all reduce baseline reserves.
The pressure-time-dose relationship
A clear dose-response pattern. Three variables drive harm risk: pressure, time, and total dose.
Pressure. Risk rises sharply above 2.4 ATA. Standard medical HBOT protocols stay at 2.0 to 2.5 ATA. Single-session emergencies (CO poisoning, decompression sickness) may reach 2.8 to 3.0 ATA, accepting higher risk. See the decompression sickness evidence atlas for the full study-by-study evidence breakdown.
Time per session. Standard sessions are 60 to 90 minutes at depth. Longer sessions raise both CNS and lung risk; the 90-minute window is a working balance between effect and risk.
Cumulative dose. Standard 40-session protocols are well within safe cumulative-dose limits. Long protocols (60+ sessions) push exposure higher, and multi-year users need lung-function checks (Plafki et al., 2000).
Why mild HBOT is lower-risk
The wellness tier at 1.3 to 1.5 ATA carries much lower oxygen-related harm risk. The math is straightforward.
At 1.3 ATA breathing 100% oxygen, PO2 is about 1.3 ATA — well below the 2.4 ATA used in medical HBOT. CNS risk at this pressure is essentially nil. Lung risk per session is also much lower.
This lower risk profile is one reason mild HBOT clinics operate without the UHMS-accredited safety oversight required of hospital programs. The trade-off: lower risk but also lower therapeutic effect for FDA-cleared indications.
For more on the pressure tradeoff, see our mild HBOT vs medical HBOT explainer.
How clinics manage oxygen-related harm risk
Standard protocols include several safeguards.
Air breaks. During longer sessions, patients are given 5-minute air breaks every 20 to 30 minutes. The air breaks reduce both CNS and lung accumulation.
Pressure limits. Standard protocols cap at 2.5 ATA for routine HBOT. Higher pressures are reserved for specific acute indications.
Session count limits. Standard protocols are 40 sessions. Some extended protocols (60 sessions for anti-aging research) are used with caution and monitoring.
Patient screening. Pre-protocol screening covers seizure history, pulmonary status, current drugs, and other risk factors. Some patients are deemed too high-risk for standard protocols and offered lower-pressure or shorter-session alternatives.
What patients should know
A practical checklist. Most patients never experience clinical oxygen-related harm at standard protocols. The known risk factors and management protocols are well-mapped.
Mention any history of seizures, lung disease, or relevant drug use during your screening visit. The clinic should adjust the protocol or decline if risk is too high.
During sessions, report any unusual symptoms — twitching, nausea, vision change, anxiety — immediately. Clinics monitor for these and can transition you to air breathing rapidly.
If you are completing multiple 40-session protocols across years, ask your physician about lung function monitoring. Cumulative-dose tracking matters for long-term users.
How HBOT-fatality cases involve oxygen risk
The rare HBOT fatalities documented in the literature are almost always associated with chamber fire, not oxygen-related harm. The FDA Healthcare Provider letter 2014 covers documented fire incidents.
A small number of CNS-related seizure deaths have been reported, generally in patients with pre-existing seizure disorders or in extreme protocols. Standard protocols at UHMS-accredited centers have an extremely low fatality rate.
For broader safety context, see our HBOT safety guide and our clinic red flags article.
Special populations
Children. Pediatric oxygen-related harm risk is similar to adult risk at equivalent pressures. Pediatric protocols generally use lower pressures (2.0 ATA) than adult protocols (2.4 ATA), which keeps cumulative dose lower.
Patients on chemo. Cisplatin, bleomycin, doxorubicin, and a handful of other chemo drugs raise lung harm risk substantially. The UHMS Indications Manual 2023 covers the no-go list and relative cautions.
Pregnancy. HBOT in pregnancy is reserved for clear medical indications (CO poisoning). The fetal oxygen-related harm risk profile is different from adult risk and not well-characterized for off-label use.
Bottom line
Oxygen-related harm is real but rare at standard HBOT protocols. CNS harm appears as seizures in 1 to 4 per 10,000 sessions at 2.0 to 2.5 ATA. Lung harm scales with cumulative oxygen exposure and is rare at standard 40-session protocols.
Risk rises sharply above 2.4 ATA. Risk falls sharply at 1.3 to 1.5 ATA mild HBOT pressures. The pressure choice is the single biggest harm-risk lever.
For patients pursuing HBOT at any tier, the protective steps are screening, monitoring, air breaks, and pressure limits. Most patients never experience clinically significant oxygen-related harm at standard protocols at UHMS-accredited centers.
Related Reading
- HBOT side effects and safety: the real risk profile
- HBOT side effects: ear pain, fatigue, and eye changes
- Mild HBOT vs medical HBOT: why 1.3 ATA is controversial
- HBOT clinic red flags: signs of unsafe practice
- UHMS-accredited HBOT facilities: what certification means
Frequently asked questions
What is oxygen toxicity in HBOT?
Oxygen-related harm has two main forms: CNS toxicity (seizures during sessions) and pulmonary toxicity (lung inflammation from prolonged exposure). Both stem from reactive oxygen species generated by elevated oxygen partial pressure.
How common are HBOT seizures?
CNS seizures occur in roughly 1 to 4 per 10,000 sessions at standard medical HBOT pressures of 2.0 to 2.5 ATA. Risk rises sharply above 2.4 ATA. Seizures during sessions are usually self-limiting once the patient breathes lower-oxygen air.
Can mild HBOT cause oxygen toxicity?
The risk is much lower at 1.3 to 1.5 ATA. CNS toxicity is essentially nil at these pressures. Lung toxicity per session is also much lower. This lower risk profile is one reason mild HBOT clinics operate with less safety oversight.
What patients are at higher oxygen-toxicity risk?
Patients with seizure history, pre-existing lung disease (COPD, prior chest radiation), or those on certain chemo drugs (cisplatin, bleomycin, doxorubicin) face elevated risk. Pre-protocol screening covers these factors.
How do clinics prevent oxygen toxicity?
Air breaks every 20 to 30 minutes during sessions, pressure limits of 2.5 ATA for routine protocols, session count limits, and pre-protocol screening for risk factors. Standard protocols at UHMS-accredited centers have an excellent safety record.
Medical disclaimer: This article is informational and does not constitute medical advice. HBOT carries real risks including ear injury, oxygen-related harm, and chamber fire. Discuss any HBOT plan with a doctor trained in hyperbaric medicine before starting. The FDA has cleared HBOT for 13 specific uses; uses outside that list are off-label.
-- The HBOT Finder Team