Answers are grounded in this paper and cite their sources.
In response to the letter by Professor Philip James,[1] since 1977 the Undersea and Hyperbaric Medical Society (UHMS) has published documents every 3–4 years reviewing the evidence and making recommendations for use of hyperbaric oxygen. The 2019 document referred to by Professor James is currently being revised, with planned publication this year. The aim of these documents is to review the medical evidence for and against use of intermittent short hyperbaric oxygen treatments for selected disease conditions. Professor James suggests the members of the UHMS Committee responsible for these documents are in error about the fundamentals of hyperbaric oxygen and has kindly offered to correct those errors in his letter. Unfortunately we must take issue with some of these “corrections.” Professor James begins by suggesting the use of terms such as “oxygen under pressure” are misleading and suggest the UHMS is implying the oxygen itself is different when delivered in a chamber. Members of the medical profession and other readers of Medical Gas Research are well aware that oxygen is always administered under some pressure, and that the various terms used (“hyperbaric oxygen” or “oxygen under pressure”) refer to administration of oxygen at pressures greater than atmospheric pressure. No one asserts the oxygen itself is any different, only the dose that can be delivered. At such higher doses, there is ample evidence that oxygen can produce a wide spectrum of potentially beneficial effects. While there is undoubtedly a dose-response for hyperbaric oxygen, scientifically supported hyperbaric treatments are usually delivered at pressures between 1.9 to 3.0 atmospheres absolute (ATA); there is presently limited evidence for efficacy of 100% oxygen breathing at lower pressures. “Mild” hyperbaric oxygen therapy is defined as oxygen administration at pressures lower than 1.5 ATA. It is quite possible that 95–100% oxygen at such doses has therapeutic effects and will be demonstrated in future clinical studies. However, during mild hyperbaric treatment clients often receive breathing gas mixes significantly less than 95% oxygen, often via oxygen concentrators or breathing devices such as masks that do not provide a tight seal, thus allowing mixing of gases with chamber air, reducing the oxygen partial pressure.[2] Whether wounds heal in dwellers acclimatized to high altitude with their vascular and mitochondrial adaptations is not relevant to the use of oxygen administered at high ambient pressure as a therapy. By analogy, the fact that vitamin B12 exists in the normal diet in low amounts is not relevant to its use in extreme doses for treatment of refractory vasoplegia.[3] Similarly, oxygen administered at high ambient pressure has several effects that are far removed from its role in maintaining metabolism, including accelerated tissue bubble resolution in decompression sickness, anti-inflammatory effects, inhibition of vascular neutrophil accumulation in ischemic flaps and attenuation of oxidative stress in carbon monoxide poisoning.[4-6] Professor James refers to arterial blood partial pressure of oxygen (PO2) measurements in individuals administered oxygen via face mask ranging from 1.11–1.59 ATA (the correct unit for partial pressure is atmospheres (ATM), not ATA), 850–1140 mmHg. When breathing 100% oxygen the partial pressure difference between oxygen delivered to a patient and arterial blood is due to humidification of the inspired gas, the presence of alveolar carbon dioxide and right-to-left shunt; ventilation/perfusion mismatch has no effect on arterial PO2 while breathing 100% oxygen.[7] Nevertheless, we agree arterial PO2 will vary between individuals and that arterial PO2 in this range has been shown to be therapeutic for several conditions. The usefulness of measuring the exact arterial PO2 in every individual in a clinical trial (let alone in clinical practice) is questionable. Pragmatic clinical trials are designed to closely mimic clinical practice to establish the effectiveness or otherwise of delivering hyperbaric oxygen for a particular condition. The PO2 achieved is at best an interim outcome of no direct value to the patient. While the U.S. Food and Drug Administration (FDA) has approved low pressure chambers for use at altitude to treat mountain sickness, they are usually employed to bring the affected individual functionally back to a lower altitude, hence attenuating hypoxemia, rather than achieving pressures greater than 1 ATA. Whether one refers to this as “hyperbaric” or not, these devices are not approved for any other use, particularly when used in conjunction with oxygen. The FDA recommends that hyperbaric chambers comply with the American Society of Mechanical Engineers document ASME-PVHO-1, Safety Standard For Pressure Vessels For Human Occupancy.[8] Most states also mandate that hyperbaric chambers comply with chapter 14 of the National Fire Protection Association NPFA 99 Healthcare Facilities Code,[9] which most “soft” chambers do not. Contrary to Professor James’ statement, the vast majority of hospitals do not use oxygen concentrators. If oxygen concentrators are used they must be approved by the FDA and prescribed by a physician. Furthermore, while delivery apparatus such as connection tubing (especially if reused) may indeed contain microorganisms, medical grade oxygen does not. Oxygen delivery equipment should only be used under appropriate guidance and is usually not recycled between patients. Professor James’ asserts that FDA decisions are made by non-medical civil servants; we know many of these individuals who are skilled clinicians. Professor James may be correct that there are conditions that could benefit from hyperbaric oxygen therapy but not yet in the UHMS guidelines. A randomized trial of 80 patients with long coronavirus disease is currently underway,[10] the outcome of which, as well as that of other studies may result in routine adoption of hyperbaric oxygen therapy for long coronavirus disease. Professor James’ use of clinical trial data to back his argument suggests our differences in this area are not about the need for clinical evidence, but rather the strength of evidence required to adopt a novel treatment as a routine approach. We believe the maxim primum non nocere demands a higher level of evidence than proposed by Professor James. Often based on conjecture or preliminary observations, medical practice has often gone down paths leading to failure or harm, hence the UHMS will continue to base its guidelines on sound evidence. Professor James’ assertion that controlled trials are not scientific or ethical is outside the realm of sensible procedure.
No takes yet. Share an insight, caveat, or question.
Moon et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: