HBOT and Long COVID: What a 2025 Patient Registry Adds, and What It Cannot
A prospective registry of 232 long-COVID patients reports quality-of-life gains after 40 HBOT sessions. Why registries inform, but cannot replace, controlled trials.
The study
Researchers published a prospective registry in Scientific Reports in 2025 to track patient characteristics and outcomes among people with long COVID who chose to undergo hyperbaric oxygen therapy. The registry included 232 patients and collected patient-reported outcomes at baseline, immediately after treatment, and at three-month and one-year follow-up. The treatment course was 40 daily HBOT sessions.
Registries like this are designed to describe real-world use. They enrol the patients who actually turn up for treatment, rather than a tightly selected and randomised trial sample, which is both their strength and their central limitation. The patients here were enrolled over about a year and were self-selected, in the sense that they had already decided to pursue hyperbaric oxygen before entering the registry.
What was measured, and what was found
The primary outcome was quality of life, measured with the mental and physical component scores of the SF-36 questionnaire. The authors defined a clinically relevant response as a change of at least ten points three months after treatment. On that measure, about 65 percent of participants showed a clinically relevant increase in quality of life at the three-month mark.
Because the registry also followed fatigue and cognitive symptoms, it offers a fuller picture of the patient experience than a single score. But it reports what happened to a group of people who elected treatment; it does not compare them against an untreated or sham-treated control group over the same period. The mental and physical component scores are separate subscales, so a person could improve on one while staying flat on the other - a granularity worth keeping in mind when a single 'quality of life' figure is quoted.
Why the design limits the conclusions
A registry is observational. There is no randomisation and no blinding, so improvement over time cannot be cleanly separated from natural recovery, placebo effects, regression to the mean, or the motivation of people willing to complete 40 sessions. The authors frame their data as hypothesis-generating rather than confirmatory, and that framing is the honest one.
Earlier randomised, sham-controlled trials in post-COVID condition have reported improvements in fatigue, cognition and quality of life, which is why the question remains open rather than settled. Registry data and trial data answer different questions: trials test whether HBOT causes a benefit, while registries describe what happens in routine use.
Safety and practical notes
The registry also recorded adverse events. The most common was barotrauma of the middle ear, reported in 11 patients, and a small number of patients stopped treatment early because their fatigue or other symptoms worsened during the course. These are the tolerability details that matter when weighing a protocol that runs to 40 visits, and they are a reminder that a treatment can be worthwhile for some people while being poorly tolerated by others. The pressures used in clinical HBOT studies are delivered in sealed, pressure-rated hard chambers; for an overview of the equipment categories involved, see our chamber-type overview at HBOT chambers.
Long COVID is a moving research target, and one registry does not change standard care. Anyone considering HBOT for persistent post-COVID symptoms should discuss the current evidence, and their own medical history, with a treating physician before starting.