Hyperbaric Oxygen for Long COVID: Promising Israeli Results Meet Skeptical Swedish Data

Holographic human brain above a hyperbaric oxygen chamber flanked by two contrasting data charts, representing conflicting long COVID trial results

For millions of people worldwide, a COVID-19 infection was not the end of the story. An estimated 10 to 30 percent of non-hospitalized patients—and up to 70 percent of those who required hospitalization—continue to experience debilitating symptoms months or even years after their initial infection. Fatigue, brain fog, memory problems, sleep disturbances, and chronic pain define what the World Health Organization calls “post-COVID-19 condition,” better known as long COVID. With conventional treatments offering limited relief, hyperbaric oxygen therapy (HBOT) has emerged as a candidate intervention. But the evidence is anything but settled.

The Israeli Success Story

At the Sagol Center for Hyperbaric Medicine and Research at Shamir Medical Center in Israel, researchers led by Dr. Shai Efrati have produced some of the most encouraging data on HBOT for long COVID. Their randomized, sham-controlled, double-blind trial, published in Scientific Reports in July 2022, randomized 73 patients to receive either 40 daily HBOT sessions or a sham treatment over two months.

The results were striking. Patients who received genuine HBOT showed significant improvements in global cognitive function, attention, and executive function compared to the control group. The benefits extended beyond cognition: participants reported better energy levels, improved sleep quality, reduced psychiatric symptoms including depression and anxiety, and decreased pain interference. Brain imaging revealed increased cerebral blood flow and microstructural changes in regions associated with cognitive and emotional processing—the left and right supramarginal gyrus, left supplementary motor area, right insula, and frontal precentral areas.

Perhaps most importantly, a follow-up study published in February 2024 tracked these patients for more than a year after their final HBOT session. The improvements persisted. Quality of life scores, sleep quality, psychiatric symptoms, and pain measures all remained significantly better than baseline a year later. The researchers concluded that HBOT induces lasting neuroplasticity and tissue repair, not merely temporary symptom relief.

The Swedish Contradiction

Then came the Swedish HOT-LoCO trial, published in BMJ Open in April 2025. This randomized, placebo-controlled, double-blind phase II study took a different approach. Rather than 40 sessions, the 80 participants received just 10 HBOT sessions over six weeks. The result: no statistically significant advantage over sham treatment.

The contrast with the Israeli findings could hardly be starker. Where the Shamir study found medium to large effect sizes across multiple domains, the Swedish trial found no meaningful difference between groups. The authors noted that their shorter protocol—10 sessions versus 40—may explain the discrepancy. “These treatments are really meant to be a Monday-through-Friday daily session,” Dr. Craig Lindsey, an emergency medicine specialist who runs a hyperbaric center in New Mexico, told Scientific American. “That’s where you get the benefit.”

What the Dutch Registry Reveals

Adding another layer of complexity, a prospective registry study from six hyperbaric centers in the Netherlands, published in Scientific Reports in August 2025, tracked 232 long COVID patients who underwent HBOT outside of a randomized trial. The results were nuanced: 56 to 63 percent of patients showed clinically meaningful improvement in mental and physical quality-of-life scores three months after treatment. But 13 to 19 percent experienced clinically significant deterioration.

The Dutch researchers emphasized that patient selection matters. Those with post-exertional malaise—the worsening of symptoms after even minor physical or mental exertion common in long COVID—may be at higher risk for adverse effects. The treatment requires patients to spend 90 minutes daily in a pressurized chamber, breathing against elevated pressure. For patients whose condition is characterized by exertion intolerance, this physical demand can potentially cause more harm than good.

Why the Discrepancy?

Several factors may explain the divergent results across studies. The most obvious is dosing: 40 sessions in the Israeli protocol versus 10 in the Swedish trial. HBOT’s proposed mechanisms—improved tissue oxygenation, reduced inflammation, mitochondrial restoration, and angiogenesis—may require sustained treatment to manifest clinically meaningful benefits.

Patient selection also varies. Long COVID is not a single disease but a syndrome with multiple potential underlying mechanisms: viral persistence, immune dysregulation, microvascular damage, mitochondrial dysfunction, and neurological injury. HBOT may benefit patients with specific pathophysiological profiles—particularly those with vascular dysfunction or neuroinflammation—while offering little to those with different dominant mechanisms.

Timing matters too. The Israeli study included patients with ongoing symptoms for at least three months after confirmed infection. The Dutch registry had a median time from infection to treatment of 20 months, selecting for patients whose symptoms had not resolved spontaneously. Natural recovery from long COVID appears to plateau after 6 to 12 months, so patients treated earlier may have different trajectories than those with established chronic symptoms.

The FDA Position: Not Approved for Long COVID

Amid the research debate, regulatory clarity is essential. The U.S. Food and Drug Administration has not cleared or approved HBOT for long COVID. The agency’s list of FDA-cleared HBOT indications includes decompression sickness, carbon monoxide poisoning, certain non-healing diabetic wounds, late radiation tissue injury, gas gangrene, severe anemia, certain skin grafts and flaps, crush injury, thermal burns, idiopathic sudden sensorineural hearing loss, and intracranial abscess. Long COVID is conspicuously absent.

The FDA explicitly warns that HBOT is not proven effective for conditions including Alzheimer’s disease, autism, cancer, Lyme disease, multiple sclerosis, stroke, or heart disease—categories that capture many off-label wellness uses. A 2025 FDA letter to healthcare providers additionally highlighted fire and safety risks associated with hyperbaric devices, particularly when used outside accredited medical facilities.

The Undersea and Hyperbaric Medical Society (UHMS), which accredits hyperbaric facilities, also does not recognize long COVID as an approved HBOT indication. However, the American Academy of Physical Medicine and Rehabilitation lists HBOT as an “emerging therapy” for long COVID, reflecting the evolving evidence base.

Access and Equity Concerns

Even for patients who wish to pursue HBOT, access presents formidable barriers. Without FDA approval or UHMS endorsement, U.S. insurers and Medicare do not cover HBOT for long COVID. A typical hospital-based hyperbaric session costs roughly $5,000; a full 40-session course approaches $200,000. This creates what researchers call “significant equity concerns,” with access largely limited to those who can self-fund treatment.

Logistical challenges compound the problem. The United States has approximately 1,000 hospital-based hyperbaric facilities—nowhere near enough to serve the millions living with long COVID. Treatment requires daily sessions for weeks, demanding substantial time commitment and geographic proximity to a facility.

Unregulated “medical spas” offering soft-shell chambers at lower pressures present another risk. These facilities often operate below therapeutic pressure thresholds and may be staffed by personnel without hyperbaric medicine certification. As Dr. Lindsey warned, “They are not certified, and that makes it dangerous as well as ineffective.”

Looking Forward

Sandra Wainwright, a hyperbaric physician at Yale New Haven Health System and incoming president of UHMS, plans to form a committee to systematically review evidence for HBOT’s brain-related indications, including long COVID. If UHMS endorses the therapy, it could accelerate acceptance. But even with professional society endorsement, widespread insurance coverage could be five to ten years away without large phase III trials conducted under FDA protocols—and currently, no one is funding such trials.

For now, the evidence for HBOT in long COVID remains promising but preliminary. The Israeli studies demonstrate that, for some patients, a 40-session protocol can produce meaningful, lasting improvements in cognition, fatigue, sleep, and pain. The Swedish trial suggests that shorter courses may be insufficient. The Dutch registry confirms that benefits are real for many but not all patients, with a notable minority experiencing worsening symptoms.

The path forward requires better patient stratification—identifying which long COVID patients have the vascular and inflammatory profiles most likely to respond to HBOT—and standardized, adequately powered randomized trials. Until then, patients and physicians must weigh the potential benefits against the significant costs, access barriers, and safety considerations that attend this off-label use of a powerful medical therapy.

Sources

← Back to blog