EBO2 (EBOO) for Long COVID: What Clinics Claim and What Has Been Studied

If a clinic has offered EBO2 (also called EBOO) for long COVID, this guide is for checking that offer against the record: what clinics’ own pages claim, what has been studied and how, what regulators have said, and what US health agencies describe as care. The short answer is that no published study has given EBO2 to people with long COVID. The studies that exist tested a different ozone method, the only randomized one was a 73-patient pilot with no sham comparison described, and CDC states that there are no approved treatments for long COVID [3]. A list of questions to take to a clinic, and the searches we ran so the check can be repeated, are near the end. For a description of the procedure itself, see what EBO2 is.

What clinics’ EBO2 pages say about long COVID

We read the EBO2 pages of the clinics in our directory and recorded, with the clinic’s own words, whether each page claims a health benefit and for what. As of October 4, 2026, we could read the pages of 174 clinics. 163 of them claim at least one health benefit, and 48 claim one for long COVID or lingering symptoms after COVID-19. The counts here and in the “From our data” section below are computed from the current records each time the site is built.

Within those 48 clinics:

• Long COVID is rarely claimed on its own. Of the 46 whose recorded wording names long COVID, post-COVID symptoms, or “long haul” COVID, 36 name it in the same sentence or list as other conditions, such as Lyme disease, mold illness, chronic fatigue, or autoimmune disease. 42 of the 48 also claim a benefit for Lyme disease or mold illness, and 37 for autoimmune conditions.
• Most pages do not say whether EBO2 is FDA-approved. 35 of the 48 EBO2 pages do not say whether EBO2 or its equipment is FDA-approved. 10 say EBO2 is not FDA-approved, and 3 say their equipment is FDA-cleared, registered, or approved, which is a statement about a device rather than about the treatment; is EBO2 FDA approved and our guide to EBOO devices and FDA status explain the difference.
• 12 of the 48 pages carry patient testimonials, which report experiences and are not evidence of effect.

We save up to four quotes per clinic, so counts drawn from the wording are a floor. We publish these claims only as counts, never clinic by clinic. A claim on a clinic’s page is the clinic’s statement, not a finding. To compare what clinics say about price, session length, and supervision, see the clinic transparency report; our guide to reading a clinic’s EBO2 page covers the wording to look for.

What the studies in our library tested

None of the post-COVID studies in our research library used EBO2. The six entries our library tags as EBOO studies are listed with their procedures in the “From our data” table below, and none of them reports on people with COVID-19 or long COVID; the 2023 ozone-uptake series among them does not say what its 12 patients were treated for. The three post-COVID studies we hold used major ozone autohemotherapy, in which a fixed volume of blood is drawn into a bottle or bag, mixed with ozone-oxygen gas, and returned through the vein. Two more entries cover COVID-19 itself: an evidence map, in the table, and an umbrella review, described after it.

For COVID-19 itself, rather than its aftermath, the newest synthesis in our library is a 2026 umbrella review of meta-analyses of randomized trials [24]. For COVID-19 it found lower PCR positivity at follow-up, which the authors judged a non-important surrogate outcome, and no significant benefit for hospital stay, intensive-care admission, or death, with the certainty of the evidence rated low or very low [24]. None of the seven meta-analyses it included concerned long COVID [24].

One newer study is not yet in our library. A 2026 open-label study from Murcia, Spain gave rectal ozone insufflation for 12 weeks to 23 people who had had long COVID for more than five months, and reported better handgrip strength and quality-of-life scores at 6 weeks, with no further significant change between weeks 6 and 12 [13]. Its own summary says there was no control group, so the authors cannot say ozone caused the improvements [13].

No study used EBO2, and none compared ozone with a sham or placebo procedure, so the effect of expecting to feel better, and of the attention that comes with repeated visits, cannot be separated from any effect of ozone. None of the papers we read reports adverse events in people with long COVID; the He trial’s full text, which we have not read, may. Our compilation of ozone therapy adverse events covers what has been reported for blood-based ozone methods in other patients.

How EBO2 differs from the method these studies used

In the Kuculmez study each session drew 100 mL of blood into a citrated bottle, mixed it with an equal volume of ozone-oxygen gas, and returned it through the vein within 15 minutes [11]. The Tirelli series treated 150 to 200 mL per session [10]. EBO2 instead circulates blood continuously through an extracorporeal circuit, as what EBO2 is describes, and the one EBOO case report in our library describes about 2 liters of blood processed in each treatment under that clinic’s protocol [14]. A result from autohemotherapy, positive or negative, does not tell us what a procedure that handles ten to twenty times as much blood does. Our comparison of EBO2 and 10-pass ozone covers the other ozone methods.

What regulators have said about ozone and COVID-19

No regulatory record in our tracker addresses long COVID by name. The records that exist concern COVID-19 claims made in 2020, and they set out the standard any health claim is held to.

• FTC, April 2020. In its third set of COVID-19 warning letters, the Federal Trade Commission said the targets included treatments “that may appear more medically sophisticated to consumers, such as acupuncture, intravenous (IV) ‘therapies’ with high doses of Vitamin C, ozone therapy, and purported stem cell treatments” [15]. Three recipients were grouped under ozone therapy, and a fourth, grouped with IV therapies, had ozone therapy among the services named [15].
• FTC letter to a Santa Rosa, California clinic, April 10, 2020. The letter cites the clinic’s “Treating Coronavirus” page and states the legal standard: health claims need “competent and reliable scientific evidence, including, when appropriate, well-controlled human clinical studies” [16]. It continues: “For COVID-19, no such study is currently known to exist for the services identified above” [16]. It also told the clinic to review all its other claims against the same standard [16].
• FDA and FTC, June 30, 2020. A joint warning letter to a clinic in Soquel, California lists, among claims “not supported by competent and reliable scientific evidence”, a weekly injection of “a small amount of your own blood, mixed with saline and ozone” [17].
• FTC, November 2020. Announcing its ninth set of letters, which brought the total to more than 330 recipients, the FTC said of the products and therapies involved, ozone therapy among them, that “currently there is no scientific evidence that these products or services can prevent or treat the disease” [18].
• Department of Justice, April 24, 2020. A federal court in the Northern District of Texas entered an agreed permanent injunction barring a Dallas ozone center and one of its principals from representing that ozone could be used to treat COVID-19 [19]. Court filings alleged the center had called its treatments 95 percent effective; the Justice Department notes these were allegations [19].
• Italy, March 26, 2020. Italy’s national public health institute, the Istituto Superiore di Sanità, answered press reports that it had approved ozone therapy for COVID-19 by saying it had issued no such authorization and that the evidence offered should be confirmed in a trial authorized by the Italian Medicines Agency [20].

Behind all of these sits the FDA’s device rule, which states: “Ozone is a toxic gas with no known useful medical application in specific, adjunctive, or preventive therapy” [21]. The same rule considers an ozone-generating device adulterated or misbranded when it is used “in any medical condition for which there is no proof of safety and effectiveness” [21]. Ozone therapy is not FDA-approved for long COVID or for any other condition.

What CDC describes as care for long COVID

This section describes what CDC tells clinicians; it is information, not a treatment plan. CDC defines long COVID as an infection-associated chronic condition that can follow SARS-CoV-2 infection and lasts at least three months, affecting one or more organ systems [1]. Its possible causes, in CDC’s list, include organ damage from the acute infection, a dysregulated inflammatory state, microvascular dysfunction, persisting virus, autoimmunity, and an inadequate antibody response [1].

On diagnosis, CDC states: “Currently, no laboratory test can be used to definitively diagnose or rule out Long COVID or to distinguish Long COVID from conditions with different etiologies” [2]. On treatment, its page for the public says living with long COVID is harder because “there are no approved tests that can determine if your symptoms or conditions are due to Long COVID and there are no approved treatments” [3]. NIH wrote in October 2024 that “there are no approved therapies to treat Long COVID or its symptoms” [4].

What CDC describes instead is symptom-based care. For most patients, it says, the goal “is to optimize function and quality of life through established symptom management approaches”, including focusing on the symptoms the patient finds most burdensome, a rehabilitation plan, management of underlying conditions, and symptom diaries [2]. It adds that treatment “should be tailored to a patient’s specific symptoms or conditions, including FDA-approved or over-the-counter medications” [2]. CDC singles out post-exertional malaise, the worsening of symptoms “following even minor physical or mental exertion”, which typically appears 12 to 48 hours later and can last days or weeks [2]. It also points clinicians to approaches used for conditions with overlapping symptoms, among them myalgic encephalomyelitis/chronic fatigue syndrome, fibromyalgia, and post-treatment Lyme disease syndrome; our Lyme disease guide covers the last [2].

What NIH’s long COVID trials are testing

NIH launched its RECOVER Initiative in 2021 [4], and its trials show what a test of a long COVID treatment looks like. RECOVER-VITAL randomized 963 people to a 15- or 25-day course of the antiviral Paxlovid or to placebo, with participants, care providers, and investigators all kept unaware of who got what [7]. RECOVER-NEURO enrolled 328 people to test a brain-training program, a goal-management program, and a brain-stimulation device against an active comparison activity and a sham device [8]. RECOVER-AUTONOMIC was designed to assign people with postural orthostatic tachycardia syndrome after COVID-19 at random to intravenous immunoglobulin, the heart-rate drug ivabradine, or placebo [6]. RECOVER-SLEEP compares two wakefulness drugs with placebo, and melatonin and bright light therapy with placebo and low-intensity light [5]. RECOVER-ENERGIZE compares personalized cardiopulmonary rehabilitation with basic exercise education, screening out people with post-exertional malaise, and for those who have it compares structured pacing with basic education about the condition [5]. NIH wrote that “structured pacing is the only intervention used to treat PEM” [5].

None of these trials involves ozone. Their sizes and controls are the contrast that matters: the largest ozone study after COVID-19 had 100 participants and no comparison group [10], and the only randomized one had 73 and no sham [9]. NIH’s next phase, RECOVER-TLC, announced in 2024, says it will assess new ideas and run further trials, and it asked for ideas for therapeutics to be submitted [4]. ClinicalTrials.gov, the registry where these trials are listed, shows which are recruiting.

Questions to ask a clinic that offers EBO2 for long COVID

These are questions, not a verdict on any clinic. Our questions for your doctor tool builds a printable list.

• Which study supports EBO2 for long COVID? If the answer is the Wuhan pilot or the Italian case series, did either use EBO2? (Neither did.)
• How was long COVID diagnosed, given CDC’s statement that no laboratory test can confirm or rule it out [2]?
• What will be measured before and after the sessions, and at what point would the clinic recommend stopping?
• How many sessions are planned, what is the total cost in writing, and what is refunded if sessions are stopped? The session cost planner and price check help with the arithmetic.
• How does the schedule account for post-exertional malaise, if it applies?
• Which anticoagulant is used in the circuit, and has it been checked against current medications? Our guide to EBOO, blood thinners, and medications covers what to ask.
• Who supervises the session, and what happens if a reaction occurs? The contraindications guide lists what clinics screen for, and the side effects guide covers reported reactions.

How to check a clinic’s study citation yourself

We ran these searches on October 2, 2026; anyone can repeat them for free.

1. Europe PMC, the free index that includes PubMed. Searching titles and abstracts with TITLE_ABS:(EBOO OR "extracorporeal blood oxygenation") AND TITLE_ABS:(COVID OR "SARS-CoV-2" OR coronavirus) returned one record, a 2020 review of drug targets for coronavirus infection; no study gave EBOO to people with COVID-19 or long COVID.
2. Europe PMC, ozone and long COVID: TITLE_ABS:(ozone) AND TITLE_ABS:("long COVID" OR "post-COVID" OR "post-acute sequelae" OR PASC) returned 22 records. Four are clinical studies of ozone therapy after COVID-19: the three in the table above and the Murcia study. The rest were a preprint of the Wuhan trial, a letter, reviews, and air-pollution research.
3. Europe PMC, spike protein: TITLE_ABS:"spike protein" AND TITLE_ABS:("ozone therapy" OR EBOO OR autohemotherapy OR ozonation) returned one record when we re-ran it on October 3, 2026: a 2022 review of oxygen-ozone in COVID-19, which did not measure spike protein after a treatment.
4. ClinicalTrials.gov. With “long COVID” as the condition and “ozone” as the intervention, the registry returned no studies. With “COVID-19” as the condition it returned 10, one of them about air pollution, and none with results posted. Two of the 10 concern lasting effects after COVID-19: an inhaled “ozone plasma” study in Mexico, last updated in 2021 [22], and an observational study in Türkiye. “EBOO” as an intervention returned none.

A clinic that cites a study should be able to give its title or PubMed number. The test is whether that study enrolled people with long COVID, used EBO2, and compared it with something.

What we could not verify

• The He trial’s full text, which would show the ozone dose and schedule, what “conventional therapy” included, whether assessors were blinded, and any adverse events. We read the abstract only [9].
• A randomized study of 140 people with post-COVID syndrome, cited in the Kuculmez paper as comparing drug treatment with and without ozone [11]. We have not found or read it.
• A 2024 letter commenting on the use of ozone in post-COVID patients, published in the journal that carried the Wuhan trial. Its authors include authors of the Tirelli series and members of SIOOT, and PubMed holds no abstract for it [23]. We have not read it.
• Whether any clinic tracks outcomes in its long COVID patients. Our dataset records what clinics claim, not results, and we found no clinic outcome data.
• How EBO2 sessions affect people with post-exertional malaise. Nothing we found measured it.

How this guide was made

This guide rests on 24 sources: CDC and NIH pages, ClinicalTrials.gov records, five peer-reviewed studies, an evidence map, an umbrella review, a published letter, and primary regulatory documents from the FTC, FDA, Justice Department, the eCFR, and Italy’s national health institute. The clinic figures come from our own dataset of clinic pages, as of October 4, 2026 (174 readable pages), the searches were run on October 2, 2026, with the spike-protein search re-run on October 3, and the study details come from our study cards, each quoted from the paper. Drafting used AI tools. A human editor has not yet checked this guide claim by claim against its sources, and no clinical reviewer has signed off yet.