Does EBO2 (EBOO) Work? The Evidence, Study by Study

EBO2 (also called EBOO) has a small research record. Of the 78 studies in our library, six are about EBOO as given to people: one randomized trial of 28 patients run at the hospital where the method was developed, two single-patient case reports, a preliminary report that gives no patient count, a review by the developers, and a series that measured ozone uptake in 12 patients at a California clinic [1][2][3][4][6][7]. None is a large or independent trial. On October 2, 2026, ClinicalTrials.gov listed no registered study of the procedure under any of the names we searched [12]. That does not show that EBOO does nothing. It shows that the question has barely been tested. This guide lays out every study in our library with what it measured and found, sets clinic claims beside them, and ends with a check anyone can run on a study a clinic cites. Ozone therapy is not FDA-approved for any condition, and the FDA’s regulation calls ozone “a toxic gas with no known useful medical application in specific, adjunctive, or preventive therapy” [13].

What counts as a study of EBOO itself

Our research library files each paper in one of four groups: studies of EBO2 or EBOO as given to people, ozone therapy given some other way, safety reports, and background science. The grouping matters because the procedures differ. Major autohemotherapy usually draws about 200 mL of blood into a bag and mixes it with ozone [7], while the EBOO developers described treating up to 4,800 ml of blood in one hour of circulation [3]. A result for one does not carry over to the other. The studies compared report sorts the same library by design.

Two papers sit near the boundary. A 2023 series from a California clinic that calls the procedure “ozone dialysis” measured how much ozone blood absorbed during routine sessions, not how patients fared [7]; the paper says the method is “commonly referred to as” EBOO, and the library files it with the EBOO studies because it measured the procedure as given to patients. Our EBOO vs dialysis guide compares that circuit with kidney dialysis. A 2007 test of the developers’ gas exchanger in a saline solution, with no blood and no patients [5], is filed as background science with the rest of their equipment work, described below. The table of procedure settings in the “From our data” section is built from the EBOO group.

The studies of EBOO itself

“Reports” means what the paper’s text says, as our study notes record it. Each study name opens our notes, with the exact wording and the page it came from.

The largest number of patients in any study of EBOO we found is 28 [4]. Four of the six come from the nephrology and physiology departments in Siena, Italy, where the method was built [1][2][3][4]; the other two were written by practitioners about their own patients, a 2023 series by the doctors who run a Santa Rosa, California clinic [7] and a 2025 case report by clinicians at a New York pain-management practice [6]. The 2005 review said that clinical trials “are under way to validate this new technique further” [3]. Apart from the 28-patient trial, published later that year [4], we found no trial from that group.

The same Siena group published the work that built the circuit, which our library files as background science. It tested equipment in saline, pig blood, and sheep and found dialysis-type membranes “unsuitable” [8]; compared several oxygenators before choosing one [9]; tested a new gas exchanger in a buffered saline solution, which it described as “very versatile and efficient” [5]; and measured ozone transfer through four dialysis filters against a purpose-built gas exchanger [10]. None measured how patients fared. Two recent animal studies of a different extracorporeal ozonation prototype, in pigs with sepsis, are listed with the background science below [53][54]. In a 2011 paper, Bocci and two colleagues described EBOO as “a procedure to be used in emergency conditions such as PAD stage 4” (PAD is peripheral arterial disease), citing its “complexity and invasiveness” [11].

What clinics claim, next to what was studied

Our clinic directory records what each clinic states about its own EBO2 service. As of October 4, 2026, 163 of the 174 clinics whose EBO2 page we could read state a health benefit, and 26 of those 174 publish patient testimonials there. The table sets the kinds of benefit clinics name beside the studies of EBOO itself. Our rule: a study counts against a claim only if it measured that kind of outcome in patients. The review and the ozone-uptake series count for none.

The claim counts come from our clinic data, which the “From our data” section recomputes each time the site is built. The matching of studies to claims is our own judgment. Our cancer claims guide looks at one category in detail.

Ozone therapy given other ways

These studies used injections, ozonated oil, or major autohemotherapy, not an EBOO circuit. Some are larger and better controlled than anything on EBOO, but the route and dose differ.

Safety reports and background science

Neither group tests whether EBOO helps anyone. The safety reports describe harms after other ozone routes. Most followed injections at or near the spine, with gas embolism or spine infection afterward; others followed ozonated blood or gas given into a vein. Our ozone therapy adverse events guide and side effects and safety guide discuss these harms in more depth.

How to read the study behind any EBO2 claim

These ten checks work on any paper a clinic cites. PubMed (pubmed.ncbi.nlm.nih.gov) and Europe PMC (europepmc.org) show the same records, and both are free.

1. Find the paper itself. Search the title the clinic gives. If a clinic cites no paper, ask which one supports the claim. A claim with no paper behind it cannot be checked.
2. Check what was given. EBOO, major autohemotherapy, an injection, and an oil are different procedures. The abstract’s methods sentence says which.
3. Read the publication type. PubMed lists it under the abstract: “Randomized Controlled Trial,” “Case Reports,” “Review.” A review or a bench test adds no patients.
4. Count the people and look for a comparison group. A before-and-after change in one group cannot be told apart from what would have happened anyway. The 2005 trial is the only study of EBOO in our library with a comparison group [4].
5. Check what was measured. A symptom score, a lab marker in blood or urine, and a property of the device are three different things. The 2025 case measured urine [6]; the 2007 study measured gas transfer in saline [5].
6. Check who judged the outcome. Look for “blind” or “masked” in the methods. None of the EBOO papers we read reports blinded assessment [1][2][4][6].
7. Check the follow-up. A result measured the day after a session says nothing about a month later.
8. Check who wrote and paid for it. The affiliations line and the conflict-of-interest statement say who the authors are. Developers studying their own method, or clinicians studying their own patients, are worth noting.
9. Check for notices. PubMed shows lines such as “Erratum in” or “Retraction in” near the top of a record. PubMed lists none for the six EBOO studies [1][2][3][4][6][7].
10. Check whether anyone repeated it. Search the title words on Europe PMC, and search ClinicalTrials.gov (clinicaltrials.gov) for a registered trial. Our searches there on October 2, 2026 for “extracorporeal blood oxygenation and ozonation,” “EBOO,” “EBO2,” “ozone dialysis,” and “recirculatory hemoperfusion” each returned no study [12]. A broader search for “extracorporeal ozone” returned only injection, shockwave, oil, and autohemotherapy trials [12].

A claim and a study can share words and still not match. A claim that the filter removes heavy metals is a different statement from a paper that measured metals in one patient’s urine [6].

Why people report feeling better

People who say they felt better after EBO2 are usually describing something real, and a report like that still cannot show what the procedure itself did. Symptoms that come and go tend to drift back toward their usual level, people often change diet, sleep, or supplements at the same time, and an elaborate, expensive procedure carries strong expectations. People who felt no change are also less likely to write a testimonial. The National Center for Complementary and Integrative Health puts the general point this way: evidence from research studies “is stronger and more reliable than something you’ve seen in an advertisement or on a website, or something someone told you about that worked for them” [14]. It suggests asking whether there is “scientific evidence (not just personal stories)” behind a claim [14].

What would change the picture

A randomized trial of EBOO with a sham or standard-care comparison, outcome assessors who do not know who was treated, a prespecified clinical outcome, and authors with no stake in the procedure would change the picture. So would a registered trial that publishes whatever it finds, or a clinic registry that tracked outcomes and harms across many patients. None of these exists in the sources we found.

What we could not verify

• We could read only the abstracts of four of the six EBOO studies, because the full texts are behind publishers. Arm sizes, randomization, blinding, ozone settings, and adverse-event methods in the 2005 trial are unknown to us [4].
• The developers’ figure of “more than 1200 treatments performed in 82 patients” [3] appears in no patient-level paper we found.
• The ozone-uptake series reports “at least 400 sessions” at the authors’ clinic “with no untoward effects observed” [7]. That is the practitioners’ own statement, not a systematic safety record.
• We searched ClinicalTrials.gov only. We did not search the WHO trial registry platform or European registers.
• Matching studies to clinic claim categories is our judgment, and the claim categories come from our reading of clinic pages, not from the clinics.
• New papers are added to our library as we read them. The counts here describe the library on October 2, 2026.

How this guide was made

This guide rests on 82 sources: 79 papers listed on PubMed, the FDA’s ozone regulation, an NIH page, and a ClinicalTrials.gov search we ran on October 2, 2026. The claim and testimonial figures come from our own dataset of clinic pages, as of October 4, 2026. Drafting used AI tools. A human editor has not yet checked this guide claim by claim against its sources. No clinical reviewer has signed off on this guide yet.