Evaluation of an extracorporeal ozone-based bactericide system for the treatment of Escherichia coli sepsis
Skorup P, Fransson A, Gustavsson J, et al. Intensive Care Medicine Experimental, 2022.
- Study type
- Animal study
- Animals studied
- 10 animals
- Population
- 10 swine with E. coli sepsis (5 randomized to ozone, 5 to none), plus 6 single-pass runs of E. coli-infected human whole blood
- Published
- 2022, Intensive Care Medicine Experimental
- Identifiers
- PMID 35467176, DOI 10.1186/s40635-022-00443-w
What the authors reported
One pass of E. coli-infected human whole blood through the extracorporeal ozonation prototype lowered viable E. coli by 27% (6 runs). In randomized septic swine (5 ozone, 5 none), 30 minutes of treatment did not change circulatory, respiratory or metabolic measures, bacteria in blood or organs, hemoglobin, leucocytes or methemoglobin; one ozone-group pig died before treatment began.
What limits this result
Not the EBOO procedure: a prototype that cools blood and mixes it with oxygen-ozone gas, run for 30 minutes in a 3.5-hour pig sepsis model. Small feasibility study; the in vitro runs had no ozone-free control, and the authors call the model likely too short to detect improvement. Funded by an unrestricted Sangair AB grant; two authors own or work for Sangair.
How ozone acts on blood and cells, how blood filters work, the laboratory and device-development studies behind the EBOO circuit, and studies that measure markers in the blood rather than how patients fare. None of these shows whether EBO2 helps patients. One study rarely settles a medical question; our guides weigh each paper against the rest of the evidence.
How the study was done
- Design
- Proof-of-concept study of ozone's effect on viable E. coli in human whole blood and of the system's technical feasibility and physiological tolerability
In this proof-of-concept study we hypothesized that ozone would decrease the concentration of viable E. coli in human whole blood and that the system would be technically feasible and physiologically tolerable
- Setting
- Section for Experimental Traumatology, Department of Neuroscience, Karolinska Institutet, Stockholm, Sweden (corresponding author's affiliation)
- Read from
- Read from the full text of the paper. Read on October 2, 2026.
Study arms
| Group | Participants | What they received |
|---|---|---|
| In vitro: E. coli-infected human whole blood | 6 | Treated with ozone in a single pass through the extracorporeal system |
| Swine: E. coli infusion and ozone treatment | 5 | Randomized to E. coli infusion and extracorporeal ozone treatment |
| Swine: E. coli infusion, no ozone treatment | 5 | Randomized to E. coli infusion without ozone treatment |
Source: Abstract on PubMed
Procedure as reported
- Ozone concentration
- < 100 g/m3 (gas flow < 100 mL/min)
- Session length
- 30 min, started after 1.5 h of the sepsis model
Ozone treatment was initiated with a duration of 30 min after 1.5 h.
- Blood flow
- Mean blood flow 49–51 mL/min
- Anticoagulant
- Heparin: 4000 E (66.6 E/kg) IV 15 min before the system started, then 2000 E/h (33.3 E/kg/h) during the 30-min treatment
4000 E heparin (66.6 E/kg) anticoagulation was administered intravenously at 15 min before start of the system, and then by a continuous infusion of 2000 E/h (33.3 E/kg/h) during the 30 min ozone treatment
- Filter
- SangAsept blood ozonation prototype (patent no WO2016/043649)
In the paper's own units, not converted.
What it measured
- Viable E. coli after a single pass (in vitro, human whole blood)
- Decreased by 27%, mean 1941 to 1422 CFU/mL; mean of differences -519.0 (95% CI -955.0 to -82.98, P = 0.0281)
- Blood gases after a single pass (in vitro)
- pO2 increased (95% CI 31.35 to 48.80), pCO2 decreased (95% CI -3.203 to -1.134), oxyhemoglobin increased (95% CI 1.010 to 3.669)
- Methemoglobin after a single pass (in vitro)
- Not affected
- Survival in the swine sepsis model
- Nine of 10 swine survived; one swine randomized to ozone treatment died from septic shock before its treatment started
- Circulatory, respiratory, and metabolic parameters (swine)
- Not affected by the ozone treatment
- E. coli in arterial blood, in organs, and in aerobic and anaerobic blood cultures (swine)
- Did not differ
- Hemoglobin, leucocytes, and methemoglobin (swine)
- Not affected by the treatment
- Lactate (swine)
- The authors note a non-significant rise in lactate
Safety as reported
The authors report the system was manageable in an intensive care setting and caused no signs of physiological adverse effects in circulation, pulmonary artery pressure, or temperature
Source: Full text on PubMed Central
Follow-up
Experiment terminated 90 min after the bacteria infusion had ended
Source: Full text on PubMed Central
Funding and conflicts
Funded by an unrestricted grant from Sangair AB (Sweden)
Source: Full text on PubMed Central
Where we cite this study
Does EBO2 (EBOO) Work? The Evidence, Study by Study
Every study in our EBO2 (EBOO) library in one place: design, size, what was measured and found, its limits, and a check for any study a clinic cites.
EBO2 (EBOO) for Lyme Disease: Clinic Claims, the Missing Evidence, and Standard Care
Nearly half the EBO2 (EBOO) clinics we read list Lyme disease. What they claim, why no study has tested ozone against Lyme, and what CDC lists as care.
What Does the EBO2 (EBOO) Filter Actually Remove?
What the EBO2 (EBOO) filter removes: what papers measured, what is theory borrowed from dialysis, what clinics claim, and what a filtrate lab report must show.