Tip · Oxygen

EWOT – Exercise With Oxygen Therapy

Q: Does training with an oxygen mask increase VO2max?

Not in the only matching randomized study. 23 trained cyclists did HIIT three times a week for six weeks under 30 or 21 percent oxygen; VO2max changed by +1.1 ± 3.8 % versus 0.0 ± 3.7 % (p = 0.55; effect size 0.08). Hemoglobin mass, blood volume and mitochondrial capacity did not differ either.

Q: Is my blood really already saturated during exercise?

During moderate cardio at sea level, essentially yes; saturation is close to the maximum. Real desaturation occurs in healthy people mainly in highly trained endurance athletes close to maximal exertion: in a study with 20 well-trained people, SpO2 fell to 89 ± 4 % and 90 ± 3 % respectively at the end of a steep time trial. That is a different situation from 15 minutes of easy cycling.

Q: But surely oxygen helps people with lung disease during training?

That is the most interesting counter-test, and it came out negative. In a multicenter randomized study with 111 COPD patients who demonstrably desaturated during exertion, both groups trained for eight weeks with gas from identical concentrators. Both improved significantly; between the groups there was no difference in either primary endpoint.

Q: When does additional oxygen help at all?

After exercise. A meta-analysis of 19 reports found a small to moderate effect of normobaric hyperoxic recovery on subsequent performance (SMD 0.42; 95 % CI 0.22–0.63; I² = 0 %), while lactate and heart rate showed no clear effect. That is a different application scenario from EWOT, but the only one with a pooled positive finding.

Q: Is it dangerous?

For healthy people in short sessions there is no indication of acute harm, and in a randomized study with 60 women with obesity, markers of oxidative stress did not worsen under oxygen during training. It should be borne in mind that hyperoxia constricts blood vessels: in healthy volunteers, flow velocity and cardiac index fell by more than 3 %. In COPD and asthma, high-dose oxygen is considered a relative contraindication.

Q: Why can’t I find any studies on EWOT?

Because there are none. A search in Europe PMC returns papers on home oxygen, pulmonary rehabilitation and intensive care medicine, but not a single study of the advertised protocol. The available evidence comes from adjacent questions: training under hyperoxia in trained people, oxygen during training in COPD patients, and hyperoxic recovery after exercise.

Training with an oxygen mask. In healthy people at sea level, the blood is already almost fully saturated – human data are lacking.

EWOT is training with an oxygen mask: cardio while you breathe heavily enriched air. The idea assumes that oxygen is the limiting factor during training. Exactly that has now been measured.

In short

There is no study on EWOT as a method. On the neighboring question — training with an increased oxygen fraction in healthy people — there is one randomized study, and it was negative: after six weeks of HIIT at FiO2 0.30, VO2max changed by +1.1 ± 3.8 % versus 0.0 ± 3.7 % (p = 0.55). Hemoglobin mass, blood volume and mitochondrial capacity did not differ either. Even in COPD patients with documented desaturation, oxygen during training brought no advantage over air. A pooled positive finding exists only for oxygen after exercise.

When oxygen is the limiting factor

The oxygen transport capacity of the blood consists almost entirely of the portion bound to hemoglobin; the portion physically dissolved in plasma is small at normal pressure. Anyone who is healthy at sea level already has a saturation close to the maximum on room air. More oxygen in the breathing gas can hardly raise it — that is the difference from the pressure chamber, where only the overpressure produces large amounts of dissolved oxygen.

There are situations with a real shortfall, for instance in highly trained endurance athletes close to maximal exertion: in a study with 20 well-trained people, all developed exercise-induced arterial hypoxemia; SpO2 fell to 89 ± 4 % in the men and 90 ± 3 % in the women at the end of a steep time trial. That is something different from easy cycling.

What was measured

The only randomized study that tests what EWOT claims ran with 23 trained cyclists over six weeks, randomized to FiO2 0.30 or 0.21 during the HIIT sessions. Mitochondrial respiration, oxygen uptake, hematological parameters and an endurance performance were measured. No endpoint differed significantly.

The counter-test in people with a real shortfall came out the same way. 111 people with moderate to severe COPD and exercise-induced desaturation trained three times a week for eight weeks, both groups with gas from identical concentrators — oxygen or air. Both improved significantly; there was no difference between them. The training worked, the oxygen did not.

Where oxygen does something

A shift of a few minutes makes the difference: oxygen in the recovery phase after exercise has a small to moderate effect on subsequent performance. A meta-analysis of 19 reports found a standardized mean difference of 0.42 (95 % CI 0.22–0.63; I² = 0 %) from 16 study entries. The accompanying markers were unremarkable.

In lung disease there is a further signal, though for individual adaptations: in a double-blind study with 28 people with non-hypoxemic COPD, maximal power rose more under oxygen (+0.20 ± 0.03 versus +0.12 ± 0.03 W/kg; p = 0.047), and an increase in quadriceps cross-sectional area (+3.9 ± 1.3 cm²; p = 0.013) appeared only there.

What is well supported

What is robust here is above all a negative finding, and it is well done. The COPD study was multicenter, randomized, with blinded outcome assessment, a genuine sham treatment from identical concentrators and two prespecified primary endpoints: 111 recruited, 97 completing. There was no difference between the groups (Endurance Shuttle Walk Test 15 s; 95 % CI −106 to 136 s; CRQ total score 0.0 points; 95 % CI −0.3 to 0.3) — and that in the most favorable conceivable starting situation for the EWOT hypothesis. Positively supported, by contrast, is hyperoxic recovery after exercise, with SMD 0.42 (95 % CI 0.22–0.63).

What the studies show

Six weeks of HIIT under hyperoxia (Cardinale et al. 2019)

23 trained cyclists (35.3 ± 6.4 years, VO2max 4.5 ± 0.7 l/min), six weeks, randomized to FiO2 0.30 (n = 12) or 0.21 (n = 11) during the HIIT sessions. VO2max +1.1 ± 3.8 % versus 0.0 ± 3.7 % (p = 0.55; effect size 0.08), no differences in blood volume, hemoglobin mass and mitochondrial capacity. Only in endurance performance a small, non-significant advantage (+6.0 ± 3.7 % versus +2.4 ± 5.0 %; p = 0.073).

Oxygen during training with documented desaturation (Alison et al. 2019)

111 people with COPD and exercise-induced desaturation (60 men, 69 ± 7 years), 97 completing. Both groups received their gas at 5 l/min via nasal cannula, three times a week for eight weeks. Primary endpoints were the time in the Endurance Shuttle Walk Test and the CRQ total score. Within both groups, both endpoints improved significantly (p < 0.01); between the groups they did not.

Hyperoxia during recovery (Liu et al. 2026)

Systematic review with meta-analysis, 19 reports, 21 study entries. For subsequent performance, 16 entries yielded an SMD of 0.42 (95 % CI 0.22–0.63; I² = 0 %). Blood lactate without a clear effect (SMD 0.14), heart rate likewise not; the SpO2 results were so heterogeneous (I² = 90.0 %) that a pooled analysis was not performed.

Where the data stop — and where they say zero

For “exercise with oxygen therapy” as a method, a search in Europe PMC returns hits on home oxygen, pulmonary rehabilitation and intensive care medicine, but not a single paper on the advertised protocol of 15 minutes of moderate cardio at 90 to 95 % oxygen. The term comes from the vendor market, not from the literature. The randomized study in healthy people also worked with FiO2 0.30 and thus below the advertised values.

Three target measures are named: VO2max, tissue oxygen saturation and energy. The first was measured and remained unchanged, as did hemoglobin mass, blood volume and mitochondrial density — not a missing proof but a measured null effect. The second was never assessed as an endpoint in healthy people. The positive individual findings come from populations with a real shortfall: five subjects at 1,600 m altitude without a control group, and 28 people with COPD.

How to do it

What has actually been studied looks different from what is on offer. In healthy people: six weeks, three HIIT sessions per week plus two at low intensity, FiO2 0.30 during the hard sessions — without a difference in VO2max, hemoglobin mass and mitochondria. In COPD: 5 l/min via nasal cannula, three times a week, eight weeks — without a difference from the air group. At 1,600 m altitude: six weeks at 95 % of maximal power with at least 70 % oxygen, five subjects without a control group; there, endurance time rose from 6.2 to 8.2 min (p = 0.012). The only variant with a positive pooled finding is oxygen after exercise.

Safety

For the method itself there are no safety data, because there are no studies on it. Two things can be said from the adjacent literature. Hyperoxia constricts blood vessels: in ten healthy volunteers, regional cerebral oxygen saturation rose by more than 6 % (p = 0.0001), while flow velocity and cardiac index fell by more than 3 %. More oxygen in the breathing gas therefore does not mean more blood flow. And in chronic CO2 retention, high-dose oxygen is not a harmless extra — COPD is considered a relative contraindication because of oxygen-induced hypoventilation, as is asthma. On oxidative stress there is an all-clear: in a randomized study with 60 women with a BMI over 30, the markers did not worsen under oxygen. Anyone who experiences shortness of breath, dizziness or chest pain during exertion needs a medical evaluation, not an oxygen mask.

BK-Score Not studied in humans

Human evidence		1
Mechanism		3
Safety data		5
Hype gap		2
Track record of use		4

On EWOT itself there are no data, but there are two randomized studies on the neighboring question – training under hyperoxia in healthy people and in people with exercise-induced desaturation – and both came out negative. The claimed mechanism presupposes an oxygen shortfall that does not exist in healthy people at sea level.

What is rated is the state of knowledge, not the effect. “Safety data 9” means well studied – not harmless.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)

Frequently asked questions about EWOT – Exercise With Oxygen Therapy

Does training with an oxygen mask increase VO2max?