The Effects of Hyperoxia on Sea-Level Exercise Performance, Training, and Recovery: A Meta-Analysis

Acute exercise performance can be limited by arterial hypoxemia, such that hyperoxia may be an ergogenic aid by increasing tissue oxygen availability. Hyperoxia during a single bout of exercise performance has been examined using many test modalities, including time trials (TTs), time to exhaustion (TTE), graded exercise tests (GXTs), and dynamic muscle function tests. Hyperoxia has also been used as a long-term training stimulus or a recovery intervention between bouts of exercise. However, due to the methodological differences in fraction of inspired oxygen (F_iO₂), exercise type, training regime, or recovery protocols, a firm consensus on the effectiveness of hyperoxia as an ergogenic aid for exercise training or recovery remains unclear.

The aims of this study were to (1) determine the efficacy of hyperoxia as an ergogenic aid for exercise performance, training stimulus, and recovery before subsequent exercise; and (2) determine if a dose-response exists between F_iO₂ and exercise performance improvements.

The PubMed, Web of Science, and SPORTDiscus databases were searched for original published articles up to and including 8 September 2017, using appropriate first- and second-order search terms.

English-language, peer-reviewed, full-text manuscripts using human participants were reviewed using the process identified in the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement.

Data for the following variables were obtained by at least two of the authors: F_iO₂, wash-in time for gas, exercise performance modality, heart rate, cardiac output, stroke volume, oxygen saturation, arterial and/or capillary lactate, hemoglobin concentration, hematocrit, arterial pH, arterial oxygen content, arterial partial pressure of oxygen and carbon dioxide, consumption of oxygen and carbon dioxide, minute ventilation, tidal volume, respiratory frequency, ratings of perceived exertion of breathing and exercise, and end-tidal oxygen and carbon dioxide partial pressures.

Data were grouped into type of intervention (acute exercise, recovery, and training), and performance data were grouped into type of exercise (TTs, TTE, GXTs, dynamic muscle function), recovery, and training in hyperoxia.

Hedges' g effect sizes and 95% confidence intervals were calculated. Separate Pearson's correlations were performed comparing the effect size of performance versus F_iO₂, along with the effect size of arterial content of oxygen, arterial partial pressure of oxygen, and oxygen saturation.

Fifty-one manuscripts were reviewed. The most common F_iO₂ for acute exercise was 1.00, with GXTs the most investigated exercise modality. Hyperoxia had a large effect improving TTE (g = 0.89), and small-to-moderate effects increasing TTs (g = 0.56), GXTs (g = 0.40), and dynamic muscle function performance (g = 0.28). An F_iO₂ ≥ 0.30 was sufficient to increase general exercise performance to a small effect or higher; a moderate positive correlation (r = 0.47-0.63) existed between performance improvement of TTs, TTE, and dynamic muscle function tests and F_iO₂, but not GXTs (r = 0.06). Exercise training and recovery supplemented with hyperoxia trended towards a large and small ergogenic effect, respectively, but the large variability and limited amount of research on these topics prevented a definitive conclusion.

Acute exercise performance is increased with hyperoxia. An F_iO₂ ≥ 0.30 appears to be beneficial for performance, with a higher F_iO₂ being correlated to greater performance improvement in TTs, TTE, and dynamic muscle function tests. Exercise training and recovery supplemented with hyperoxic gas appears to have a beneficial effect on subsequent exercise performance, but small sample size and wide disparity in experimental protocols preclude definitive conclusions.