Oxygen is an important substance for maintenance of human life activities and plays a key role in daily life, especially during exercise (Hill and Flack, 1910). To the best of our knowledge, athletes constantly need to improve their performance to continue excelling in their chosen disciplines. Thus, athletes are constantly required to perform continuous high-intensity technical movements involving shifting and changing directions during training and in competitions (Kellmann et al., 2018). The prolonged and intense oxygen utilization by tissues and organs causes hypoxia in the internal environment of the body, which greatly mobilizes the glycolysis energy supply (Carreau et al., 2011), leading to the accumulation of lactic acid in muscle or blood and depletion of glycogen in muscle (Bassett and Howley, 2000). Likewise, since oxygen delivery and utilization are influenced by hypoxia, exercise performance and recovery can lead to negative developments such as overtraining syndrome, injuries, or illnesses (Kellmann et al., 2018). Therefore, different interventions have been developed to improve performance and recovery, including nutrition therapy (Lynch et al., 2018), oxygen therapy (Sperlich et al., 2017), and cryotherapy (Rose et al., 2017). Among these, oxygen therapy has been applied as an ergogenic aid to enhance performance and accelerate recovery after exercise due to the importance of oxygen in the aerobic energy system (Hodges et al., 2003), and the low tolerance of human tissues and organs to hypoxia (Millet et al., 2010). Notably, according to the international standard prohibited list of the World Anti-doping Agency (WADA), methods for oxygen inhalation and replenishment are not classified under doping, and therefore can be applied to sports practice (Sperlich et al., 2017). With advancements in technology, the equipment and configurations used in different oxygen-therapy methods have been continuously evolving, allowing their use as auxiliary techniques in different stages of training and competition to improve the effects of training and the physical and physiological conditions of athletes as well as prevent exercise-induced fatigue.

In the field of sports science, various oxygen-therapy methods have been studied and applied to date, including hyperoxic gas supplementation, hyperbaric oxygen (HBO₂) therapy, and micro-pressurized oxygen therapy (Sperlich et al., 2017; Ishihara, 2019). Among these, HBO₂ therapy is defined as a treatment in which 100% oxygen is administered under a pressure greater than 1 atm absolute (ATA), and the patient breathes intermittently (Hodges et al., 2003; Branco et al., 2016). HBO₂ therapy is not only beneficial in recovery from sports injuries and other injuries, but also contributes to an improvement in the body’s sports state and functional level (Sperlich et al., 2017; Ishihara, 2019). During long and intense training sessions, the increasingly pronounced hypoxia and the resultant variations in arterial oxygen saturation (SaO₂) necessitate additional cardiorespiratory effort to compensate for the reduced delivery and utilization of oxygen and thereby maintain performance and muscle activity (Wolfel et al., 1991). Under these conditions, HBO₂ therapy can increase the amount of dissolved oxygen in arterial plasma, potentially accelerating recovery (Powers et al., 1993; Peltonen et al., 1995). In addition, under conditions of exercise-induced fatigue, HBO₂ therapy can increase the oxygen supply to the skeletal muscle system, which may activate cell activities, increase the synthesis of adenosine triphosphate, and promote the metabolic clearance of fatigue-inducing substances (Sperlich et al., 2017). Nevertheless, there is no consensus on the practical application effect of HBO₂ therapy on performance and recovery, and the underlying mechanisms of action of this therapy require further research.

Therefore, this study systematically evaluated the effects of pre-, post-, and intra-exercise HBO₂ therapy on performance and recovery, critically summarized the studies on the use of HBO₂ therapy to improve performance and recovery, and provided an evidence-based basis for the application of this method in exercise practice.

A total of 1,712 identified records were screened within the five databases, of which 1,387 duplicates were removed, and 17 articles with full text available were potentially assessed for eligibility. We reviewed the full text of the manuscripts according to the predetermined inclusion and exclusion criteria and included 10 studies enrolling 166 participants in the qualitative analysis and six studies with 69 participants in the quantitative synthesis (meta-analysis). The low number of studies in the quantitative synthesis was primarily due to variations in therapy phases (e.g., pre-exercise, post-exercise, intra-exercise) and outcomes (e.g., blood lactate concentration, creatine kinase, lactate dehydrogenase, maximum oxygen uptake, peak power output, etc.), which prevented meta-analysis. The PRISMA diagram in Figure 1 summarizes the results of the screening and selection process of the study.

To our knowledge, this is the first systematic review and meta-analysis to evaluate the effectiveness of HBO₂ therapy in improving exercise performance and recovery in healthy adults. This systematic review included 10 studies with 166 participants that examined the effects of HBO₂ therapy on exercise performance, and its purpose was to identify the effects of therapy administered in different phases. The principal findings of the review were as follows: (1) pre-exercise HBO₂ therapy appears to have no significant effect on subsequent exercise performance in comparison to the control group; (2) post-exercise HBO₂ therapy during the recovery phase did not affect muscle damage or physiological responses in comparison with an NN environment; (3) HBO₂ therapy during exercise appears to induce an effective improvement in muscle oxygenation and muscle fatigue.

The beneficial effects of hyperoxia supplementation on oxygen transport system capacity, lactic acid metabolism capacity, power output performance, and endurance exercise tolerance have been described in multiple previous studies (Sperlich et al., 2017; Cardinale and Ekblom, 2018). With regard to exercise performance, Knight et al. (1993) used a combination of femoral arterial and venous measurements of blood flow to show that HBO₂ treatment could increase the VO_2max of the exercising leg, and three other studies showed that hyperoxic conditions can enhance performance (Cunningham, 1966; Ekblom et al., 1975; Wilson and Welch, 1975). Likewise, BLA, HR_max, and RPE are commonly used monitoring indices in training practice and an effective basis for evaluating training intensity, motor ability, and metabolic function (Glass et al., 1992; Weinstein et al., 1998; Eston, 2012). The absence of a significant effect in our quantitative synthesis thus seems to be at odds with some of the previous research. However, none of the previous studies were RCTs, nor were they design-blind, and all were conducted before the 1980s, necessitating a rational approach to the evaluation of their experimental rigor and conclusions. Horrigan et al. (1979) showed that muscle PO₂ decreased exponentially after HBO₂ exposure; therefore, the time interval between the end of HBO₂ exposure and the start of exercise was relevant in these analyses. These four studies in the quantitative analysis shortened the time between the two events as much as possible for experimental design considerations; however, these investigators did not measure the tissue and blood PO₂. Furthermore, two studies indicated that pre-exercise HBO₂ did not enhance aerobic performance (Webster et al., 1998; McGavock et al., 1999). The ergogenic effect of HBO₂ was not obvious, because under HBO₂ conditions, the increased dissolved oxygen content was released immediately instead of being released in a timely manner to enhance aerobic exercises. For the slight improvement in HBO₂ therapy with isometric knee extensor torque pre-exercise (Kawada et al., 2008), the possible mechanism is that hyperoxic gas supplementation can increase the activity of neurons and keep motor units in a stable state of activation during intense exercise, thus reducing peripheral fatigue and delaying muscle contraction fatigue (Cardinale and Ekblom, 2018). The differences in outcome indicators among the included studies may be related to the individual status of participants in the study, exercise intensity, exercise duration, and other factors. Accordingly, future studies should adopt sports performance indicators close to the specific technical movements of different sports, and on this basis, conduct a rigorous RCT.

In the section titled The effect of post-exercise HBO₂ therapy on recovery (Figure 3), there was no statistical difference in BAP, CK, and LDH levels between HBO₂ and NN in the quantitative synthesis. Reactive oxygen species (ROS) are produced due to incomplete reduction of oxygen during intensive exercise, which leads to a continuous increase in oxidative stress (OS) levels (Powers and Jackson, 2008), necessitating a greater supply of oxygen. Additionally, the levels of serum BAP serve as an important indicator of exercise-induced OS and can directly reflect the OS level and antioxidant capacity. Specifically, existing research has reported that BAP levels increase significantly after the maximal incremental treadmill test (Sugita et al., 2016), mountain bike exercise (Martarelli and Pompei, 2009), and cycling at 75% VO_2max intensity (Aoki et al., 2012). Thus, the BAP indexes of the two studies (Park et al., 2018; Woo et al., 2020) were quantitatively synthesized to reveal the effect of HBO₂ therapy on recovery. The negative conclusion of the meta-analysis could be due to the high antioxidant activity of participants who were trained for exercise over time (Park et al., 2018) and the higher VO_2max levels of the participants (48.20 ± 2.19 mL/kg/min) (Woo et al., 2020) in comparison with their peers (Thompson et al., 2013), which led to higher levels of serum BAP at rest (2348.95 ± 266.02 μmol/L, 2647.61 ± 245.45 μmol/L, respectively) in comparison with those reported in previous studies (from 1938.5 to 2347.3 μmol/L) (Martarelli and Pompei, 2009; Aoki et al., 2012; Sugita et al., 2016). However, high-intensity exercise may be one of the main causes of increased OS, and future studies are needed to more fully demonstrate the recovery effect of post-exercise HBO₂ therapy by analyzing serum ROS levels as well as superoxide dismutase (SOD) and catalase levels. Among the markers of cell damage, CK and LDH are important blood indicators for assessing muscle injury after intensive exercise, and can also be used for fatigue monitoring during the recovery period after exercise (Zajac et al., 2014; Nowakowska et al., 2019). HBO₂ therapy during the post-exercise recovery phase can effectively alleviate exercise-induced muscle injury (Branco et al., 2016; Woo et al., 2020). The negative conclusion in this meta-analysis probably occurred because the intensity of the exercise did not adequately break down the muscle tissue, limiting the expression of serum CK and LDH levels and the subsequent recovery effect. Therefore, in the recovery stage after exercise, no significant difference was observed between the HBO₂ and NN groups. Meanwhile, serum CK and LDH levels tended to increase within 72 h after exercise. Thus, considering the half-life of CK and LDH, additional studies are needed to verify the effect of HBO₂ on fatigue recovery.

RPR is an important indicator of human sensory recovery and is of particular significance for evaluating the effect of fatigue recovery (Laurent et al., 2011). Branco et al. (2016) showed that HBO₂ therapy improved RPR, which is consistent with the findings reported in another study (Kim et al., 2011). However, with varying RPR being the only significant difference in this part of the systematic review, attention should be paid to the potential influence of the placebo effect, as demonstrated by previous research (Broatch et al., 2014; Pinho Júnior et al., 2014). PO₂ and/or tcPO₂ are reliable assessments of tissue oxygen availability (Sheffield, 1998). HBO₂ therapy improves the respiration efficiency of mitochondria, increases the partial pressure of oxygen in blood vessels, improves the dispersion of oxygen in capillaries, and improves the oxygen transport capacity of the body (Sperlich et al., 2017; Cardinale and Ekblom, 2018; Shankar et al., 2018). However, data obtained by Hodges et al. (2003) showed that plasma and tissue oxygen levels did not increase after HBO₂ therapy. This was probably because the oxygen in the blood was mainly transported by binding to hemoglobin, only a small part of which was directly dissolved, and the hemoglobin levels in the body were stable (Wajcman and Kiger, 2002; Klotz, 2003). The data also support the rationale that HBO₂ therapy does not improve exercise performance (Hodges et al., 2003).

Burgos et al. (2016) suggested that antioxidant capacity did not improve during 3 weeks of HBO₂ training, training in an HBO₂ environment did not change oxidative stress in volunteers, and that changes in cellular antioxidant defenses may mediate these results (Radak et al., 2008). Thus, future studies should aim to determine the mechanism of this response to HBO₂. The balance between oxygen supply and consumption was tested using the indicators PO₂, S_aO₂, and TSI (Zinner et al., 2015), which indicates that intermittent oxygen supplementation can improve muscle oxygenation level during high-intensity interval training. HBO₂ treatment during exercise conditions may have increased alveolar-capillary oxygen exchange (Weaver et al., 2009; Casey et al., 2011); therefore, further research is required to clarify this aspect. Stewart et al. (2011) confirmed that short, sustained exposure to HBO could enhance the ability of forearm muscles to generate force during maximum sustained contractions. One possible mechanism for fatigue during reduced muscle oxygenation may be an increase in intracellular ADP, which may be associated with the inhibition of cross-bridge dissociation and, therefore, actin movement during muscle contraction (Debold et al., 2008). Studies have shown that an increase in intracellular inorganic phosphate (Pi) during muscle contraction may lead to muscle fatigue by limiting the release of Ca⁺⁺ from the sarcoplasmic reticulum (Westerblad et al., 2002), and increased tissue oxygen tension prior to muscle contraction may inhibit intracellular ADP and/or Pi concentrations and promote PCr synthesis of ATP, resulting in a cross-bridge closed loop (Stewart et al., 2011). This could explain why sustained HBO₂ therapy maintains power output at a higher level throughout muscle contraction. According to Shimoda et al. (2015), MVC torque values, RMS, and MG were continuously higher in the HBO₂ group than in the NN group throughout the repetitive movement (50 repetitions) (during repetitions 41–50; during repetitions 31–40; during repetitions 41–50, respectively). These results indicate that during exercise, HBO₂ therapy could inhibit the progression of muscle fatigue and bring about sustained output of muscle strength, but the effect on short-term maximum strength generation was not obvious (Shimoda et al., 2015). Reductions in the motor unit complement, synchronicity, and firing rates were considered to be the main reasons for the decrease in the RMS of EMG signals in active muscle groups during maximum contraction exercise induced by muscle fatigue (Kaufman et al., 1984; Crenshaw et al., 2000). During exercise, HBO₂ therapy could increase the synchronization and firing rates of motor units (Gosselin et al., 2004), and then improve the activity ability of neurons, so that motor units can maintain a stable activation state during high-intensity exercise, and then reduce the degree of peripheral fatigue degree (Cardinale and Ekblom, 2018). Because the mechanism by which HBO₂ affects muscle recovery from fatigue is unclear, it is necessary to investigate the benefits of HBO₂ therapy for athletes in greater depth.

In this systematic review and meta-analysis, the PRISMA declaration list was strictly followed (Moher et al., 2015), and the included studies were generally high-quality studies, but the current study had some limitations: (a) the inclusion and exclusion criteria were designed to allow enrollment of healthy adults who were considered physically active; as a result, the number of studies included in the systematic review was insufficient; (b) the exercise protocols differed among studies (e.g., maximal incremental cycling test, TTF at 30% of MVC of bench press; MVC of knee extensors, etc.); (c) the studies used a variety of oxygen equipment and oxygen dosages (Table 1); (d) most of the studies were small and only investigated male participants, except Stewart et al. (2011); (e) and the conclusions of qualitative analysis needed to be confirmed by a large number of further empirical studies.

This systematic review and meta-analysis clearly indicated that pre-exercise HBO₂ therapy had no significant effect on subsequent exercise performance, and the effect of post-exercise HBO₂ therapy on recovery was not obvious. However, HBO₂ therapy administered during exercise can improve muscle endurance performance. Despite the limitations discussed above, this systematic review included studies describing exercise protocols of various types and durations as well as HBO₂ equipment with various set-ups and oxygen dosages; therefore, our conclusions, including the dose-effect relationship between exercise design and oxygen supplementation, need to be confirmed by a large number of follow-up studies and more abundant evidence-based evidence should be obtained.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

XH and RW participated in the study design and drafted the manuscript. XH and ZZ were responsible for writing the manuscript. BG, GW, and RW participated in the overall editing and approval of the manuscript. BG was in charge of financial support. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.