Physical exercise promotes the modulation of a large panel of cellular signaling pathways to promote metabolic and/or morphological changes that enhance performance (Bishop et al., 2019; Sanchez et al., 2019; Solsona et al., 2021). In recent years, repetition of short (≤30 s) all-out exercises were boosted in popularity because they can induce similar or higher gains of performance for a lower training volume compared to conventional endurance protocols (MacDougall et al., 1998; Barnett et al., 2004; Burgomaster et al., 2006, 2008; Gibala et al., 2006). Two distinct repeated sprint protocols are distinguished. Repeated sprint training (RST) consists of maximal sprint exercises of short duration (≤10 s) interspaced with short recovery periods (i.e., exercise to rest ratio <1:4). On the other hand, sprint interval training (SIT) includes repeated long sprints (~30 s) interspaced by longer rest periods (~2–4 min) (Buchheit and Laursen, 2013a,b; Brocherie et al., 2017). In recent years, studies have suggested that the addition of a hypoxic stimulus to chronic SIT and RST may favor several training responses (Faiss et al., 2013; Brocherie et al., 2017; Brechbuhl et al., 2020; James and Girard, 2020). For example, systemic hypoxia (HYP) may promote higher muscle perfusion and oxygenation with greater modulations of molecular adaptations (Faiss et al., 2013; Brocherie et al., 2017). These changes include increases in HIF-1α, myoglobin, and the expression of genes involved in mitochondrial biogenesis.

In addition to HYP, blood flow restriction (BFR) training strategies can also favor several training adaptations (Preobrazenski et al., 2021). BFR results in reduced arterial and/or venous blood flow according to the level of pressure exerted, thus generating local hypoxia and an accumulation of metabolites in the working muscles (Sugaya et al., 2011; Teixeira et al., 2018; Okita et al., 2019). BFR exercise training was found to be beneficial for enhancing adaptations such as muscle hypertrophy and aerobic capacity (Preobrazenski et al., 2021). Acute exercise with BFR may increase neuromuscular activation and type II muscle fibers recruitment (Moritani et al., 1992). In addition, BFR aerobic exercise stimulates ventilatory and cardiac response through the stimulation of group III and IV afferents (Adreani and Kaufman, 1998). These acute responses result in increased energy expenditure leading to a loss of muscle efficiency during submaximal exercise (Ozaki et al., 2010; Mendonca et al., 2014; Silva et al., 2019). However, to the best of our knowledge, BFR studies were conducted with low-intensity exercises or during RST (Willis et al., 2019), and little is known about SIT protocols using BFR.

Furthermore, different BFR models, such as cuff-, pressure chamber-, and gravity-models rise in popularity, and studies are needed to highlight their specific effects and related adaptations. Recently, a gravity-induced BFR (G-BFR) aerobic protocol has been investigated in a study conducted by Preobrazenski and colleagues (Preobrazenski et al., 2020). An ergocycle titled 45° was used to generate ischemia within working muscles. In this study, muscle oxygenation was effectively found to be reduced during submaximal aerobic exercises, as previously indicated during supine exercise where faster O₂ uptake kinetics was also observed (Hughson et al., 1991, 1993). Of note, Preobrazenski et al. suggested that the G-BFR model seems favorable to enhance aerobic adaptations (Preobrazenski et al., 2020). However, nothing is known about the use of such a protocol during SIT. Importantly, to our knowledge, the acute effects of HYP and BFR models on both cardiovascular and muscular responses have never been compared during SIT protocols. It appears important to understand the acute effects of these models to identify potential training adaptations for the efficient prescription of training strategies.

Thus, the goal of this study was to assess the effects of each condition on mechanical output, cardiorespiratory responses, and muscle oxygenation in healthy moderately trained men. In the present study, it has been chosen to set maximal stress for both HYP and BFR conditions that could be tolerated in this cohort in combination with the SIT protocol proposed based on preliminary work in our laboratory. Concerning the G-BFR condition, the inclined position was maintained during the entire exercise session. The hypotheses were that we would observe different responses depending on the condition with (i) a higher power output during the successive exercise bouts in NOR since the other conditions induce additional stress that may lead to a reduction of acute muscle efficiency and subsequent performance, (ii) a greater impact of HYP on the cardiorespiratory responses because HYP causes hypoxemia, and (iii) a more pronounced impact of BFR models on muscle oxygenation because they generate local hypoxia.

No interaction between conditions and sprint number was found for any variable (p > 0.05).

The aim of the present study was to compare the acute effects of systemic hypoxia, BFR₆₀, and G-BFR on mechanical output, cardiorespiratory responses, and O₂ muscle extraction during SIT exercises in healthy moderately trained men. The main results were that SIT associated with G-BFR displayed lower mechanical and cardiorespiratory responses than the other modalities. G-BFR also showed lower skeletal muscle oxygenation. BFR₆₀ induced greater blood accumulation within working muscles compared to the other conditions. Moreover, HYP at 13% FiO₂ and G-BFR increased local hypoxia within the working muscles, with a higher level of hypoxia observed for G-BFR.

The primary results showed that SpO₂ was lower for HYP throughout the exercise session and decreased at the end of the session (i.e., sprints four and five) for all the conditions. According to Fick's law of diffusion, systemic hypoxemia occurred in HYP because alveolo-capillary oxygen pressure difference decreases under hypoxic conditions. Consequently, V.E_peak increased in HYP, as a compensatory response aiming to maintain oxygen consumption. In accordance, previous studies observed that the magnitude of the ventilatory response appears a critical factor of performance during a Wingate test in hypoxia (Fallon et al., 2015). Indeed, the authors showed that ventilation was elevated at the beginning and throughout the exercise when F_iO₂ was decreased from 20.9 to 10% (Fallon et al., 2015). However, the augmentation of ventilation observed in the present study was not sufficient, as total and peak oxygen uptake were still lower in HYP compared to NOR. In addition, mean ventilation was higher in HYP during the first period of recovery. Hypoxemia had an impact on muscle oxygenation parameters. Specifically, oxyhemoglobin concentration during sprints was lower in HYP compared to NOR. Therefore, as total work was the same between the conditions, except for G-BFR, this suggests that the anaerobic metabolism is more solicited in HYP. Furthermore, we analyzed SpO₂ during recovery when the cuffs were applied and did not observe any difference between the conditions. Similarly, no effect was detected in the study from Willis et al. (2019) using continuous BFR during the RSH protocol at 45% of arterial occlusive pressure.

In the current protocol, fatigue increased progressively throughout the session as performance variables (i.e., peak and mean power, time to reach mean power) were altered, and HR increased with the repetition of sprints. On the other hand, ΣVO₂ and V.O_2peak were lower during the first sprint probably because metabolic demands are ensured by the anaerobic metabolism (i.e., phosphocreatine hydrolysis and glycolysis). Interestingly, V.O_2peak was higher in NOR compared to the other conditions, which is in accordance with a study from Willis et al. comparing NOR and BFR (Willis et al., 2018). Furthermore, no effect of the condition or the number of sprints was found on FI. A study by Fallon and coworkers (Fallon et al., 2015) observed higher FI in HYP but their study included a single sprint of 30 s with a higher level of hypoxia (FiO₂ = 10%). In the current study, the BFR₆₀ application induced a decrease in gas exchange rates, which increased after its release. This phenomenon has already been observed during partial occlusion in dogs where oxygen consumption was decreased during BFR application and increased when pressure was removed. These responses relied on blood flow, which means that BFR causes vascular resistance (Fales et al., 1962). Importantly, in the current study, BFR was applied for the first 2 min of recovery. This probably delayed and reduced overall recovery, which may influence in turn performance, energy system usage, and metabolite accumulation in the following sprints. The BFR protocol used in the present study was applied during recovery periods between exercise bouts and as such has a different effect to the continuous application that may be seen during other studies. For HYP, performance variables were unaffected but average and maximal oxygen consumption and muscular oxyhemoglobin availability decreased. For all conditions, peak and average power decreased after each sprint excepting between sprints four and five where values were similar. Time to reach peak power increased between the three first sprints and sprint five. RPE increased throughout the exercises. Altogether, these data suggest that during SIT, both BFR and HYP enhance cellular stress (i.e., metabolite accumulation and/or hypoxia) without affecting total work during the training sessions.

Importantly, several cardiorespiratory parameters (i.e., ΣVT, V.E_peak, V.E_min, ΔV.E, ΣVO₂, ΣVCO₂, HR, HR_peak, HR_min, ΔHR), mean TSI (exercise and recoveries), TSI_max and TSI_min during exercise and/or recovery were lower for G-BFR compared to the other conditions. Of note, during sprints, the supine position induced lower mean, maximal, and minimal tissue saturation. Concerning training data, RPE appeared lower for G-BFR compared to BFR₆₀ and HYP, peak power was lower compared to NOR, and time to achieve peak power was higher compared to BFR₆₀ and HYP. Average power also appeared lower for G-BFR compared to the other conditions. Overall, these data are consistent with the literature. Indeed, an important decrease in strength of the knee flexor and extensor muscles is inherent to the supine vs. seated position (Houtz et al., 1957). Moreover, maximal exercise performance (such as maximal work rate, V.O_2max, V.E_max, and HR_max) during an incremental test has been shown to be impaired in supine position compared to upright cycle position, probably due to a lower perfusion (Hughson et al., 1991). Our results are consistent with the recent study of Preobrazenski and colleagues who showed a decrease in muscle oxygenation in their G-BFR model compared to the control group. However, they reported a higher RPE with G-BFR whereas we obtained the opposite result. This could be explained by the difference in exercise modality (aerobic vs. repeated all-out exercises). Altogether, the G-BFR condition may alter biomechanical and cardiorespiratory responses. However, G-BFR induced lower mean tissue oxygenation. Thus, G-BFR seems to induce specific stress compared to the other modalities.

Concerning BFR₆₀ and HYP, present data indicate no difference with NOR for mechanical outcomes. These results are in line with the data from Willis et al. (2019) who compared NOR, BFR, and HYP during an RST protocol. Indeed, the authors found no difference in mean power, mean ventilation, or HR_peak. They also found a significant increase in total hemoglobin concentration with BFR (constant pressure of 45%). The present results also suggest that BFR₆₀ promotes higher blood accumulation within working muscles than the other conditions, meaning that BFR₆₀ may additionally affect training adaptations by confining metabolites. Importantly, BFR was suggested to induce greater neuromuscular fatigue while HYP produces central fatigue by impairing corticospinal excitability because of cerebral deoxygenation (Willis et al., 2018; Peyrard et al., 2019). During BFR application, deoxyhemoglobin concentration was higher and oxyhemoglobin content was not different compared to NOR. This result means that a partial occlusion (60%) has a greater impact on venous return than on arterial blood supply. Of note, power output strongly decreases for G-BFR, which may explain in turn the decrease of HR values while it should be increased with higher venous return. Furthermore, enhanced deoxyhemoglobin concentration matches the blood accumulation (increased Δ[tHb]) observed during BFR application. Its release induces bigger changes in oxyhemoglobin, deoxyhemoglobin, and total hemoglobin concentrations compared to the other conditions at the same period. On the contrary, our results suggest that G-BFR does not have a significant impact on venous return, as recently demonstrated by another group during aerobic exercises (Preobrazenski et al., 2021). During sprints, TSI was higher for BFR₆₀ compared to the other conditions, due to an increased oxyhemoglobin concentration ([O₂Hb]). On the other hand, peak and mean HR were lower in BFR₆₀ during the exercises. Interestingly, HYP and G-BFR induced greater local hypoxia within skeletal muscles, which was more prominent in G-BFR when considering both exercise bouts and recovery periods. Thus, G-BFR would represent an alternative to HYP to promote additional hypoxic stress within skeletal muscles cells. Additional studies are needed to compare the chronic effects of these conditions on both cellular adaptations to training and gains in muscle performance, especially because G-BFR lowers power output during sessions.

Furthermore, some limits must be acknowledged. First, the modalities of HYP, BFR₆₀, and G-BFR conditions were established according to preliminary work that allowed to identify the maximal stress that could be tolerated by the participants in combination with the SIT protocol of this study. Thus, further studies are needed to evaluate more precisely the nature and the degree of the stress generated by these conditions. On the other hand, BFR₆₀ was the only stress which was not continuously applied. In such a situation, the BFR application still had a delayed impact on muscle oxygenation parameters during sprints. However, based on preliminary work, this mode of BFR has been chosen to set maximal stress that could be tolerated in this population in combination with the SIT protocol. Another limit can be that participants were moderately trained individuals, and it was a demanding protocol. This also makes it difficult to generalize the results, especially to untrained populations for which these training methods would be hard to complete. Moreover, TSI measurements cover a limited zone of the vastus lateralis muscle thus interpretations of the results at the whole muscle are to be considered with these limitations. Finally, the present study did not include women athletes. Indeed, some adaptations such as cardiovascular responses differ between men and women (Patel et al., 2021), thus introducing variability. Additional works are needed to compare outcomes according to gender because women remain underrepresented in sport science literature.

In conclusion, this study conducted in healthy moderately trained men showed that a session of SIT in combination with G-BFR showed lower mechanical, cardiorespiratory responses, and skeletal muscle oxygenation than the other conditions. Another important insight is that both SIT protocols conducted under HYP at 13% FiO₂ and G-BFR amplified local hypoxia within the working muscles. Importantly, a higher level of hypoxia was found with G-BFR when considering the measurement of the entire exercise session (i.e., exercise bouts and recovery periods). Furthermore, a single session of SIT associated with BFR₆₀ promotes higher blood accumulation within working muscles compared to the other exercise modes. This suggests that BFR₆₀ may additionally or differentially affect cellular homeostasis. Thus, each condition generates specific stress and further studies are needed to better understand subsequent consequences on long-term adaptation.

Further studies are needed (i) to compare the effects of these protocols on-field performance, (ii) to evaluate more precisely the degree of stress generated by each condition, even if high values for both BFR₆₀ and hypoxia have been used based on literature for this kind of exercise, and (iii) to examine mechanistic insight since the mechanisms of action may be different. Investigations of the involvement of different cellular pathways, for example, those involved in mitochondrial adaptations (e.g., the axis of peroxisome proliferator-activated receptor-gamma coactivator 1-alpha) are needed to improve our knowledge about the molecular benefits of these training methods. These research directions are important because they may help to improve the ability to develop more efficient hypoxic or BFR training modalities and to improve skeletal muscle function and whole-body metabolism. Finally, adding stress to training would promote adaptations in the long term, at least when recovery processes are sufficient.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The experimental protocol was approved by the Local Ethics Committee (VD-2021-00597). The patients/participants provided their written informed consent to participate in this study.

RS, FB, and AS: conceived and designed the experiments, analyzed the data, designed the tables and figures, wrote the manuscript, contributed reagents, materials, and analysis tools. RS, HB, and FB: performed the experiments. All authors read the manuscript and approved the final version.

An internal Bonus Qualité Recherche from the University of Perpignan Via Domitia was obtained to perform this study.

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.