Unpacking the mitochondrial bioenergetics of blood flow restricted resistance exercise

While resistance training (i.e. strength training) is the most prominent exercise modality for improving skeletal muscle strength and hypertrophy, it is traditionally only able to elicit these adaptations when prescribed as heavy-load resistance training (HLRT). This generally involves loads greater than 70% of an individual's one-repetition maximum (1RM). However, HLRT may be contraindicated among many populations including frail individuals, people living with chronic diseases, or those with post-operative musculoskeletal injuries due to the associated mechanical stress or elevations in haemodynamic responses. In such cases, low-load resistance training (≤30% 1RM) combined with blood flow restriction (BFR-RT) presents as a viable alternative capable of achieving comparable increases in muscle strength and hypertrophy. In practice, BFR-RT utilises pneumatically pressurised cuffs fitted to active limbs during exercise, partially restricting blood flow to the muscle groups distal to the cuffs with the lower loads reducing the mechanical stress associated with resistance exercise. Despite the established benefits and adaptations, the underlying mechanisms that support the efficacy of BFR-RT have not been thoroughly elucidated. Traditional skeletal muscle strength and hypertrophy adaptations to BFR-RT are thought to be related to a combination of local and systemic stimuli that affect pathways within skeletal muscle responsible for protein synthesis and degradation, such as the mTOR and MAPK pathways. Other mechanisms contributing to this may be related to compression-induced hypoxia in the blood flow-restricted muscle groups, changes in ATP concentrations, or physiological factors related to oxidative stress. These latter factors suggest that acute exercise-mediated changes in mitochondrial bioenergetics may be a contributory factor to the adaptations that occur following BFR-RT. Interestingly, changes in mitochondrial function, including protein fractional synthesis rate and mitochondrial respiration, are traditionally associated with aerobic exercise. However, Groennebaek et al. (2018) recently demonstrated that both HLRT and BFR-RT may also improve these mitochondrial functions. That BFR-RT may also elicit these improvements is of particular importance for ageing and various chronic disease populations who are often contraindicated to high-intensities of aerobic activity or the high mechanical stress associated with HLRT, as they stand to benefit most from the reduced physical demands of BFR-RT. While low load resistance training (LLRT) to volitional fatigue may also induce favourable muscular adaptations, BFR-RT has been shown to produce a similar degree of hypertrophy to that of LLRT while requiring a lower training volume to achieve this, which may be an important consideration when working with clinical populations (Farup et al. 2015). Mitochondrial function is also negatively affected among many of these clinical populations due to their propensity for prolonged physical inactivity as well as the progression of many chronic diseases. This further exacerbates both their physical limitations and underlying metabolic disorders. Yet, despite the findings from Groennebaek et al. (2018), the role of mitochondrial bioenergetics in BFR-RT adaptations is still poorly understood and evidence underpinning it is often contradictory. In a recent article in The Journal of Physiology, Petrick et al. (2019) explored the mitochondrial bioenergetic responses to BFR-RT performed to volitional fatigue. Increased mitochondrial reactive oxygen species (ROS) emission can promote signalling for transcription of mitochondrial genes. While evidence is equivocal, hypoxic conditions and reductions in oxygen partial pressure (), such as that induced during BFR-RT, could be capable of increasing mitochondrial ROS production by causing the electron transport chain to be overly reduced and optimised for superoxide radical production. This could be an important signalling event mediating the adaptations to BFR-RT. A reduction in tissue oxygenation (measured via near-infrared spectroscopy) with decreased blood flow during BFR-RT in vivo was also hypothesised to elicit a similar effect. However, this hypothesis also opposes the findings of prior studies which have demonstrated that BFR-RT does not induce acute oxidative stress, albeit through the indirect measurement of oxidative stress using blood and plasma markers (Goldfarb et al. 2008; Neto et al. 2018). Petrick et al. directly examined skeletal muscle tissue from 10 healthy young adult males (24 ± 1 years) to assess mitochondrial ROS emission rates, rather than the indirect measurement of blood and plasma markers which are more commonly employed as a proxy for oxidative stress in similar BFR-RT studies. Additionally, the influence of reduced on mitochondrial bioenergetics was examined in vitro from a second group of six healthy young adult males (25 ± 2 years). As tissue oxygenation was reduced during BFR-RT compared with LLRT (∼30% 1RM), which may explain the differences in mitochondrial ROS emission rates between exercise protocols, this in vitro approach served to assess the direct influence of oxygen tension on mitochondrial bioenergetics in the absence of confounding factors present during exercise. A combined approach such as this helps to reconcile some of the contrasting findings between solely in vivo studies and solely in vitro studies. In contrast to their hypothesis though, the authors found that acute BFR-RT attenuated muscle mitochondrial ROS emission rates in vivo by approximately 30% from pre-exercise levels (P < 0.01), compared with no change in LLRT from pre-exercise levels. In addition, in vitro oxygen restriction substantially attenuated maximal (∼4-fold) and submaximal (∼50-fold) ROS emission rates compared with a room air condition (P < 0.05). Therefore, the induced hypoxic state during BFR-RT may be important for regulating ROS emissions, particularly as in vitro reductions in dramatically attenuated ROS emissions. In turn, this creates a somewhat paradoxical preclusion of the authors’ hypothesis regarding the role of mitochondrial ROS emissions in explaining the metabolic adaptations to BFR-RT. These findings differ from previous studies that have observed no difference in oxidative stress, albeit through indirect blood and plasma markers, compared with HLRT following traditional BFR-RT protocols (4 sets of 30, 15, 15 and 15 repetitions) rather than to volitional fatigue (Centner et al. 2018; Neto et al. 2018). This may also highlight the importance of using direct measurements of skeletal muscle in assessing ROS emission rates, rather than indirect markers from blood sampling. While the findings from Petrick et al. further expound the understanding of basic mitochondrial redox, particularly in conjunction with the presence of hypoxia, it also highlights the paucity of evidence that actually explains the metabolic and chronic training adaptations to BFR-RT. The strength of the study from Petrick et al. is the use of direct measures of mitochondrial ROS emissions using human permeabilised muscle fibres alongside comparative conditions in a more controlled in vitro environment. This combination represents a robust method for assessing mitochondrial ROS emissions that builds on the results of previous studies assessing blood and plasma markers (Goldfarb et al. 2008; Neto et al. 2018) or using electron paramagnetic resonance spectroscopy (Centner et al. 2018) for assessing oxidative stress following BFR-RT. However, the timing of the muscle biopsies conducted in the study by Petrick et al. presents a potential confounding variable. Samples were taken in conditions 2 h post-exercise, common protocol among studies utilising aerobic exercise modalities. While aerobic exercise at moderate or higher intensities are prescribed prolonged work periods (>30 min), the total volume of both the LLRT and BFR-RT work periods in this study were less than 5 min, potentially evoking a more transitory response that was not maintained 2 h post-exercise. An improvement to the study design would be to take a more immediate post-exercise muscle biopsy in addition to the ones taken 2 h post-exercise, similar to the protocol used by Groennebaek et al. (2018). That would then allow for the detection of any short-term mitochondrial bioenergetic responses as well as those reported in the study. A broader limitation of many similar BFR-RT studies examining metabolic responses is the heterogeneity of prescribed training variables. While Petrick et al. identify that repetitions to volitional fatigue may be paramount to achieving metabolic adaptations, it is also not the single most common prescription of BFR-RT. Similar studies by Goldfarb et al. (2008) and Groennebaek et al. (2018) also used three to four sets to volitional fatigue, while studies by Neto et al. (2018) and Centner et al. (2018) used a ‘traditional’ BFR-RT prescription of four sets consisting of 30, 15, 15 and 15 repetitions. The degree of occlusion pressure is also inconsistently applied among these studies, some employing between 50% and 70% of limb or arterial occlusion pressure (Centner et al. 2018; Groennebaek et al. 2018; Petrick et al. 2019), and others using arbitrarily applied pressures including 20 mmHg below systolic blood pressure (Goldfarb et al. 2008), or 1.3 times systolic blood pressure (Neto et al. 2018). Comparator RT groups are less diverse, with most similar studies comparing BFR-RT with HLRT using three to four sets of 70–80% 1RM. However, Petrick et al. compare only load-matched LLRT to the BFR-RT group. While these methodological differences do not invalidate conclusions drawn from among these studies, they simply add a layer of complexity that makes direct comparison of results more difficult. In order to derive a clearer understanding of the mechanism and metabolic adaptations that occur with BFR-RT, a greater degree of consistency is needed between study designs and the protocols utilised for BFR-RT and comparator groups. The work conducted by Petrick et al. does indeed advance our understanding of the physiological processes contributing to the metabolic adaptations to BFR-RT through relatively robust methodology and ultimately a rejection of their hypothesis. This also highlights what is still unknown with regards to both the mechanisms supporting BFR-RT, as well as the measurement and function of exercise-mediated changes in mitochondrial bioenergetics. Further research with equally stringent measures and the utilisation of more consistent BFR-RT protocols will help to develop the font of knowledge regarding these acute metabolic responses to this form of exercise training. Understanding these responses and the mechanisms that generate them will dictate how these novel modalities of exercise training are ultimately prescribed in order to maximise the beneficial training adaptations that are evoked. The author confirms there are no conflicts of interest relevant to the contents of this article. Sole author. This research was supported only by local funds made available by the School of Exercise and Nutrition Sciences, Faculty of Health, Deakin University, Victoria, Australia.