Effect of hypoxia conditioning on physical fitness in middle-aged and older adults—a systematic review and meta-analysis

Introduction

The global aging trend has led to a continuous increase in the middle-aged and older population (aged over 40 years) (McNicoll, 2002). Aging is a primary risk factor for chronic diseases, including cardiovascular disease, neurodegenerative conditions, which also impairs physical and cognitive functions, contributing to muscle atrophy (Goodpaster et al., 2006; Niccoli & Partridge, 2012; Partridge, Deelen & Slagboom, 2018). The loss of muscle strength, reduced oxygen uptake, and diminished physical function in the older population are closely associated (Arnett et al., 2008; Gouveia et al., 2020; Sillanpää et al., 2014), further reducing the quality of life. Thus, identifying strategies to address the decline in physical fitness among middle-aged and older adults is crucial.

Exercise has been proven to be an effective mean of improving physical fitness in middle-aged and older adults (Valenzuela et al., 2023; Villareal et al., 2017). Various forms of exercise, including aerobic exercise, resistance training, and high-intensity interval training (HIIT), have shown positive effects on the physical fitness of this population (Hunter, McCarthy & Bamman, 2004; Marzuca-Nassr et al., 2020; Nash, 2012). For instance, sedentary middle-aged and older adults who participated in a 16-week aerobic exercise intervention showed significant improvements in their 36-item short-form health survey (SF-36) scores compared to those who did not exercise (Reid et al., 2010). Additionally, healthy middle-aged and older individuals demonstrated significant enhancements in flexibility, stability, strength, and maximal oxygen uptake (VO₂max) after 12 weeks of HIIT (Marzuca-Nassr et al., 2020). The benefits of resistance training in enhancing muscle strength and function in middle-aged and older adults have also been well-documented (Csapo & Alegre, 2016; Frontera et al., 1988). However, despite the widely recognized benefits of exercise on physical fitness, approximately 25% of adults globally do not meet the recommended levels of physical activity (WHO, 2021). The percentage of older adults meeting the recommended weekly physical activity levels is even lower than that of younger populations (Watson et al., 2016), with only 2.5% to 22% of community-dwelling older adults achieving the World Health Organization-recommended levels of physical activity (PA) (150 min of moderate-intensity PA per week) (Harris et al., 2009; Sagelv et al., 2019). Therefore, more efficient methods are needed to help middle-aged and older adults manage their physical fitness.

Hypoxic conditioning (HC) is a non-pharmacological therapeutic intervention that induces systemic and/or localized tissue hypoxia by exposing individuals to low-oxygen conditions, either at rest or during exercise. Hypoxia elicits multisystemic physiological adaptations, including the upregulation of erythropoietin synthesis to stimulate erythropoiesis (Ehrenreich et al., 2022; Rodríguez et al., 1999), the potentiation of sympathetic nervous system activity with concomitant augmentation of ventilatory response (Burtscher et al., 2024) and stimulating the increase in mitochondrial electron transfer efficiency to maintain adenosine triphosphate (ATP) production (Fukuda et al., 2007). This approach can lead to similar or enhanced physiological and functional benefits compared to normoxic training (Hobbins et al., 2017; Samuel & Franklin, 2008; Verges et al., 2015). This technique is often used as a training strategy to enhance athletic performance (Millet et al., 2010), and has also been shown to improve physical function in non-athletic individuals. Additionally, it can be safely and effectively used to manage elevated heart rate and blood pressure in older patients with cardiovascular diseases (Glazachev et al., 2021; Maher, Jones & Hartley, 1974; Nishimura et al., 2010). For example, eight weeks of hypoxic conditioning led to greater muscle hypertrophy in healthy young individuals compared to exercise alone (Kurobe et al., 2015). In the middle-aged and older population, eight weeks of hypoxic conditioning significantly improved peak oxygen consumption (VO₂peak) and maximal power output (PO_max) in obese participants (Chacaroun et al., 2020). A four-week intervention combining aerobic exercise with intermittent hypoxia significantly increased the time to exhaustion in older adults during an exercise load test (Schega et al., 2016). Importantly, hypoxic conditioning does not increase perceived exercise intensity and may even reduce the overall exercise workload (the blend of volume and the intensity in training) (Girard, Malatesta & Millet, 2017; Scott et al., 2018; Wiesner et al., 2013). The efficacy of hypoxic conditioning remains an ongoing scientific debate. Study have demonstrated that training under hypoxic conditions may surpass normoxic exercise in enhancing lean body mass and muscle fiber cross-sectional area. This purported superiority is attributed to the synergistic effects of mechanical loading and hypoxia-induced metabolic stress, which collectively enhance motor unit recruitment, stimulate anabolic hormone secretion, and promote myofibrillar protein synthesis (Jung et al., 2021). However, conflicting evidence suggests that hypoxic exposure may paradoxically accelerate skeletal muscle catabolism through impairment of protein synthesis mechanisms (Chen et al., 2014). Notably, this debate also exists among middle-aged and older people. For instance, after eight weeks of hypoxic endurance training, participants with an average age of 62 years showed no differences in constant workload test results or VO₂peak compared to those who trained under normoxic conditions (Chobanyan-Jürgens et al., 2019). Additionally, studies involving obese middle-aged and older participants found that five or 18 weeks of hypoxic conditioning did not result in significantly higher VO₂peak and PO_max compared to equivalent training under normoxia (Camacho-Cardenosa et al., 2020; Törpel, Peter & Schega, 2020).

Therefore, this study aims to systematically review and conduct a meta-analysis to analyse the effects of training under hypoxic conditions in physical fitness outcomes compared to the same training under normoxic conditions in middle-aged and older adults. Adding hypoxia to exercise may further enhance physical fitness outcomes in this population.

Materials & Methods

This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) 2020 Statement (Page et al., 2021) and registered at PROSPERO (CRD42023474570). We determined the inclusion criteria based on the PICOS principles: (1) Participants, (2) Interventions, (3) Comparisons, (4) Outcomes, (5) Study Design.

Results

Search yield

During the study selection phase, a total of 215 articles were identified as potentially eligible (PubMed: n = 18, Embase: n = 41, Web of Science: n = 128, Cochrane: n = 26; manual search: 2). Following the exclusion of duplicates, 186 articles remained for screening. A total of 13 articles (Allsopp et al., 2022; Camacho-Cardenosa et al., 2020; Chacaroun et al., 2020; Chobanyan-Jürgens et al., 2019; Gatterer et al., 2015; Ghaith et al., 2022; Klug et al., 2018; Park et al., 2019; Schega et al., 2016; Schega et al., 2013; Schreuder et al., 2014; Törpel, Peter & Schega, 2020; Wiesner et al., 2013) were finally included in the meta-analysis after screening full-texts. This meta-analysis assessed an experimental group with 188 participants and a control group with 180 participants (Fig. 1).

Results of risk of bias evaluation

Detailed information on the risk of bias in randomized controlled trials can be found in Fig. 2. We found that out of 13 articles, 11 did not clearly specify how they generated random sequence and conducted allocation concealment, leading to an unclear risk of selection bias. Approximately half of the studies exhibited performance bias and attrition bias. Only three studies were assessed with low risk of bias in blinding of outcomes assessment and three with high risk of bias in incomplete outcome data.

Meta-analysis

Results of functional outcomes analysis

Three experiments involving 77 participants reported functional outcomes using the 36-Item Short Form Health Survey - physical component, time to exhaustion, and timed “Up and Go” test. Moderate certainty evidence (Fig. 3A) revealed that the combined strategy had the similar effect on functional outcomes compared to exercise alone (SMD = −0.21, 95% CI [−0.66–0.24]; low heterogeneity), the certainty of evidence was downgraded due to differences in outcome measures. Subgroup analysis (Table 2) did not reveal significant effects based on age, obesity, intervention duration, hypoxia duration, inspired oxygen fraction, and intervention type. Sensitivity analysis did not reveal any differences in the pooled results after removing a single study.

Analysis of VO₂max

Five experiments involving 142 participants reported an outcome on VO₂max. Very low certainty evidence (Fig. 4A) found that the combined hypoxic conditioning and exercise strategy had a similar effect on VO₂max as exercise alone (SMD = −0.39, 95% CI [−1.12–1.90]; moderate to high heterogeneity), the certainty of evidence was downgraded due to high heterogeneity and wide 95% CI. Subgroup analysis (Table 2) did not reveal significant effects on types of age, obesity, intervention duration, hypoxia duration, inspired oxygen fraction, and intervention type. Sensitivity analysis did not reveal any differences in the pooled results after removing a single study.

Analysis of exercise workload

Four studies involving 102 participants reported an outcome on workload. Very low certainty evidence (Fig. 5) found that the combined hypoxic conditioning and exercise strategy had a similar effect on workload as exercise alone (MD = −10.07, 95% CI [−34.95–14.80]; moderate to high heterogeneity), the certainty of evidence was downgraded due to high heterogeneity and wide 95% CI. Subgroup analysis (Table 2) did not show any significant effects based on age, obesity, intervention duration, hypoxia duration, inspired oxygen fraction, and intervention type. Sensitivity analysis did not reveal any differences in the pooled results after removing single study.

Discussion

The concern for the health of middle-aged and older population has become a primary focus of public health (Concha-Cisternas et al., 2023; Edwards, 2012; Gell et al., 2023; McNicoll, 2002; Shirazi et al., 2023; Sillanpää et al., 2014). In this context, we evaluated a strategy for improving physical fitness through the application of hypoxia as an intervention outside of exercise training. High certainty evidence revealed that training under hypoxic conditions led to a greater improvement in VO₂peak compared to training under normoxic conditions, while it had consistent effects on functional outcomes, muscle strength, power output, and VO₂max in middle-aged and older individuals as equivalent training in normoxia.

The meta-analysis found that the use of hypoxia conditioning resulted in significantly greater benefits for VO₂peak. Oxygen uptake capacity, a key indicator of cardiorespiratory fitness and aerobic capacity (Caspersen, Powell & Christenson, 1985; Myers et al., 2002; Sugie et al., 2018), was found to have marked improvement with hypoxic conditioning compared to normoxic conditions. One research has found that middle-aged and older individuals who undergo 8 weeks of HC experience a more significant improvement in VO₂peak compared to subjects who exercise under normoxic conditions. This may be attributed to the higher exercise intensity used in the study compared to other studies with similar intervention durations and training methods (Chacaroun et al., 2020). However, contrasting outcomes have been observed in other studies. For instance, after an eight-week intervention combining hypoxia and HIIT in adults under 65 years old, no significant difference in VO₂peak was found between the two groups. One possible explanation is the 20% reduction in training workload during HIIT under hypoxic conditions compared to normoxic conditions (Ghaith et al., 2022). In middle-aged individuals, eight weeks of HC resulted in a 10% increase in VO₂peak (Schreuder et al., 2014). While hypoxic conditioning effectively improved the participants’ physical fitness, no intergroup differences were observed. This might be due to the relatively low exercise intensity, the use of maximal heart rate as a reference for training intensity did not account for the fact that maximum heart rate is lower under hypoxia compared to normoxic ones (Ozcelik & Kelestimur, 2004; Schreuder et al., 2014). Our study supported the effectiveness of HC in enhancing oxygen uptake in middle-aged and older population. Notably, the observed superior effect of HC on VO₂peak improvement primarily derives from a single included study, which underscores the necessity of rigorous validation through additional high-quality studies. VO₂max, determined through maximal exercise testing, represents the maximum rate of oxygen consumption during physical exertion, while VO₂peak refers to the highest measured oxygen uptake attained during submaximal exercise testing and is commonly used to estimate VO₂max (Bassett Jr & Howley, 2000; Ryan, 2014). Older adults frequently undergo submaximal exercise testing due to age-related challenges in sustaining maximal exertion levels during peak performance assessments (Church et al., 2008; Schultz et al., 2020). This physiological limitation in achieving maximal effort may explain why hypoxia conditioning demonstrate limited efficacy in improving VO₂max within middle-aged and older populations. Only two studies reported information on both VO₂max and VO₂peak, and future studies need more comprehensive data to compare the effects of hypoxia conditioning on aerobic capacity in middle-aged and older adults, and to explore potential associations between these two measures before and after the intervention.

For middle-aged and older population, hypoxia conditioning may not be the optimal choice for improving muscle strength. This finding aligns with the outcomes from a meta-analysis examining the effects of HC on muscle strength (Csapo & Alegre, 2016). While healthy young participants showed significant improvements in both absolute and relative 1RM after a seven-week resistance training regimen under hypoxic conditions compared to exercise alone, this effect may be attributed to the utilization of a lower volume, high-intensity training protocol, which is better suited for enhancing maximal strength (Inness et al., 2016). Such adaptive changes may also arise from the higher metabolic stress induced by hypoxia, triggering functional muscle adaptations (Schoenfeld, 2013; Scott et al., 2017). However, in healthy older individuals, an 8-week combination of hypoxia and resistance training did not yield significant changes in muscle strength compared to those exercising in normoxic conditions (Allsopp et al., 2022). This could stem from the delayed vasodilatory response of skeletal muscles due to aging and a sluggish upregulation of genes related to myogenic cell differentiation, collectively resulting in a delayed adaptation of older participants to resistance training in hypoxic conditions (Casey et al., 2011; Gnimassou et al., 2018).

In the context of aging-related conditions, about 2%–34% of people are affected by sarcopenia, resulting in diminished muscle strength and functionality (Reijnierse et al., 2015; Shafiee et al., 2017). A study involving 2,318 high-altitude residents over the age of 60 found that the incidence of sarcopenia was significantly higher than in those living at lower altitudes (Ye et al., 2020). After an 8-week stay in high-altitude areas, 24 healthy male subjects at 34.9 years showed that body weight was significantly decreased and association between lower oxygen saturation and loss of lean mass was established (Wandrag et al., 2017). Sarcopenia stems primarily from aging but is also influenced by external environmental factors. For instance, hypoxic conditions can lead to muscle atrophy and diminished contractile strength. This may be attributed to the impact of hypoxia, which not only accelerates the breakdown of skeletal muscle tissue, but also disrupts protein synthesis pathways (Chen et al., 2014). Therefore, hypoxia may be a risk factor for sarcopenia, and the application of hypoxic conditioning requires caution (Ye et al., 2020). Future research is needed to establish the causal relationship between hypoxia and sarcopenia in middle-aged and older populations.

The limited effectiveness of hypoxia conditioning in enhancing cardiorespiratory function and muscle strength may help explain why significant improvements in functional outcomes and maximal power output were not observed. Physical fitness such as muscle strength, cardiorespiratory capacity, and functionality typically decline with age (Arnett et al., 2008; Sillanpää et al., 2014; Trombetti et al., 2016). Although some studies suggest that hypoxic conditioning can improve physical fitness in middle-aged and older individuals, the evidence remains mixed. For instance, a 12-week intervention that combined hypoxia with aerobic and resistance trainings in older adults led to a notable increase in chair sit-to-stand repetitions compared to a control group engaged in exercise alone (Park et al., 2019), suggesting that hypoxia conditioning may enhance both aerobic and anaerobic capacities and muscle functionality more effectively than normoxic conditions. In middle-aged individuals, an 8-week hypoxia and aerobic exercises program resulted in a significant improvement in PO_max, potentially linked to the slight but non-significant increase in lean body mass (Chacaroun et al., 2020). However, our meta-analysis indicates that the effects of hypoxia conditioning combined with exercise on functional outcomes and power output are comparable to exercise alone. In older individuals, a 5-week exercise program that paired hypoxia with low to moderate level of resistance training did not yield additional benefits in muscle strength and PO_max, possibly due to the insufficient exercise load, although resistance training has been shown to increase PO_max in younger cohorts (Törpel, Peter & Schega, 2020). Additionally, an 18-week intervention involving whole-body vibration under hypoxic conditions showed no significant improvements in the timed “Up and Go” test or lean body mass among older individuals compared to the baseline, underscoring the potential limitations of hypoxia conditioning in this age group and highlighting the need for further research to determine the most effective combinations of physical activities and hypoxic conditions (Camacho-Cardenosa et al., 2020).

The variability of individual responses to exercise interventions may also contribute to the suboptimal efficacy of HC interventions. For instance, although physical exercise demonstrates superior effects compared to usual hospital care in improving functional outcomes, gait speed, and muscle strength among hospitalized older adults aged ≥75 years, at least 8.7% of individuals still exhibited no response to the intervention (non-responders) (Sáez de Asteasu et al., 2019). Evidence from insulin-resistant middle-aged women similarly reveals the presence of non-responders to both HIIT and resistance training (RT) interventions, with the prevalence of non-responders showing significant associations with exercise modalities (Álvarez et al., 2017). This phenomenon may originate from multiple factors including the difference in skeletal muscle microRNA expression, training status, psychological stress, and habitual physical activity (Davidsen et al., 2010; Mann, Lamberts & Lambert, 2014). Given the consistent emergence of non-responders across various exercise modes and the elevated prevalence of chronic diseases in older populations (Guo et al., 2022; Izquierdo et al., 2021; Wang et al., 2020), comprehensive exercise-related assessments, screenings, and individualized exercise prescriptions should be implemented prior to administering HC interventions to middle-aged and older adults.

The lower oxygen concentration, resulting in reduced actual training volume and exercise intensity, could explain the absence of significant intergroup differences. In middle-aged overweight individuals, a 4-week exercise program combining hypoxia and aerobic activities did not lead to significant changes in VO₂max. However, participants using HC experienced a significantly lower workload compared to the exercise-only group (Wiesner et al., 2013). In studies with lower training volume and intensity compared to normoxic training, it was revealed that even when power output and muscle strength were similar in both the HC group and the exercise-only group, HC may still serve as an effective alternative, particularly for populations unable to maintain a fixed training load, such as patients with osteoarthritis and middle-aged and older adults with lower physical fitness (Ghaith et al., 2022; Schreuder et al., 2014). Additionally, the influence of hypoxia on heart rate or exercise load should be taken into consideration, and conducting exercise experiments under hypoxic conditions may be more beneficial in determining the appropriate exercise intensity for participants undergoing HC.

A monitored, sustained low-oxygen supply combined with higher exercise intensity may be a crucial factor influencing the effectiveness of hypoxic conditioning. The study conducted by Chacaroun et al. (2020) introduced bias in the pooled analysis of VO₂peak, leading to significantly improved VO₂peak in the subgroup analysis when hypoxia was combined with exercise compared to exercise alone. In comparison to other studies (Camacho-Cardenosa et al., 2020; Gatterer et al., 2015; Schega et al., 2016; Schega et al., 2013; Törpel, Peter & Schega, 2020), the research of Chacaroun et al. (2020) utilized sustained and stable low-oxygen supply (80% SpO₂) and higher exercise intensity (75% HR_max), which may have been more effective in inducing greater improvements in aerobic capacity than normoxic training.

Some limitations should be acknowledged. The limited number of studies included in this meta-analysis restricted our ability to perform subgroup analyses across various study characteristics. This constraint also impeded our ability to distinguish real asymmetry from chance, thereby precluding the creation of a funnel plot (Sterne et al., 2011). Additionally, the absence of detailed descriptions of the randomization processes in most included studies may introduce potential biases. Moreover, only four studies reported the mean age of the final study population (Chacaroun et al., 2020; Gatterer et al., 2015; Ghaith et al., 2022; Wiesner et al., 2013), which could potentially introduce age-related biases by including individuals outside the intended middle-aged and older adult demographic. Although most high-quality research were published in English (Morrison et al., 2012), it may be necessary to consider material in a wider range of languages to broaden the scope of future research. While the observed heterogeneity was low, sensitivity analyses revealed reduced robustness in the pooled meta-analytic results for VO₂peak. This finding not only underscores the need for cautious interpretation of the efficacy of HC on improving VO₂peak outcomes, but also emphasizes the necessity to incorporate and systematically analyze additional studies in future investigations.

The randomization procedures in 11 out of 13 RCTs were marked as “Not clear,” potentially impacting the validity of the final results. For instance, the study by Chacaroun et al. (2020), which involved higher training intensity combined with a sustained, stable hypoxic supply, contributed to the poor robustness in the pooled analysis for power output. Similarly, the lack of robustness in the pooled analysis for VO₂max can be attributed to the younger age of the participants (Wiesner et al., 2013). Furthermore, the analysis of workload robustness was compromised by the relatively severe hypoxic levels used in another study (Ghaith et al., 2022), which resulted in a significant 20% reduction in absolute workload among participants undergoing hypoxia conditioning.

Conclusions

Hypoxic conditioning has a greater effect on enhancing VO₂peak compared to equivalent normoxic training in the middle-aged and older population, while both have similar effects on maximal power output, functional outcomes, muscle strength, and workload. Future researches should explore different combinations of oxygen concentrations and exercise intensities in hypoxic conditioning to better understand its impact on middle-aged and older adults. Adopting more rigorous randomized controlled trial designs is also recommended to enhance the reliability of existing evidence.

Supplemental Information