Journal Logo


Hypoxic Training: Effect on Mitochondrial Function and Aerobic Performance in Hypoxia


Author Information
Medicine & Science in Sports & Exercise: October 2014 - Volume 46 - Issue 10 - p 1936-1945
doi: 10.1249/MSS.0000000000000321

L“ive low–train high”, i.e., training in hypoxia, is a training method that aims to improve exercise performance by combining exercise and hypoxic exposure during a portion of the training process. Superimposing hypoxia on exercise, a condition known to reduce intramuscular oxygen partial pressure (27), is thought to induce a higher metabolic stress on the skeletal muscle, in turn promoting a selective adaptive response, ultimately leading to improved athletic performance, particularly aerobic performance.

Although training in hypoxia is considered a strategy to increase exercise performance, its actual ergogenic effects remain controversial. In a recent review (16), of 21 studies, 13 found some additional benefit of hypoxic training for sea level performance. More recent data repeatedly provided contrasting findings, showing either an additional effect on aerobic performance at sea level (15,39) or no effect (14,20,25). Of note, the only two studies using a double-blind design and whole-body exercise training intervention demonstrated either a larger increase in maximal oxygen uptake (V˙O2peak) after training in hypoxia (1) or no additional benefit (35).

Assuming that training in hypoxia has some additive beneficial effects on exercise performance, the question as to which mechanism(s) may mediate these changes arises. First, the possibility that training in hypoxia may enhance performance through systemic hematological changes, i.e., an increase in total red blood cell volume (RCV), is not supported by studies conducted with intermittent hypoxic exposure at rest, indicating no robust hematological change with this technique (3). Hypoxic exposure during training is likely also too short to stimulate erythropoiesis sufficiently and expand RCV (30).

Second, the hypothesis that training in hypoxia does promote adaptive responses favorable to aerobic performance in skeletal muscle is suggested by several factors, such as selective increases in i) messenger RNA levels of hypoxia-inducible factor 1alpha and vascular endothelial growth factor (37,40), ii) citrate synthase activity (23), iii) capillary-to-fiber ratio (37), and iv) mitochondrial respiratory capacity and/or efficiency (26,33). These positive findings are, however, counterbalanced by other data, showing either no specific benefit in capillarization (22), citrate synthase activity (22,34), or mitochondrial respiratory capacity (25) or even a decrease in mitochondrial respiratory capacity after hypoxic training (2).

In summary, the exact role of training in hypoxia on the respiratory capacity of the skeletal muscle and ultimately on systemic oxygen transport and aerobic performance remains difficult to establish.

Another question refers to the effect of training in hypoxia on exercise capacity at moderate altitude. Because aerobic performance is decreased under hypoxic conditions, it is tempting to speculate that physical conditioning in this particular milieu may lead to adaptive responses, ultimately minimizing the performance decrement at altitude. A recent review not only suggests limited evidence for an advantage of hypoxic training for competition at altitude but also points out the lack of available data (16). Other data, indicating either an advantage (9,32) or no additive benefit of hypoxic training when exercise is performed in acute hypoxia (4,15,20), further question its ergogenic effects at altitude.

Using a randomized, double-blind, placebo-controlled design, the aim of the present study was to test the following hypotheses: i) training in hypoxia increases V˙O2peak and endurance performance at sea level and even more so in moderate hypoxia and ii) the mechanism underlying the improvement in aerobic performance after training in hypoxia is of peripheral origin (increase in respiratory capacity of the skeletal muscle) but not systemic (no increase in RCV).



Seventeen physically active male subjects (mean ± SD: age, 27 ± 3 yr; height, 181 ± 7 cm; body mass, 78 ± 9 kg; V˙O2peak, 47 ± 5 mL·kg−1·min−1) participated in this study. Body composition was assessed by a dual-energy x-ray absorptiometer scan (Lunar iDXA™ GE Healthcare, Madison, WI). This study involving human subjects was approved by the ethical committee for the Eidgeno¨ssische Technische Hochschule Zürich (EK 2011-N-51) in accordance with the Declaration of Helsinki. Before the start of the experiments, a written informed consent was obtained from all participants.

Experimental Design

The experimental design consisted of i) familiarization period followed by 2 wk of lead-in training phase, ii) baseline testing, iii) experimental intervention, i.e., a 6-wk endurance training period in either normoxia or normobaric hypoxia, and iv) postintervention measurements.

Familiarization and Lead-in Training

Familiarization consisted of an incremental cycling test to exhaustion followed by a time trial. Each subject repeated the familiarization procedure on two separate occasions before undergoing the baseline measurements. During the 2-wk period preceding the baseline measurements, the subjects performed six sessions of low-volume high-intensity training (18). The objectives of this 2-wk period were i) to homogenize the physical activity of subjects before starting the intervention period (lead-in training) and ii) to separately investigate the effect of low-volume high-intensity training (18).

Exercise Testing

Incremental exercise to exhaustion

Subjects performed an incremental cycling test to volitional fatigue on an electronically braked cycle ergometer (Monark, Varberg, Sweden) to determine maximal workload (Wmax). Minute ventilation and O2 and CO2 concentrations in the expired gas were continuously measured by using an online breath-by-breath gas collection system (Innocor; Innovision, Odense, Denmark). The gas analyzers and the flowmeter of the applied spirometer were calibrated before each test. Subjects began the exercise at three consecutive 5-min workloads at a fixed pedaling cadency, increasing by 50-W increments, beginning with 50 W and finishing with 150 W. From that point, the workload was increased by 30 W every 75 s thereafter until volitional fatigue. Wmax was calculated as Wmax = Wcompl + 30(t/90), where Wcompl is the last completed workload and t is the number of seconds that the final, not completed, workload was sustained and 30 W is the workload increment. V˙O2peak was determined as the highest value averaged over 30 s for each subject. The subjects were then allowed a 1-h rest before beginning a time trial.

The reproducibility of our gas collection system was assessed during a separate experiment in seven subjects who performed duplicate incremental tests until exhaustion (on different days) in normoxia and in acute normobaric hypoxia. The coefficient of variation for V˙O2peak, expressed as percent typical error (i.e., SD of difference scores/

), was 3.5% in normoxia and 3.6% in hypoxia. The present gas collection system was also tested against another metabolic cart (Quark; COSMED, Rome, Italy) during an incremental exercise to exhaustion in four subjects, with the two systems connected in series. In this setting, the percent typical error was 1.4%.

Time trial.

After rest, each subject performed a time trial in which they completed a given workload as quickly as possible. The criterion for completion of the time trial was completing a set relative workload determined as 25% more work than that completed during the incremental cycling test at baseline (thus, the absolute work done before and again after exercise was the same absolute workload). This relative workload across subjects was selected for the time trial exercise tests because, first, it controlled for differences in aerobic capacity across subjects and, second, it made all baseline tests last approximately 20 min (1246 ± 215 s). This allowed for a more standardized test of endurance across all subjects opposed to a test with a set absolute workload. All time trials were performed on an electronically braked ergometer. The ergometer was set to newton mode in an attempt to closely model more realistic race conditions so that each subject could increase or decrease resistance manually and/or via their cadence as desired to complete the predetermined workload. Subjects were instructed to complete the tests as quickly as possible and were provided no temporal or physiological feedback. The only feedback provided was the remaining workload until completion. Exercise duration and average power were recorded upon completion of each test.

Exercise tests in hypoxia.

On a separate day, the subjects performed the same exercise procedure as in normoxia but in normobaric hypoxia (FIO2 = 0.15) (AltiTrainer; SMTEC, Nyon, Switzerland). Recovery between the two tests (incremental exercise then time trial) was performed in normoxia.

Hypoxic Training

Subjects were paired according to their baseline V˙O2peak and randomly divided, in a double-blind, placebo-controlled manner, into two groups: eight subjects training in normoxia and the other nine in normobaric hypoxia (FIO2 = 0.15, corresponding to an altitude of approximately 2500 m). Subjects performed a total of 20 training sessions over 6 wk (3–4 sessions per week), each session lasting 60 min. Four different intensity profiles were alternately given to the subjects to keep motivation high (Table 1). In both groups (i.e., training in normoxia and training in hypoxia), subjects trained at the same relative workload, calculated from their individual Wmax previously determined in normoxia or hypoxia. To reach similar amounts of total work during training in both groups (because absolute workloads were lower during training in hypoxia), the subjects training in normoxia were furtively unloaded at the start and end of each session. During all training sessions, all subjects wore a face mask and inhaled into the air mixing system (AltiTrainer) connected to compressed air (delivering normoxia) or nitrogen (delivering hypoxia). By doing so, neither investigators nor subjects could identify the treatment. Subjects filled out a questionnaire after the training intervention, indicating that 47% guessed the right training group they were in.

Training profiles.

Skeletal Muscle Sampling, Preparation, and High-Resolution Spirometry

Skeletal muscle biopsies were obtained from the vastus lateralis muscle at baseline and after the 6 wk of endurance training. Samples were collected under local anesthesia using the Bergstro¨m technique with a needle modified for suction. The biopsy was immediately dissected free of fat and connective tissue and divided into sections for measurements of mitochondrial respiration, as reported elsewhere (18). All biopsies were taken approximately 48 h after the last bout of exercise. Some of the samples were collected in the morning, and others were collected in the afternoon, but all biopsies were collected under the same standardized conditions. The procedures are also described in the present article (see Text, Supplemental Digital Content 1,, which describes the procedures for mitochondrial respiration measurements).

Total Hemoglobin Mass

Total hemoglobin mass (Hbmass) was measured with a carbon monoxide (CO) rebreathing technique, as previously described (18). Each time, the subject would come to the laboratory and first rest for 20 min in a semirecumbent position. Thereafter, 2 mL of blood was sampled from an antecubital vein via a 20-gauge catheter and analyzed immediately in quadruplicate for i) percent carboxyhemoglobin and Hb concentration ([Hb]) using a hemoximeter (ABL800; Radiometer, Copenhagen, Denmark); and ii) hematocrit (Hct) with the micromethod (4 min at 13,500 rpm). After baseline collection and control, the subject breathed 100% O2 for 4 min to flush the nitrogen from the airways. The breathing circuit (previously flushed with O2) was then closed, and a bolus (1.5 mL·kg−1) of 99.997% chemically pure CO (CO N47; Air Liquide, Paris, France) was administrated into the now closed rebreathing apparatus. The subject rebreathed this gas mixture for 10 min. At the end of the rebreathing period, an additional 2-mL blood sample was obtained and analyzed as before. The change in percent carboxyhemoglobin between the first and second measurement was used to calculate Hbmass, taking into account the amount of CO that remained in the rebreathing circuit at the end of the procedure (2.2%). Total RCV, blood volume, and plasma volume were derived from measures of Hbmass, [Hb], and Hct. All CO rebreathing tests were performed by the same researcher. Baseline and postendurance training values reported here correspond to the average of duplicate measurements conducted on separate days within these testing sessions. The coefficient of variation for Hbmass, assessed from duplicate measures and expressed as the percent typical error (i.e., SD of difference scores/√2), was 3.3% at baseline and 2.4% after endurance training. Of note, during baseline measurements, duplicate Hbmass recordings differed by 122 g in one subject. Without including this result, the typical error of Hbmass would have been 2.5% at baseline.


The Kolmogorov–Smirnov test was applied to examine the normality in the distribution of data. The Bartlett test was used to evaluate the uniformity of variance between conditions. Resting, maximal exercise, and endurance performance data were analyzed using a two-way ANOVA with repeated measures, which included one between-subject (group, i.e., normoxic training and hypoxic training) and one within-subject (timing, i.e., pretraining and posttraining) factor. When necessary, differences between values obtained pre- and posttraining for a particular group were analyzed using Student’s paired t-test. The effect of acute hypoxia on exercise parameters for a particular group was analyzed using Student’s paired t-test. Statistics were done with the StatView software, version 5.0. (SAS Institute, Cary, NC). The values are reported as arithmetic means ± SD. A P value of <0.05 was considered significant.


Body composition.

Six weeks of endurance training was associated with reductions in total body mass and body fat percentage, regardless of training environment (Table 2).

[Hb], Hct, Hbmass, and intravascular volumes.

Maximal exercise.

In both groups, endurance training increased normoxic V˙O2peak (mL·min−1) (mean changes, 9.7% ± 7.2% after placebo and 6.3% ± 7.4% after training in hypoxia) and hypoxic V˙O2peak (mean changes, 15.7% ± 9.8% after placebo and 6.4% ± 12.4% after training in hypoxia) (Table 3). The magnitude of the increase in normoxic or hypoxic V˙O2peak after training was not different between groups.

Maximal exercise and time trial data in normoxia and acute hypoxia.

Time trial.

Endurance training improved normoxic time trial performance among our subjects, who increased their mean power output by 52% and were able to complete the time trial 32% faster (n = 17). The changes in normoxic time trial performance after training were similar in both groups (Table 3 and Fig. 1). In acute hypoxia, training increased time trial performance in both groups; however, mean power output tended to increase more in the group training in hypoxia (52%) than that in the placebo group (32%) (nonsignificant interaction, P = 0.09). Although training increased V˙O2peak and time trial performance concomitantly, there was no relation between the individual changes in V˙O2peak and the corresponding changes in mean power output during time trial, whatever the group (hypoxic training or placebo) or the condition (normoxia or acute hypoxia) (results not shown).

Individual response of mean power output (W) developed during a time trial after completion of a 6-wk endurance training period, either in normoxia (n = 8) or normobaric hypoxia (FIO2 = 0.15) (n = 9). Before and after training intervention, all subjects performed the time trial in normoxia (A) and in acute normobaric hypoxia (FIO2 = 0.15, n = 8 for the group training in hypoxia) (B) on separate occasions. Histograms indicate means. SD and P values are presented in Table 3. ‡Denotes a nonsignificant interaction between training and hypoxia on mean power output (P = 0.09).

Hbmass and intravascular volumes.

Training elicited an overall effect on Hct, Hbmass, RCV, plasma volume, and blood volume (Table 2). The increases in Hct, RCV, and plasma and blood volumes after training did not significantly differ between subjects training in hypoxia and placebo subjects. By contrast, Hbmass increased more after hypoxic training (8.4%) than that after placebo (3.3%) (significant interaction, P = 0.02). Although training increased Hbmass and V˙O2peak concomitantly, there was no relation between the individual changes in Hbmass and the corresponding changes in V˙O2peak, whatever the group (hypoxic training or placebo) or the condition (normoxia or acute hypoxia) (results not shown).

Skeletal muscle respiratory capacity.

Hypoxic training had largely no specific effect on the respiratory capacity of the skeletal muscle. No mass-specific (normalized to tissue wet weight) respiratory capacities were altered by the 6 wk of endurance training (Table 4). In addition, mitochondrial content as assessed by cytochrome c oxidase (COX) activity did not change with training and was not affected by hypoxia. Electron coupling efficiency during fat oxidation (assessed by leak respiration in the absence of adenylates (LN), LN normalized to COX (LN/COX), and leak coupling control during fat oxidation) slightly diminished with hypoxic training, whereas it slightly improved with normoxic training (significant interactions, P ≤ 0.05) (Table 4). There was no relation between the individual training-induced changes in mean power output during time trial and the corresponding changes in skeletal muscle respiratory capacities and measures of efficiency, whatever the group (hypoxic training or placebo) or the condition (normoxia or acute hypoxia) (results not shown).

Skeletal muscle respiratory capacities and measures of efficiency.


Using a randomized, double-blind, placebo-controlled design in physically active subjects, we found that regular exercise training in hypoxia for 6 wk did not induce additional positive effects on maximal aerobic performance or endurance capacity in normoxic conditions. The absence of ergogenic effects coincided with the unchanged respiratory capacity of the skeletal muscle after hypoxic training. When exercise was conducted in acute hypoxia, training in hypoxia also failed to further improve maximal aerobic performance. Finally, hypoxic training intervention demonstrated a trend toward a greater improvement in endurance capacity.

Training in hypoxia confers no specific advantage for aerobic performance at sea level.

Six weeks of training improved V˙O2peak, regardless of the environmental training condition. The magnitude of this increase (approximately 8%) was small in comparison with the 25% increase in V˙O2peak reported after 9 wk of endurance training in sedentary subjects (31). The modest response among our subjects may be attributed to the previous lead-in training (see Methods), which was associated with an 8% increase in V˙O2peak (18). Contrary to hypothesis 1, training in hypoxia did not improve sea level aerobic performance more than normoxic training. This negative finding confirms previous studies (14,16,20,25) but conflicts with others (15,16,39). In particular, we could not confirm the beneficial effect of hypoxic training on V˙O2peak previously demonstrated when using a double-blind design (1). The divergent outcomes may, at least to some extent, be attributed to the considerable differences existing between studies in terms of experimental conditions and subjects’ training status. However, the potential existence of an optimal dose of hypoxia, training, or aerobic fitness leading to a more robust performance improvement remains to be established even if the recent definition of an optimal altitude (2000–2500 m) for “live high–train low” altitude training (7) encourages further research on hypoxic training using this altitude range. Alternatively, the lack of a distinct functional phenotype associated with hypoxic training highlights the large variability in the individual physiological responses during exercise in hypoxia. Finally, in the present study, it may be that hypoxic training at the same relative intensity did not actually impose a higher metabolic stress on the skeletal muscle (as it is the case with the same absolute intensity), in turn explaining the absence of positive performance changes. This assumption is, however, not supported by previous data showing similar V˙O2peak responses after hypoxic training, regardless of whether exercise was carried out at the same relative or absolute intensity (10). Whatever the causes of these divergent findings, in the present study, we recruited moderately trained subjects presenting with substantial scope for performance change and a prolonged training duration to increase the likelihood to induce worthwhile adaptive responses.

Underlying mechanisms.

Contrary to hypothesis 2, neither skeletal muscle respiratory capacity nor electron coupling efficiency was enhanced by hypoxic training in moderately trained subjects. In comparison with normoxic training, hypoxic training slightly decreased the bioenergetic efficiency of fat respiration. Such response remains unexplained but may not be related to higher CHO reliance in hypoxia because our subjects trained at the same relative intensities in both conditions (21). Alternatively, it may be that hypoxic training did promote glycolytic activation (possibly at the expense of fat oxidation) in our subjects who underwent high-intensity bouts in hypoxia in some training profiles (Table 1), as previously reported during supramaximal exercise in hypoxia (24). If this is true, such response may explain, at least in part, why time trial performance in acute hypoxia tended to increase more after hypoxic training than that after placebo. Indeed, in the present setting, greater glycolytic activation might be of importance because the short length of the time trial test implies that muscle metabolism would mostly depend on glycogen oxidation and this would be accentuated after training, when the time trail was even shorter.

Our finding is in line with data from sedentary subjects showing no augmented mitochondrial function after hypoxic training (2,25) but contrasts with studies on athletes where mitochondrial function improved with hypoxic training (26,33). In the present study, after training in hypoxia, the absence of hypoxia-induced alterations in muscle respiratory capacity coincided with unchanged aerobic performance. The fact that endurance training induced a marginal effect on mitochondrial function is surprising because skeletal muscle oxidative capacity is suggested as a predictor of endurance performance at sea level (17), and furthermore, normoxic endurance training was found to simultaneously improve the V˙O2peak and skeletal muscle respiratory capacity of untrained subjects (through increases in adenosine diphosphate-stimulated respiration (38), citrate synthase activity (2), or fatty acid oxidation capacity (25)), suggesting a causal link. In our study, it may be that the lead-in training period (associated with an increase in muscle oxidative capacity and mitochondrial content (18)) prevented subsequent endurance training to further improve muscle oxidative capacity. Nevertheless, the observation of a substantial training effect on aerobic performance without major changes in mitochondrial function i) confirms that respiratory capacity of the skeletal muscle is not the main limiting factor of O2 transport during maximal incremental exercise (5) (although they do correlate well (19)) and ii) highlights the integrative aspects, opposed to individual adaptations, that facilitate training-induced improvements in exercise performance.

At the systemic level, Hbmass was found to increase after training in hypoxia, as compared with that in placebo (contrary to hypothesis 2). One previous study found no change in Hbmass after 3 wk of hypoxic training in athletes (30). First, endurance training is known to promote plasma volume, Hbmass, and RCV expansion, resulting in blood volume expansion (8). Such response occurred in our moderately trained subjects after 6 wk of training. Of note, little is known about the mechanisms of red blood cell adaptation to endurance training (8). Erythropoietin (Epo) stimulation by exercise is equivocal, suggesting that factors other than hypoxemia, such as exercise-induced plasma volume contraction, also modulate Epo production in response to exercise (29). Second, hypoxia promotes red blood cell synthesis via erythropoietic stimulation. However, at rest, short-term hypoxia, even severe, is not a sufficient stimulus to induce RCV expansion (13). Our data showing a slight increase in Hbmass with hypoxic training suggest that systematically superimposing hypoxemia during exercise may potentiate red blood cell adaptation to endurance training. This response, however, had no functional role on maximal aerobic performance or endurance performance at sea level, therefore questioning its physiological significance (posttraining Hbmass values were virtually identical in both groups, but two “nonresponding” subjects in the normoxic training group prevented the difference to reach statistical significance in that group). In the broader context of altitude training, this also questions the actual role of small Hbmass changes on the V˙O2peak response. It may be that Hbmass changes after hypoxic training were too small to elicit an ergogenic effect or, alternatively, that performance enhancement was rather related to plasma/blood volume changes, potentially leading to improved stroke volume during exercise (8).

Does training in hypoxia confer some advantages for aerobic performance at altitude?

With acute hypoxia, V˙O2peak did not increase more after hypoxic training than that in placebo, contrary to hypothesis 2. This negative finding confirms previous reports (15,20,32,36) but differs from other data showing an advantage with hypoxic training (9,12). On the other hand, under acute hypoxia, time trial performance tended to increase more after hypoxic training than that after normoxic training (Fig. 1B). We are not aware of other studies examining endurance performance in hypoxia after hypoxic training. Although the nonsignificant difference between groups (P = 0.09) does not allow us to conclude, this observation nonetheless supports the contention that hypoxic training intervention may be advantageous for endurance performance at moderate altitude. The present study is limited by the small number of subjects, and we do not exclude that a larger study sample reducing type 2 error would have provided a significant outcome.

Assuming an effect, enhanced endurance performance in hypoxia could not be attributed to improved respiratory capacity of the skeletal muscle. By contrast, it may be that the slight Hbmass expansion observed after hypoxic training—although too small to increase V˙O2peak in acute hypoxia—nonetheless contributed to improved time trial performance by enhancing convective O2 transport during submaximal exercise in hypoxia. In support of this, it was shown that Hbmass expansion after recombinant Epo treatment exerted its strongest effect on maximal O2 transport when exercise was performed at moderate altitude (28).

In the present study, moderate acute hypoxia blunted V˙O2peak only marginally. This atypical finding contrasts with the common observation that V˙O2peak decreases at moderate altitude. Such response, however, remains plausible, given the large variability in the decrement of V˙O2peak reported at similar altitudes, ranging from 0% to 17% (11). Here, the fact that normoxic experiments were performed at 460 m may have contributed to lessen the drop of V˙O2peak at moderate altitude (11). The possibility that analytical aspects may have biased V˙O2peak measurements cannot be totally excluded but is not suggested because i) reproducibility of our metabolic cart was proven satisfactorily (see Methods) and ii) the achievement of V˙O2peak was systematically established in all tests by standard criteria such as high RER. Of note, the high levels of end-tidal carbon dioxide pressure observed in our subjects during maximal exercise may be indicative of a specific respiratory pattern (6), but whether this did play any role on V˙O2peak at moderate altitude remains unknown.

Because we did not evaluate lung function or exercise-induced arterial hypoxemia, we cannot exclude the possibility that some pulmonary gas exchange limitations did contribute to minimize the beneficial effects of hypoxic training. However, we can speculate that our physically active subjects, presenting with lower endurance capacities than that in elite endurance athletes, were likely not prone to exercise-induced arterial hypoxemia, suggesting that pulmonary limitations were not the main cause of our negative results.

Finally, the question arises as to whether the use of hypobaric hypoxia, instead of normobaric hypoxia, would have led to different results because the two conditions are thought to induce differential responses (e.g., in terms of arterial saturation or ventilatory work). This question is, however, still a matter of debate, and previous data using hypobaric hypoxia during hypoxic training failed to demonstrate a positive outcome (10). Further research on this aspect is warranted.

In conclusion, 6 wk of training in hypoxia i) did not improve skeletal muscle respiratory capacity or coupling control and ii) did not confer any advantage for maximal aerobic performance or endurance performance at sea level. Under acute hypoxic condition, training in hypoxia did not improve V˙O2peak, but we do not rule out that hypoxic training may exhibit some functional advantage associated with endurance capacity. Our data support further research on endurance performance at moderate altitude after hypoxic training in endurance athletes.

This work was supported by the Swiss National Science Foundation, grant number 320030-143745 (C. L.), and by the Ministère des Sports, de la Jeunesse, de l’Education Populaire et de la Vie Associative/Institut National du Sport, de l’Expertise et de la Performance, grant number 11-R-011 (P. R.).

The authors’ contributions were as follows: P. R., R. A. J., and C. L., significant manuscript writers; P. R., D. F., R. A. J., and C. L. significant manuscript reviewers/revisers; P. R., R. A. J., and C. L., concept and design; T. B., D. F., S. B., M. T., R. A. J., and C. L., data acquisition; P. R., T. B., D. F., S. B., M. T., R. A. J., and C. L., data analysis and interpretation; P. R., R. A. J., and C. L., statistical expertise.

There is no conflict of interest to be reported.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. Bailey DM, Davies B, Baker J. Training in hypoxia: modulation of metabolic and cardiovascular risk factors in men. Med Sci Sports Exerc. 2000; 32 (6): 1058–66.
2. Bakkman L, Sahlin K, Holmberg HC, Tonkonogi M. Quantitative and qualitative adaptation of human skeletal muscle mitochondria to hypoxic compared with normoxic training at the same relative work rate. Acta Physiol (Oxf). 2007; 190 (3): 243–51.
3. Bärtsch P, Dehnert C, Friedmann-Bette B, Tadibi V. Intermittent hypoxia at rest for improvement of athletic performance. Scand J Med Sci Sports. 2008; 18 (1 Suppl): 50–6.
4. Beidleman BA, Muza SR, Fulco CS, et al. Intermittent hypoxic exposure does not improve endurance performance at altitude. Med Sci Sports Exerc. 2009; 41 (6): 1317–25.
5. Boushel R, Gnaiger E, Calbet JA, et al. Muscle mitochondrial capacity exceeds maximal oxygen delivery in humans. Mitochondrion. 2011; 11 (2): 303–7.
6. Bussotti M, Magri D, Previtali E, et al. End-tidal pressure of CO2 and exercise performance in healthy subjects. Eur J Appl Physiol. 2008; 103 (6): 727–32.
7. Chapman RF, Karlsen T, Resaland GK, et al. Defining the “dose” of altitude training: how high to live for optimal sea level performance enhancement. J Appl Physiol. 2014; 116 (6): 595–603.
8. Convertino VA. Blood volume: its adaptation to endurance training. Med Sci Sports Exerc. 1991; 23 (12): 1338–48.
9. Dufour SP, Ponsot E, Zoll J, et al. Exercise training in normobaric hypoxia in endurance runners. I. Improvement in aerobic performance capacity. J Appl Physiol. 2006; 100 (4): 1238–48.
10. Engfred K, Kjaer M, Secher NH, et al. Hypoxia and training-induced adaptation of hormonal responses to exercise in humans. Eur J Appl Physiol Occup Physiol. 1994; 68 (4): 303–9.
11. Fulco CS, Rock PB, Cymerman A. Maximal and submaximal exercise performance at altitude. Aviat Space Environ Med. 1998; 69 (8): 793–801.
12. Geiser J, Vogt M, Billeter R, Zuleger C, Belforti F, Hoppeler H. Training high–living low: changes of aerobic performance and muscle structure with training at simulated altitude. Int J Sports Med. 2001; 22 (8): 579–85.
13. Gore CJ, Rodriguez FA, Truijens MJ, Townsend NE, Stray-Gundersen J, Levine BD. Increased serum erythropoietin but not red cell production after 4 wk of intermittent hypobaric hypoxia (4,000-5,500 m). J Appl Physiol. 2006; 101 (5): 1386–93.
14. Hamlin MJ, Marshall HC, Hellemans J, Ainslie PN, Anglem N. Effect of intermittent hypoxic training on 20 km time trial and 30 s anaerobic performance. Scand J Med Sci Sports. 2010; 20 (4): 651–61.
15. Haufe S, Wiesner S, Engeli S, Luft FC, Jordan J. Influences of normobaric hypoxia training on metabolic risk markers in human subjects. Med Sci Sports Exerc. 2008; 40 (11): 1939–44.
16. Hoppeler H, Klossner S, Vogt M. Training in hypoxia and its effects on skeletal muscle tissue. Scand J Med Sci Sports. 2008; 18 (Suppl 1): 38–49.
17. Jacobs RA, Rasmussen P, Siebenmann C, et al. Determinants of time trial performance and maximal incremental exercise in highly trained endurance athletes. J Appl Physiol. 2011; 111 (5): 1422–30.
18. Jacobs RA, Flück D, Bonne TC, et al. Improvements in exercise performance with high-intensity interval training coincide with an increase in skeletal muscle mitochondrial content and function. J Appl Physiol. 2013; 115 (6): 785–93.
19. Jacobs RA, Lundby C. Mitochondria express enhanced quality as well as quantity in association with aerobic fitness across recreationally active individuals up to elite athletes. J Appl Physiol. 2013; 114 (3): 344–50.
20. Lecoultre V, Boss A, Tappy L, et al. Training in hypoxia fails to further enhance endurance performance and lactate clearance in well-trained men and impairs glucose metabolism during prolonged exercise. Exp Physiol. 2010; 95 (2): 315–30.
21. Lundby C, Van Hall G. Substrate utilization in sea level residents during exercise in acute hypoxia and after 4 weeks of acclimatization to 4100 m. Acta Physiol Scand. 2002; 176 (3): 195–201.
22. Masuda K, Okazaki K, Kuno S, Asano K, Shimojo H, Katsuta S. Endurance training under 2500-m hypoxia does not increase myoglobin content in human skeletal muscle. Eur J Appl Physiol. 2001; 85 (5): 486–90.
23. Melissa L, MacDougall JD, Tarnopolsky MA, Cipriano N, Green HJ. Skeletal muscle adaptations to training under normobaric hypoxic versus normoxic conditions. Med Sci Sports Exerc. 1997; 29 (2): 238–43.
24. Morales-Alamo D, Ponce-González JG, Guadalupe-Grau A, et al. Increased oxidative stress and anaerobic energy release, but blunted Thr172-AMPKα phosphorylation, in response to sprint exercise in severe acute hypoxia in humans. J Appl Physiol. 2012; 113: 917–28.
25. Pesta D, Hoppel F, Macek C, et al. Similar qualitative and quantitative changes of mitochondrial respiration following strength and endurance training in normoxia and hypoxia in sedentary humans. Am J Physiol Regul Integr Comp Physiol. 2011; 301 (4): R1078–87.
26. Ponsot E, Dufour SP, Zoll J, et al. Exercise training in normobaric hypoxia in endurance runners. II. Improvement of mitochondrial properties in skeletal muscle. J Appl Physiol. 2006; 100 (4): 1249–57.
27. Richardson RS, Noyszewski EA, Kendrick KF, Leigh JS, Wagner PD. Myoglobin O2 desaturation during exercise. Evidence of limited O2 transport. J Clin Invest. 1995; 96 (4): 1916–26.
28. Robach P, Calbet JA, Thomsen JJ, et al. The ergogenic effect of recombinant human erythropoietin on V˙O2max depends on the severity of arterial hypoxemia. PLoS One. 2008; 3 (8): e2996.
29. Roberts D, Smith DJ, Donnelly S, Simard S. Plasma-volume contraction and exercise-induced hypoxaemia modulate erythropoietin production in healthy humans. Clin Sci (Lond). 2000; 98 (1): 39–45.
30. Robertson EY, Saunders PU, Pyne DB, Gore CJ, Anson JM. Effectiveness of intermittent training in hypoxia combined with live high/train low. Eur J Appl Physiol. 2010; 110 (2): 379–87.
31. Roca J, Agusti AG, Alonso A, et al. Effects of training on muscle O2 transport at V˙O2max. J Appl Physiol. 1992; 73 (3): 1067–76.
32. Roels B, Bentley DJ, Coste O, Mercier J, Millet GP. Effects of intermittent hypoxic training on cycling performance in well-trained athletes. Eur J Appl Physiol. 2007; 101 (3): 359–68.
33. Roels B, Thomas C, Bentley DJ, Mercier J, Hayot M, Millet G. Effects of intermittent hypoxic training on amino and fatty acid oxidative combustion in human permeabilized muscle fibers. J Appl Physiol. 2007; 102 (1): 79–86.
34. Terrados N, Melichna J, Sylvén C, Jansson E, Kaijser L. Effects of training at simulated altitude on performance and muscle metabolic capacity in competitive road cyclists. Eur J Appl Physiol Occup Physiol. 1988; 57 (2): 203–9.
35. Truijens MJ, Toussaint HM, Dow J, Levine BD. Effect of high-intensity hypoxic training on sea-level swimming performances. J Appl Physiol. 2003; 94 (2): 733–43.
36. Ventura N, Hoppeler H, Seiler R, Binggeli A, Mullis P, Vogt M. The response of trained athletes to six weeks of endurance training in hypoxia or normoxia. Int J Sports Med. 2003; 24 (3): 166–72.
37. Vogt M, Puntschart A, Geiser J, Zuleger C, Billeter R, Hoppeler H. Molecular adaptations in human skeletal muscle to endurance training under simulated hypoxic conditions. J Appl Physiol. 2001; 91 (1): 173–82.
38. Walsh B, Tonkonogi M, Sahlin K. Effect of endurance training on oxidative and antioxidative function in human permeabilized muscle fibres. Pflugers Arch. 2001; 442 (3): 420–5.
39. Wang JS, Wu MH, Mao TY, Fu TC, Hsu CC. Effects of normoxic and hypoxic exercise regimens on cardiac, muscular, and cerebral hemodynamics suppressed by severe hypoxia in humans. J Appl Physiol. 2010; 109 (1): 219–29.
40. Zoll J, Ponsot E, Dufour S, et al. Exercise training in normobaric hypoxia in endurance runners. III. Muscular adjustments of selected gene transcripts. J Appl Physiol. 2006; 100 (4): 1258–66.


Supplemental Digital Content

© 2014 American College of Sports Medicine