Journal Logo

Applied Sciences: Physical Fitness and Performance

Skeletal muscle adaptations to training under normobaric hypoxic versus normoxic conditions


Author Information
Medicine & Science in Sports & Exercise: February 1997 - Volume 29 - Issue 2 - p 238-243
  • Free


The progressive decline in maximal aerobic power and work capacity with ascent to increasing altitudes is well known and documented(22,27). It is also well known that the physiologic adaptations that occur with altitude training and acclimatization will improve physical performance at altitude. Considerable controversy exists, however, as to whether performance at sea level can be enhanced by altitude training. If one uses ˙VO2max as the criterion, most studies indicate no improvement in sea level values when athletes return from a period of altitude training(1,6,8,9,28). Despite this, many coaches and athletes believe that sea level performance in endurance events will benefit from brief periods of training at moderate altitude and offer numerous anecdotal accounts attesting to its effectiveness.

Investigation of the literature on altitude acclimatization reveals that following 10 or more days of exposure to moderate altitude a significant improvement is seen in physical work capacity at altitude(4,22) and a reduction in plasma lactate occurs when subjects exercise at the same power output (22). There is, however, little or no change in ˙VO2max over this time(1,25). This disproportionately large increase in work capacity and apparent decrease in lactate production, despite minimal change in ˙VO2max, suggests that adaptations may have occurred at the muscle level rather than adaptations that may have enhanced oxygen delivery.

We have investigated muscle adaptations to 40 d of extreme hypoxia in a“live-in” hypobaric chamber (11,21) and concluded that hypoxia on its own is not a stimulus for increased mitochondria synthesis or oxidative enzyme activity, although there was a slight increase in muscle capillary density as a result of fiber atrophy. In contrast, Terrados et al. (29) demonstrated that when hypoxia is combined with exercise significantly greater increases occur in oxidative enzyme activity and myoglobin than when the same training is performed in normoxia. This has been substantiated by Bigard et al. with rats(3) and Kaijser et al. (18) with humans, who found greater increases in citrate synthase activity in muscles that were trained under conditions of local ischemia than in muscles trained under normal conditions. The purpose of this study was to determine whether training under normobaric hypoxic conditions (simulating medium level altitude) would enhance physical performance and selected muscle adaptations over and above that which would occur with normoxic training. By using a unilateral training model whereby one leg was trained under normoxic conditions and the other under hypoxic conditions (29), we were able to control for adaptations that may affect performance other than those at the muscle level.


Subjects. Ten healthy male volunteers (19-25 yr) served as subjects. Although they were physically active, none had undergone previous endurance training. All participants were fully informed of the purposes of the study and associated risks as approved by the Human Ethics Committee of McMaster University. Participants received remuneration for participating in the study and training compliance was 100%.

Design. Exercise performance characteristics and biochemical and morphometric properties of the vastus lateralis were assessed for each leg before and after 8 wk of endurance training.

Training consisted of unilateral cycle ergometry so that one leg was trained while subjects breathed an inspirate of 13.5% oxygen (balance nitrogen) and the other leg was trained while breathing normal ambient air. Performance characteristics included measurements of maximum aerobic power(˙VO2max) and maximum aerobic capacity (time to fatigue at 95%˙VO2max) for each leg. Biochemical measurements included assays for oxidative and glycolytic enzyme activity and morphometric measurements included assessments of muscle capillary density, fiber area and% fiber type, and mitochondrial and lipid volume density.

The hypoxic condition was achieved by having subjects breathe out of a 350-l Tissot gasometer which was coupled to an electrically controlled valve which diluted ambient air by bleeding nitrogen into it at a controlled rate. The configuration was such that ambient air was first bubbled through a smaller tank filled with water to be moisturized before being mixed with nitrogen and drawn into the Tissot tank by vacuum pump. Subjects breathed the gas mixture through a standard Rudolph valve. The system was capable of delivering the inspirate on line up to a maximum inspired ventilation of approximately 120 l·min-1 (BTPS). The inspirate was continuously monitored by a second oxygen analyzer and a variation of less than 0.2%(FIO2) was maintained throughout all training sessions.

Training program. Subjects trained 3 times per week for 8 wk. For the first 6 wk, each training session consisted of 30 min of continuous unilateral cycling in the normoxic condition and 30 min of cycling by the opposite leg in the hypoxic condition. The order of conditions was alternated each training session and our design was such that, for five subjects, the hypoxically trained leg was randomly designated as the left leg and for the remaining five subjects it was the right leg. Initial training intensity was set at 75% of the lower of the two legs' pre-training maximum power output. This value was increased on an individual basis every 2 wk to provide a progressive overload. The magnitude of this increase depended upon each subject's tolerance ability and amounted to ≈2% every second week. The increased training load was always applied first in the hypoxic condition so that, if subjects were unable to complete the full 30 min, the load could be reduced and the training load for the normoxic condition matched to it. Thus, the same absolute power output was always maintained for the two legs on any given day.

Training for the final 2 wk consisted of a combination of interval and continuous training. In each session and for each leg, subjects performed five 3-min intervals at 100% of the pre-training maximum power output with 3-min recovery periods. This was followed by 10 min of continuous training at the power output reached in the sixth week of training. The interval training was added to simulate the training programs of middle distance athletes who often combine interval training with continuous training as part of their normal preparation for competition. All training sessions were supervised, and inspired FO2 was continuously monitored and held constant during the hypoxic training condition.

Measurements. ˙VO2max was recorded for each leg separately during unilateral cycling on an electrically braked cycle ergometer. The test began at an initial power output of 50 W and the load was increased 15 or 30 W every 2 min until fatigue. Heart rate was continuously recorded by a 3 lead ECG throughout the test, as was oxygen uptake by means of a computerized open circuit system which calculated ˙VO2 on-line every 30 s. The peak ˙VO2 attained was considered to be˙VO2max and the highest power output that was sustained for at least 1 min was considered to be maximum power output.

The maximum aerobic capacity (MAC) test consisted of unilateral cycling to fatigue on the ergometer at a power output which corresponded to that at 95% of pre-trained ˙VO2max for each leg. Fatigue was considered to occur when subjects could no longer maintain a pedaling rate of 60 rpm and test duration was recorded to the nearest 1.0 s.

Needle biopsies were extracted from the vastus lateralis of the right leg before the training period and from both legs after the training period using the Bergström technique (1962) and applying “suction” with a 50-ml syringe. Post-training biopsies were taken 2 or 3 d after the last training sessions. On each occasion two biopsy samples were taken. The first was divided for electron microscopy and histochemistry, and the second was immediately frozen in liquid nitrogen for subsequent biochemical analysis.

For histochemistry, cryostat sections (7 μm) were stained for myofibrillar ATPase activity (following pre-incubation at pH of 4.3,4.6, and 10.0) or with hematoxylin and eosin. The slides were then photographed under a light microscope at 40 × magnification and the photographs projected onto either a 144 square grid for capillary counting or a computerized digitizer for measurement of fiber area. Capillaries were counted from the hematoxylin and eosin stained tissue (for a field of ≈250 fibers) and expressed per square millimeter as well as per fiber. Cross-sectional area was determined for an average of 125 Type I fibers and 125 Type II fibers per biopsy. Percent fiber type was estimated by counting an average of ≈300 fibers per biopsy.

For electron microscopy, tissue was prepared as has been described(21). Serial ultrathin sections were made at a slightly oblique angle to the fibers, stained with uranyl acetate and lead citrate, and mounted on copper/rhodian grids. These sections were photographed at approximately 50,000 × magnification under a Philips EM 301 (Eindhoven, The Netherlands). Where possible, 50 fibers were randomly selected per biopsy, and for each a photographic field for the interior of each fiber was randomly selected and photographed. Stereological analysis was performed on each micrograph by means of a 168 point shortline test system(30) according to the method as described by Hoppeler et al. (15). For each biopsy, volume densities were calculated for myofibrils, interior mitochondria, lipid, and cytoplasm.

The second biopsy samples were stored at -80° C until biochemical analysis. Tissue was freeze dried, dissected free of blood and connective tissue, and homogenized in 50% glycerol, 20 mM sodium phosphate buffer (pH = 7.4), 5 mM B-mercaptoethanol, 0.5 mM EDTA, and 0.02% BSA(19). The activities of citrate synthase (CS), succinate dehydrogenase (SDH) and phosphofructokinase (PFK) were determined fluorometrically (12) and expressed as either mmol·g-1 tissue or mmol·hr-1·g-1 tissue.

Data were analyzed with a two-factor (pre-post-training × training condition) analysis of variance. When a significant interaction was found, apost-hoc test (Tukey A) was used to identify significant differences among mean values. Statistical significance was accepted at P ≤ 0.05.


Performance measurements. Compared to pre-training values,˙VO2max was significantly (P < 0.05) higher for both legs following training (Fig. 1). The mean increase was≈13% in the normoxically-trained leg and ≈11% in the hypoxically-trained leg with no difference between conditions.

Following training, time to fatigue during unilateral cycling at 95% of the pre-training ˙VO2max, markedly increased for both legs (P< 0.05). Subjects were able to maintain exercise ≈400% longer with the normoxically-trained leg and ≈510% longer with the hypoxically-trained leg, but this difference between conditions did not achieve statistical significance (Fig. 2).

Muscle enzyme activities. The activity of CS, SDH, and PFK increased significantly in the muscle of both legs following training(Fig. 3). CS activity was ≈51% higher in the normoxically-trained leg and ≈71% higher in the hypoxically-trained leg compared with pre-training values. The increase in CS activity in the hypoxically-trained leg was also significantly greater than that in the normoxically-trained leg (P < 0.05). SDH activity was ≈35% higher in the normoxically-trained leg and ≈63% higher in the hypoxically-trained leg, but the difference between conditions was not statistically significant. Similarly PFK activity was ≈23% higher in the normoxically-trained leg and 32% higher in the hypoxically-trained leg, but again the difference between training conditions was not statistically significant.

Morphometric measurements. The effects of training on mitochondrial volume density, fiber type and area, and muscle capillarization are summarized in Table 1. Although capillary/fiber ratio, capillary density, and mitochondrial volume density tended to be higher following training and especially in the hypoxically-trained leg, none of these changes achieved statistical significance. Percent fiber type and cross-sectional area of Type I fibers were unaffected by training although there was a tendency for Type II fiber areas to be greater following training.

Additional measurements. Following training there was no change in body mass, pre-exercise hemoglobin concentration increased significantly from 14.7 g% to 15.8 g%, exercise heart rate was significantly lower at the same submaximal power outputs, peak power output was significantly higher for both legs, and peak plasma lactate was significantly higher following the˙VO2max test for each leg.


Since the purpose of this study was to isolate and examine the effects of the combination of exercise training and hypoxia, the subjects breathed the hypoxic mixture only while they were training the designated leg. They were thus only exposed to the hypoxic environment for a total of 90 min·wk-1, and it should be recognized that our study was not intended to simulate a condition in which training is conducted under chronic hypoxia. Our selection of an inspirate with a fractional oxygen concentration of 13.5% was based on a pilot study of different hypoxic mixtures. In that study five healthy young subjects performed progressive and constant load exercise on separate occasions while breathing either normal ambient air or inspirates that ranged from 11.0-14.0% O2 in a randomized and blinded manner. An FIO2 of 13.5% was the lowest that these subjects could tolerate for 30 min while performing cycle exercise at 75% of their maximum normoxic power output. Peak ˙VO2 at this inspirate was ≈83% of their normoxic ˙VO2. An FIO2 of 13.5% results in a PIO2 of ≈103 mm Hg and corresponds to an altitude of ≈3,292 m.

Our subjects trained each leg at the same absolute intensity. Since absolute ˙VO2 measurements during training were the same under both conditions, the level of oxidative phosphorylation was probably the same for both legs. Because peak ˙VO2 is reduced under hypoxic conditions our training protocol was such that the hypoxic training represented a higher relative intensity (>90% ˙VO2peak) than the normoxic training(≈78% ˙VO2max). The question thus arises as to whether any differences in adaptive response between the two legs are a result of the hypoxia per se or simply a result of the differences in relative training intensity imposed by the hypoxic condition. While one would normally expect differences in relative training intensity to affect the nature and magnitude of the adaptive response in muscle, such differences are normally accompanied by differences in absolute power output, oxygen uptake, and enzyme kinetics. In the present study, care was taken to ensure that each leg trained for the same duration and at the same absolute power output each training day so that the only difference was the presence or absence of the hypoxic condition. Thus it is valid to attribute any between-leg differences in muscle adaptation to the hypoxia per se.

The relatively large increases in activity of CS, SDH, and PFK indicate that the training program resulted in considerable adaptation at the muscle level in both legs. It is also apparent that the hypoxic condition combined with exercise training resulted in a significant increase in CS activity over and above that which occurred with the same training under normoxic conditions. SDH activity also increased approximately 28% more in the hypoxically-trained leg than in the normoxically-trained leg, but this difference was not statistically significant. Although one might normally expect changes in one oxidative enzyme to parallel changes in another, we have previously noted greater changes in CS activity than in SDH activity in a training study of similar duration and intensity (23). The changes in PFK activity were somewhat surprising and may be related to our inclusion of high intensity interval training in the final weeks of the program.

Although capillary density, capillary/fiber ratio, and mitochondrial volume density tended to be higher following training, the magnitude of these changes was not statistically significant. Again, although one normally expects increased oxidative enzyme activity to be closely coupled with an increase in mitochondial volume density (13,14), we have previously observed significant increases in CS activity with training in the absence of significant increases in mitochondial volume density(23). We interpret this greater sensitivity of CS as an oxidative marker as being due to the relatively lower precision of the morphometric technique for quantifying mitochondial density.

Our finding that ˙VO2max increased to the same extent for each leg indicates that the enhanced enzymatic adaptations in the hypoxically trained leg had little or no effect on ˙VO2max. This result is consistent with the commonly held belief that ˙VO2max is primarily determined by an individual's maximum cardiac output(26), whereas MAC is affected to a greater extent by muscle respiratory capacity (13). We did not measure cardiac output in the present study, but since HR was lower at the same submaximal power outputs following training, it is probable that training-induced increases in stroke volume, coupled with the slight increase in hemoglobin concentration, resulted in increased oxygen delivery to the muscles in the post-trained state.

The more than four- and five-fold increases in MAC which were found for the normoxically- and hypoxically-trained legs, respectively, were considerably greater than expected. Since ˙VO2max increased for both legs, part of this improvement can be attributed to the fact that the same absolute power output (95% of the pre-training ˙VO2max) represented a lower relative intensity following training. In addition, our subjects probably improved their mechanical efficiency for unilateral cycling as a result of the training period since ˙VO2 during the MAC test was lower(P < 0.05) following training. The combined effects of an improvement in ˙VO2max and unilateral cycling economy probably inordinately prolonged fatigue time (in excess of 90 min in some subjects) to the extent that the factors causing fatigue may have been quite different in the two tests. Consequently, the validity of this performance test as a true measure of the improvement in aerobic work capacity is questionable. However, since ˙VO2max increased to the same extent in each leg and each leg did the same absolute amount of training, between-leg comparisons in the post-trained state are justifiable. Time to fatigue was approximately 4 min 7 s longer in the hypoxically-trained leg than in the normoxically-trained leg, but this difference was not statistically significant. We are thus left to conclude that either the changes that occurred in CS concentration had little or no effect on MAC or that the power output that was used in our MAC test was not appropriate for discriminating changes in exercise capacity.

Our data confirm those of Terrados et al. (29) indicating that training under a moderate hypobaric hypoxic condition increases CS activity to a greater extent than does the same amount of training under a normoxic condition. Since even extreme chronic hypobaric hypoxia on its own does not increase oxidative enzyme activity and mitochondial density (11,21), the results of the present study indicate that the combination of exercise training with a moderate hypoxic environment provides an enhanced stimulus for adaptation and not the hypoxia per se. Theoretically, one might expect these enzymatic changes to also enhance exercise performance in the hypoxically-trained leg (18,29). In this regard, however, our results were inconclusive and may have been obscured by the method that was used to quantify exercise capacity.

In summary, our results indicate that training under moderate normobaric hypoxic conditions results in a greater increase in muscle oxidative enzyme activity than does the same volume of training under normoxic conditions. These changes have little or no effect on ˙VO2max but may enhance aerobic exercise capacity. A training protocol in which subjects only experience hypoxia while they are training (as in the present study) may be superior to actual training at altitude. Although a sojourn at altitude may boost oxygen carrying capacity owing to elevated Hgb, such adaptations may be offset by maladaptations such as reduced maximal Qc(20,24), muscle atrophy(11,16,17,21), or reduced buffer capacity and anaerobic power(2,5,7,11). With a protocol like the one in the present study, the hypoxic exposure time is so brief that these negative affects are probably avoided. The method used to simulate altitude is inexpensive, adaptable to a number of sports, and does not necessitate transportation of athletes to altitude for training.

Figure 1-Single leg ˙VO2max before and after 8 wk of unilateral cycle ergometry training. The left panel indicates the leg that trained under a normoxic condition and the right panel indicates the leg that trained under a normobaric hypoxic condition. * denotes
Figure 1-Single leg ˙VO2max before and after 8 wk of unilateral cycle ergometry training. The left panel indicates the leg that trained under a normoxic condition and the right panel indicates the leg that trained under a normobaric hypoxic condition. * denotes:
P > 0.05. Values are means ± SE( N = 10).
Figure 2-Single leg maximum aerobic capacity (time to fatigue at 95% of pre-training ˙VO2max) following training. The left panel indicates the normoxically trained leg and the right panel the hypoxically trained leg. * denotes
Figure 2-Single leg maximum aerobic capacity (time to fatigue at 95% of pre-training ˙VO2max) following training. The left panel indicates the normoxically trained leg and the right panel the hypoxically trained leg. * denotes :
P > 0.05. Values are means ± SE ( N = 10).
Figure 3-Muscle enzyme activity of citrate synthase (CS), succinate dehydrogenase (SDH), and phosphofructokinase (PFK) before training and following training in the leg that trained under normoxic conditions and the leg that trained under hypoxic conditions. * denotes
Figure 3-Muscle enzyme activity of citrate synthase (CS), succinate dehydrogenase (SDH), and phosphofructokinase (PFK) before training and following training in the leg that trained under normoxic conditions and the leg that trained under hypoxic conditions. * denotes:
P > 0.05. All three enzymes showed a main effect for training ( P > 0.05). In addition CS activity was also significantly higher in the hypoxically trained leg than in the normoxically trained leg. Values are means ± SE, ( N = 10).


1. Adams, W. C., E. M. Bernauer, D. B. Dill, and J. B. Bormar, Jr. Effects of equivalent sea level and altitude training on˙VO2max and running performance. J. Appl. Physiol. 39:262-268, 1975.
2. Bender, P. R., M. G. Bertron, R. G. McCullough, et al. Decreased exercise muscle lactate release after high altitude acclimatization.J. Appl. Physiol. 67:1456-1462, 1989.
3. Bigard, A., A. Brunet, C. Y. Guezennec, and H. Monad. Skeletal muscle changes after endurance training at high altitude. J. Appl. Physiol. 71:2114-2121, 1991.
4. Billings, C. R., R. Bason, D. Mathews, and E. Fox. Cost of submaximal and maximal work during chronic exposure at 3,800 m. J. Appl. Physiol. 30:406-408, 1971.
5. Brooks, G. A., G. E. Butterfield, R. R. Wolfe, et al. Decreased reliance on lactate during exercise after acclimatization to 4,300 m. J. Appl. Physiol. 71:333-341, 1991.
6. Buskirk, E. R., J. Kollias, R. E. Akers, E. K. Prokop, and E. P. Reatequi. Maximal performances at altitude and on return from altitude in conditioned runners. J. Appl. Physiol. 23:259-266, 1967.
7. Cerretelli, P., A. Veicsteinas, and C. Marconi. Anaerobic metabolism at high altitude: the lactacis mechanism. In: High Altitude Physiology and Medicine, W. Brendel and R. A. Zink (Eds.). New York: Springer Verlag, pp. 94-102, 1983.
8. Essen, B., E. Jansson, J. Henriksson, A. W. Taylor, and B. Saltin. Metabolic characteristics of fiber types in human skeletal muscle.Acta Physiol. Scand. 95:153-165, 1975.
9. Faulkner, J. A., J. T. Daniels, and B. Balke. Effects of training at moderate altitude on physical performance capacity. J. Appl. Physiol. 23:85-89, 1967.
10. Green, H. J. Muscle metabolism in chronic hypoxia. In:Hypoxia, The Tolerable Limits. J. R. Sutton, C. S. Houston, and G. Coates (Eds.)., Indianapolis: Benchmark Press, pp. 101-120, 1988.
    11. Green, H. J., J. R. Sutton, P. M. Young, A. Cymerman, and C. S. Houston. Operation Everest: II. adaptations in human skeletal muscle. J. Appl. Physiol. 66:2454-2461, 1989.
    12. Henriksson, J., M. M. Chi, C. S. Hintz, et al. Chronic stimulation of mammalian muscle: changes in enzymes of six metabolic pathways.Am. J. Physiol. 251:C614-C632, 1986.
    13. Holloszy, J. O. and E. F. Coyle. Adaptations of skeletal muscle to endurance training and their metabolic consequences.J. Appl. Physiol. 56:831-838, 1984.
    14. Hoppeler, H. Exercise induced ultrastructural changes in skeletal muscle. Int. J. Sports Med. 7:187-204, 1986.
    15. Hoppeler, H., P. Luthi, H. Claassen, E. R. Weibel, and H. Howald. The ultrastructure of the normal human skeletal muscle: a morphometric analysis on untrained men, women, and well-trained orienteers.Pflugers Arch. 344:217-232, 1973.
    16. Hoppeler, H., E. Kleinert, C. Schlegal, et al. Morphological adaptations of human skeletal muscle to chronic hypoxia.Int. J. Sports Med. 11(Suppl 1):S3-S9, 1990.
    17. Howald, H., D. Pette, J. A. Simoneau, A. Uber, H. Hoppeler, and P. Cerretelli. Effects of chronic hypoxia on muscle enzyme activities. Int. J. Sports Med. 11(Suppl 1):S10-S14, 1990.
    18. Kaijser, L., C. J. Sundberg, O. Eiken, et al. Muscle oxidative capacity and work performance after training under local leg ischemia. J. Appl. Physiol. 69:785-787, 1990.
    19. Lowry, O. H. and J. V. Passonneau. A Flexible System of Enzymatic Analysis, 1st Ed. New York: Academic Press, 1972.
    20. MacDougall, J. D., W. G. Reddan, J. A. Dempsey, and H. Forester. Acute alterations in stroke volume during exercise at 3,100 m altitude. J. Human Erg. 5:103-111, 1976.
    21. MacDougall, J. D., H. J. Green, J. R. Sutton, et al. Operation Everest II. structural adaptations in skeletal muscle in response to extreme simulated altitude. Acta. Physiol. Scand. 142:421-427, 1991.
    22. Maher, J. R., L. C. Jones, and L. H. Hartley. Effects of high altitude exposure on submaximal endurance capacity of men. J. Appl. Physiol. 37:895-901, 1974.
    23. Sale, D. G., J. D. MacDougall, I. Jacobs, and S. Garner. Interaction between concurrent strength and endurance training.J. Appl. Physiol. 68:260-270, 1990.
    24. Saltin, B. Aerobic and anaerobic work capacity at an altitude of 2,250 m. In: The International Symposium on the Effects of Altitude on Physical Performance. R.F. Goddart (Ed.). Chicago: The Athletic Institute, pp. 97-102, 1967.
    25. Saltin, B., R. F. Grover, C. G. Blomqvist, L. H. Hartley, and R. L. Johnson. Maximum oxygen uptake and cardiac output after 2 weeks at 4,300 m. J. Appl. Physiol. 25:400-409, 1968.
    26. Saltin, B. and L. B. Rowell. Functional adaptations to physical activity and inactivity. Fed. Proc. 39:1506-1513, 1980.
    27. Stenberg, J., B. Ekblom, and R. Messin. Hemodynamic Response to Work at Simulated Altitude. J. Appl. Physiol. 21:1589-1593, 1966.
    28. Terrados, N., J. Melichna, C. Sylven, E. Jansson, and L. Kaijser. Effects of training at simulated altitude on performance and muscle metabolic capacity in competitive road cyclists. Eur. J. Appl. Physiol. Occup. Physiol. 57:203-209, 1988.
    29. Terrados, N., E. Jansson, C. Sylven, and L. Kaijser. Is hypoxia a stimulus for synthesis of oxidative enzymes and myoglobin? J. Appl. Physiol. 68:2369-2372, 1990.
    30. Weibel, E. R. Stereological Methods. Vol I: Practical Methods for Biological Morphometry. Toronto: Academic Press, 1979, Chs.4,6.


    ©1997The American College of Sports Medicine