Elite endurance athletes frequently reside and/or train at moderate altitudes to improve sea-level performance via adaptations acquired from the hypoxic stimulus (1). General recommendations regarding training at altitude suggest a reduction in absolute running speed, especially during the initial phase of a camp, to minimize the risk of overtraining and facilitate the acclimatization process (2,3). Previous research has shown, however, that the reduction in running speed at altitude is coupled with a lower oxygen flux, resulting in a potential deconditioning effect impairing subsequent sea-level performance (4,5). As such, the maintenance of absolute exercise intensity is likely to be an important factor contributing to improved sea-level performance after altitude training (6).
The physiological responses associated with submaximal and maximal aerobic exercise in hypoxia are well described (7,8). Due to the reduced partial pressure of oxygen at natural altitude, or reduced fraction of inspired oxygen in simulated altitude environments, maximal oxygen consumption (V˙O2max) is reduced (9,10). The decline of V˙O2max (~ 6% per 1000 m) increases with altitude (9) and has been observed in elite athletes at altitudes as low as 580 m (11).
Greater physiological and metabolic adjustments are required to maintain homeostasis and performance for a given absolute workload when exercise is performed at altitude compared with sea level (8). Due to the reduction in V˙O2max, training sessions in hypoxia at submaximal intensities are completed at a higher relative intensity than equivalent sessions in normoxia. Although submaximal V˙O2 has been shown to remain unchanged at increasing altitudes (12), this is at an increased overall physiological cost, as oxygen transport to the working muscles is maintained by increasing cardiac output (as a function of increased HR) and muscle blood flow, compensating for the reduction in arterial oxygen content (7,8).
At supramaximal intensities, to compensate for the reduced oxygen availability in hypoxia, the anaerobic contribution to exercise increases to maintain performance during sprint exercise (13,14,15). We have recently demonstrated (16) that running speed during self-paced intervals (lactate threshold, V˙O2max and middle-distance race pace) at 2100 m was reduced to different degrees, with the greatest reductions observed in those with the greatest aerobic contributions (i.e., threshold and V˙O2max), confirming previous findings in elite cyclists (6). This suggests that physiological responses attempting to compensate for the reduced oxygen availability at altitude are insufficient to maintain performance at certain intensities more so than others. However to our knowledge, studies investigating these responses at multiple training intensities having not been conducted. Furthermore, although the physiological responses to single bouts of submaximal and supramaximal exercise in hypoxia are generally well understood, studies investigating physiological responses during interval training (i.e., repeated bouts) are limited to supramaximal intensities (15,17).
Training with higher levels of oxygen flux characteristic of lower altitudes is typically viewed as beneficial in facilitating adaptation and improved performance (and a key reason for the recommendation of “Live High Train Low” over “Live High Train High”) (4,5). However, remaining at moderate altitudes for high-intensity training may result in greater levels of muscle deoxygenation, which has been proposed to stimulate muscular adaptations (e.g., improved muscle pH regulation, buffer capacity, and anaerobic glycolytic activity, and increased muscle blood perfusion, mitochondrial volume, and capillary density) (18,19,20,21,22). Such adaptations may also enhance competitive performance during middle-distance and distance events, where both aerobic and anaerobic contributions to performance are relevant (23). In practical terms, knowledge of how physiological responses differ when training at different intensities would inform intensity specific modifications to training sessions (e.g., stay high or descend to lower altitude, increase recoveries, modify pace) designed to maintain exercise intensity and oxygen flux, or amplify the anaerobic contribution to interval training at altitude.
We therefore sought to determine the effect of low (1400 m) and moderate (2100 m) normobaric hypoxia on V˙O2, anaerobic contribution and other physiological parameters during acute exercise at three different intensities in highly trained runners. To ensure any physiological differences observed were due to hypoxia, and not a lower self-paced intensity of exercise (6,16), constant work load intervals were prescribed to athletes at the same absolute running speed for each training intensity. Compared to exercise in normoxia (580 m), we hypothesized that V˙O2 would be lower, and anaerobic contribution higher across all training intensities at simulated altitudes of 1400 and 2100 m.
METHODS
Subjects
Eight highly trained male runners and triathletes (age, 25 ± 7 yr, body mass: 71 ± 5 kg) participated in the investigation. The investigation took place during the precompetition phase of the season with participants regularly engaged in training consisting of continuous and interval running 5 to 7 d·wk−1 and were habituated to running on a motorized treadmill. None of the participants had prolonged exposure to altitude in the 12 months before participating in the investigation. All procedures and risks were explained to participants before they provided written informed consent to participate. Ethical approval for the investigation was granted by the institutional ethics committee (University of Canberra-Human Research Ethics Committee ref. no. 16–233) and all procedures complied with the Declaration of Helsinki.
Study design
To assess the effect of low and moderate hypoxia on the physiological responses to interval running at three different intensities compared to normoxia, an unblinded, randomized, repeated measures design was employed. The simulated altitudes chosen for training sessions are typically used by elite endurance athletes during both Live High Train High and Live High Train Low altitude training (16,24). The altitude for each training session was known to participants to preserve ecological validity, as elite athletes engaged in altitude training are aware of the altitude at which they are exercising. Furthermore, it was determined that the physiological impact of blinding would be inconsequential as the running speeds were constant across altitudes and based on the incremental exercise test in normoxia (580 m).
The investigation took place in Canberra, Australia (elevation: 580 m), and participants were required to complete 12 exercise sessions on a motorized treadmill (pulsar 3p, h/p/cosmos, Germany) over a 5-wk period, with all sessions taking place in an environmental chamber (ATS-1000BLHP, Altitude Training Systems, Lidcombe, Australia). The first week involved completing three incremental exercise tests in normoxia (elevation: 580 m, FiO2: 0.21), as well as low (1400 m, FiO2: 0.195) and moderate (2100 m, FiO2: 0.18) simulated normobaric hypoxia, to characterize V˙O2max, velocity at V˙O2max (vV˙O2max), 4 mM lactate threshold (LT), and to prescribe running speeds for the subsequent interval sessions. Over the following 3 to 4 wk, participants completed three different interval training sessions: threshold (4 × 5 min with 2 min recovery), V˙O2max (8 × 90 s with 90 s recovery), and race pace (10 × 45 s with 1 min 45 s recovery) at each of the altitudes, for a total of nine sessions. Participants stood at rest during all recovery periods. Running speeds for these sessions were calculated from the incremental exercise test completed at 580 m; threshold, V˙O2max and race pace sessions were completed at 4 mM LT speed, vV˙O2max, and 110% of the vV˙O2max respectively. Participants maintained regular training commitments during the investigation but were instructed to refrain from strenuous exercise for the 12 h preceding each trial. For each participant, testing sessions were conducted at a similar time of day, with trials separated by at least 48 h.
Baseline trials—Incremental exercise testing
Each incremental exercise test to exhaustion comprised of four submaximal stages completed at 12, 14, 16 and 18 km·h−1 (0% gradient) for determination of V˙O2 and capillary blood lactate concentration ([BLa]), immediately followed by an incremental ramp to exhaustion to determine V˙O2max. The initial submaximal stage at 12 km·h−1 was 4 min in length, with the subsequent three stages each being 3 min in length. Immediately, after each submaximal workload, a small capillary blood sample was taken from the fingertip, to measure [BLa] (Lactate Pro, Arkray, Kyoto, Japan). After the completion of the fourth submaximal stage, the gradient of the treadmill was increased by 0.5% every 30 s, until the participant reached volitional exhaustion. A final capillary blood sample was taken 1 min after cessation of exercise to determine maximal [BLa]. Heart rate (Polar Electro, Kempele, Finland) was measured throughout the test. Expired gas was collected throughout both the submaximal and maximal portions of the test for determination of ventilation (V˙E), V˙O2, V˙CO2 and respiratory exchange ratio (RER), using a metabolic cart (True One 2400; ParvoMedics, USA). Before each test, the ParvoMedics system was calibrated with normoxic gas (20.93% O2 and 0.04% CO2) and a gas of known concentration (16.01% O2 and 4.00% CO2). A 3-L syringe was used to calibrate flow. Submaximal V˙O2 for each stage was indicated by mean V˙O2 during the final minute, and V˙O2max (mL·kg−1·min−1) as the highest 30 s value achieved during the incremental ramp portion of the test. We acknowledge that this may be taken to represent V˙O2peak (25), as evidenced by the RER values obtained (Table 1), however we will refer to this as V˙O2max throughout the manuscript for consistency and clarity in differentiating between peak V˙O2 obtained during the interval training sessions. Individual running speed at 4 mM [BLa] was calculated using freely available software (26), and vV˙O2max was calculated from the running speed:V˙O2 relationship using the four submaximal speeds.
Experimental trials
Participants completed three different interval training sessions at each of the three altitudes, for a total of nine sessions. Before each trial, participants completed a 15-min standardized warm-up in normoxia including 10 min of low-intensity continuous running, some stretches and mobility exercises, and strides. Participants then entered the environmental chamber for 10 min before commencing the interval session, and were fitted with a HR monitor, safety harness, nose-clip and Hans-Rudolph mouthpiece for collection of expired gas. To start the interval session, participants straddled the treadmill belt while it was brought up to speed, with the first interval starting with them lowering themselves onto the moving treadmill belt. At the end of each interval, participants lifted themselves clear of the treadmill belt where they remained straddling the moving belt for the duration of the recovery period. For safety purposes, the safety harness remained fitted throughout the session. V˙O2peak (highest 15-s value), total V˙O2 and accumulated oxygen deficit (AOD) were measured for each interval, which was then averaged across the session. Heart rate was measured throughout the test, and RPE was obtained after each interval (27). After each interval during the threshold sessions, and every second interval during the V˙O2max and race pace sessions, a capillary blood sample was taken from the fingertip to measure [BLa]. Additionally, RPE for each interval was divided by running speed to determine the ratio of perceived exertion to running velocity (i.e., RPE units per kilometer per hour running speed) (16).
Normobaric hypoxia
All trials were conducted in the same 100 m3 environmental chamber. The low and moderate simulated altitudes were achieved through nitrogen injection (flow rate, 1000 L·min−1, 89% nitrogen, 11% oxygen), creating a normobaric hypoxic environment. The room had an in-built barometric pressure compensation, with the percentage oxygen in the chamber adjusted to account for the ~ 10% lower barometric pressure compared to sea level of Canberra (altitude, 580 m). The percentage of inspired oxygen for 580, 1400, and 2100 m were 20.94% ± 0.05%, 19.45% ± 0.06% and 18.00% ± 0.08%, respectively. Mean temperature, pressure and relative humidity for the trials was 21.4°C ± 1.5°C, 708.4 ± 4.1 mm Hg and 53.6% ± 11.4%, respectively.
Calculation of accumulated oxygen deficit
The AOD (28) arising from each interval was calculated as the difference between estimated oxygen requirements of the work achieved (derived from the running speed: V˙O2 regression for each individual athlete) and the total V˙O2 consumed during each interval. AOD and total V˙O2 for each interval completed during a session were summated to give total AOD and V˙O2 for the session. The relative aerobic and anaerobic contributions of each interval were calculated as the percentage of measured V˙O2 compared with the predicted V˙O2 and then averaged to give a value for the session.
Statistical analysis
All statistical calculations were performed using performed using the SPSS statistical package version 23 (IBM, New York). Differences between altitudes for measures obtained during incremental exercise testing (V˙O2max, vV˙O2max, and 4 mM LT) were quantified using one-way ANOVA. To test changes in measured variables (V˙O2peak, V˙E, RER, V˙O2peak as percentage of altitude specific V˙O2max [% V˙O2max], total V˙O2, AOD, aerobic contribution, anaerobic contribution, HR, RPE, exertion/velocity ratio and [BLa]) within training sessions (threshold, V˙O2max and race pace) between altitudes (580, 1400, and 2100 m) and over time (4, 8, and 10 intervals for threshold, V˙O2max, and race pace, respectively), two-way (altitude–time) repeated measures ANOVA were performed. Data from submaximal speeds (12, 14, 16, and 18 km·h−1) during the incremental exercise test (V˙O2, VE, RER, HR, [BLa]) were similarly compared using two-way (altitude–speed) repeated-measures ANOVA. To compare between the three training sessions across the three altitudes, two-way (altitude–session) repeated measures ANOVA was also performed. ANOVA assumptions were verified preceding all statistical procedures; however, none of the data violated the assumption of sphericity. Where significant effects were established, pairwise differences were identified using the Bonferroni post hoc analysis procedure adjusted for multiple comparisons. P values less than 0.05 were considered statistically significant. Effect size was measured using partial eta-squared (η2) values with η2 > 0.06 representing a moderate effect and η2 > 0.14 a large effect. All values are expressed as means ± SD.
RESULTS
Incremental exercise testing
Summary data from the incremental exercise tests in each altitude are shown in Table 1. There were significant reductions in V˙O2max at simulated 1400 m (3.4% ± 2.5%, P = 0.02) and simulated 2100 m (7.3% ± 2.4%, P < 0.001) compared with 580 m, with a significant reduction observed from simulated altitudes of 1400 to 2100 m (4.0% ± 3.2%, P = 0.046). V˙O2 at each of the submaximal workloads (12, 14, 16, and 18 km·h−1) was not significantly different between altitudes (P > 0.05, Figure 1A). No significant differences were observed for running speed at 4 mM LT between altitudes, however vV˙O2max was significantly lower at 2100 m compared to both 580 and 1400 m (P = 0.003 and 0.004, respectively).
;)
FIGURE 1: Changes in (A) V˙O2, (B) HR, (C) [BLa], (D) minute ventilation at submaximal and maximal workloads during incremental exercise testing in normoxia (black), low (white) and moderate (gray) moderate hypoxia. *Significantly different to 580 m; #Significantly different to 1400 m. Significance, P < 0.05.
Threshold (4 × 5 min) sessions
Threshold intervals were completed at 17.5 ± 0.9 km·h−1 across all altitudes. A main effect of altitude was observed (P = 0.017, η2 = 0.44) for %V˙O2max, with intervals completed at 2100 m (89.9% ± 5.7%) significantly higher (P = 0.04, % change = 4.8% ± 4.0%) than 580 m (85.9% ± 5.7%). There were no significant differences (P = 0.69) between 580 and 1400 m (87.2% ± 6.0%). A significant altitude effect (P = 0.002, η2 = 0.59; Figure 2A) was observed for total V˙O2 (averaged over the four intervals), being significantly lower at 2100 m versus 580 m (270.9 ± 14.6 mL·kg−1 vs 283.4 ± 11.8 mL·kg−1, P = 0.003, % change = 4.4% ± 2.1%), with the difference between 1400 m (278.1 ± 15.7 mL·kg−1) and 2100 m tending toward significance (P = 0.065). An altitude effect (P = 0.001, η2 = 0.48) was observed for AOD, with higher values being observed at 1400 m (27.0 ± 17.4 mL·kg−1) and 2100 m (35.8 ± 15.8 mL·kg−1) compared with 580 m (23.7 ± 17.3 mL·kg−1; P vs 580 m = 1.000 and 0.001 respectively). A significant altitude effect was observed for aerobic contribution (P = 0.01, η2 = 0.48), being significantly lower at 2100 m (88.6% ± 4.0%) compared with 580 m (92.5% ± 4.7%; P = 0.002), but not significantly different between 580 and 1400 m (91.4% ± 4.8%; P = 1.000). Consequently, anaerobic contribution (Fig. 2B) was higher at 1400 m (8.6% ± 4.8%) and 2100 m (11.4% ± 4.0%) compared to 580 m (7.5% ± 4.7%), with this difference being significant at 2100 m (P = 0.002). There was a significant altitude by interval interaction (P = 0.05, η2 = 0.25) for [BLa]. [BLa] remained stable during all four intervals at 580 and 1400 m (P > 0.05) but was significantly higher after the fourth interval compared to the first at 2100 m (P = 0.038). At 2100 m, [BLa] was significantly higher than all intervals at 580 m (P < 0.05) and after intervals three and four at 1400 m (P < 0.05). Significant effects for time were observed for V˙O2peak (P = 0.001, η2 = 0.67), V˙E (P = 0.001, η2 = 0.54), HR (P < 0.001, η2 = 0.96), RPE (P < 0.001, η2 = 0.71) and RPE·Speed−1 (P < 0.001, η2 = 0.72), with values increasing with each interval.
FIGURE 2: Changes in (A) total O2 consumption and (B) anaerobic contribution during threshold training (4 × 5 min at 4 mM with 2-min recovery) at 580, 1400 and 2100 m. a, Significant altitude effect; b, significant time effect. Significance, P < 0.05.
V˙O2max (8 × 90 s) sessions
V˙O2max intervals were completed at 20.1 ± 1.3 km·h−1 across all altitudes, which was 1.0 and 5.8% greater than the vV˙O2 peak at 1400 m and 2100 m, respectively. The %V˙O2max at each altitude was 90.9% ± 4.5% (580 m), 90.3% ± 4.4% (1400 m) and 92.5% ± 3.7% (2100 m), with no significant effect for altitude observed (P = 0.511, η2 = 0.09). A main effect of altitude (P = 0.003, η2 = 0.57) was observed for V˙O2peak, which compared with 580 m (65.4 ± 4.2 mL·kg−1·min−1) was significantly lower at 2100 m (61.6 ± 3.1 mL·kg·min−1; P = 0.013) but not 1400 m (62.8 ± 4.1 mL·kg−1·min−1; P = 0.121). Compared to 580 m (82.8 ± 6.9 mL·kg−1) total V˙O2 (averaged over eight intervals) was lower at both 1400 m (79.7 ± 6.4 mL·kg−1, % change = −3.6% ± 4.0%) and 2100 m (78.9 ± 5.1 mL·kg−1, % change = −4.5% ± 3.8%), with altitude (P = 0.07, η2 = 0.31) and interaction (P = 0.06, η2 = 0.197) effects both approaching significance (Fig. 3A). Significant altitude effects were observed for AOD (P = 0.047, η2 = 0.35), aerobic contribution (P = 0.03, η2 = 0.396), and anaerobic contribution (P = 0.03, η2 = 0.396). AOD at 1400 m (25.1 ± 6.0 mL·kg−1) and 2100 m (26.3 ± 5.5 mL·kg−1) was higher than 580 m (22.5.1 ± 4.6 mL·kg−1), with this difference approaching significance at 2100 m (P = 0.06). The aerobic contribution was significantly lower at 2100 m (75.2% ± 3.1%) compared to 580 m (78.7% ± 3.4%; P = 0.047), but not significantly different between 580 and 1400 m (76.2% ± 4.2%; P = 0.37). Consequently, anaerobic contribution at 2100 m (24.8% ± 3.1%) was significantly higher compared to 580 m (21.3% ± 3.4%; P = 0.047); however, there was no significant difference with 1400 m (23.8% ± 4.2%; Fig. 3B). A significant interaction effect was observed for RER (P < 0.001, η2 = 0.36). During intervals one and two, RER was higher at 2100 m than both 580 m and 1400 m (P < 0.05), with no significant difference between altitudes for the remaining intervals. Significant effects of time were observed for V˙O2peak (P < 0.001, η2 = 0.77), V˙E (P < 0.001, η2 = 0.58), [BLa] (P < 0.001, η2 = 0.62), HR (P < 0.001, η2 = 0.89), RPE (P < 0.001, η2 = 0.74) and RPE·km−1·h−1 (P < 0.001, η2 = 0.75), with values increasing with each interval.
FIGURE 3: Changes in physiological parameters during V˙O2max (8 × 90 s at vV˙O2 peak with 90 s recovery) training at 580, 1400 and 2100 m. (A) Total O2 consumption. (B) Anaerobic contribution, a, altitude effect, P < 0.1; b, significant time effect; c, significant altitude effect. Significance, P < 0.05.
Race pace (10 × 45 s) sessions
Race pace intervals were completed at 22.1 ± 1.4 km·h−1 across all altitudes, which was 1.4 and 5.7% greater than 110% of the vV˙O2 peak at 1400 m and 2100 m respectively. The %V˙O2max at each altitude was 83.7% ± 3.0% (580 m), 85.7% ± 5.2% (1400 m) and 86.9% ± 3.4% (2100 m), with no significant effect of altitude observed (P = 0.182, η2 = 0.216). A main effect of altitude (P = 0.019, η2 = 0.43) was observed for V˙O2peak, whereby it was lower at 2100 m (58.0 ± 3.8 mL·kg−1·min−1; P = 0.043) but not 1400 m (59.6 ± 4.9 mL·kg−1·min−1; P = 0.304) compared with 580 m (60.3 ± 4.2 mL·kg−1·min−1). No significant altitude effects were found for total V˙O2 (P = 0.273, η2 = 0.169), AOD (P = 0.56, η2 = 0.08), aerobic and anaerobic contribution (P = 0.47, η2 = 0.10), and [BLa] (P = 0.18, η2 = 0.22) (Fig. 4). Significant altitude (P = 0.015, η2 = 0.453) and interaction (P = 0.021, η2 = 0.214) effects were observed for RPE·km−1·h−1. RPE·km−1·h−1 remained relatively stable at 1400 m and with no significant differences between any intervals (P > 0.05); however, at 580 m RPE·km−1·h−1 for interval one was significantly lower compared to intervals six, seven, and eight (P = 0.03). At 2100 m, RPE·km−1·h−1 for intervals seven, nine, and ten was significantly higher than intervals one through five (P < 0.05). Significant effects for time were observed for V˙O2peak (P < 0.001, η2 = 0.62), V˙E (P < 0.001, η2 = 0.69), [BLa] (P < 0.001, η2 = 0.58), HR (P < 0.001, η2 = 0.82) and RPE (P < 0.001, η2 = 0.77), with values increasing over the course of the session.
FIGURE 4: Changes in physiological parameters during race pace (10 × 45 s at 110% vV˙O2 peak with 105 s recovery) training at 580, 1400, and 2100 m. (A) Total O2 consumption. (B) Anaerobic contribution. a, significant time effect. Significance, P < 0.05.
TABLE 1: Summary data from the maximal incremental exercise tests.
Between session comparisons
A comparison of the mean values for each session at the different altitudes is presented in Table 2. A significant interaction effect (P < 0.001, η2 = 0.74) was observed for mean relative intensity of exercise (running speed relative to percent of altitude specific vV˙O2max). Relative intensity at 2100 m (92% ± 2%, 106% ± 2% and 116% ± 3% for threshold, V˙O2max and race pace sessions respectively) was significantly higher than both 1400 m (89% ± 2%, 101% ± 3%, and 111% ± 3%) and 580 m (88% ± 3%, 100% ± 0%, and 110% ± 0%) for all three sessions (P < 0.01). However, there were no differences between 580 and 1400 m (P > 0.80). Significant differences between sessions (P < 0.001, η2 = 0.95) were observed for anaerobic contribution, with race pace sessions (mean anaerobic contribution across altitudes = 39% ± 5%) significantly higher (P < 0.001) than both threshold (9% ± 4%) and V˙O2max (23% ± 4%) sessions (threshold vs V˙O2max, P < 0.001). A significant effect between sessions was also found for total V˙O2 (normalized to mL·kg−1·min−1), with race pace sessions (mean total V˙O2 across altitudes = 47.0 ± 6.5 mL·kg−1·min−1) significantly lower (P < 0.01) than both threshold (55.5 ± 2.8 mL·kg−1·min−1) and V˙O2max (53.6 ± 4.1 mL·kg−1·min−1) sessions.
TABLE 2: Mean values per interval for the threshold, V˙O2max and race pace sessions in normoxic (580 m), low (1400 m), and moderate (2100 m) hypoxic conditions.
DISCUSSION
We sought to determine the effect of low and moderate simulated normobaric hypoxia on total V˙O2 and anaerobic contribution to interval running at different intensities, but at the same absolute running speed across altitudes, in highly trained athletes. We confirm the results of previous research (9,29,30) suggesting that for high-intensity exercise completed at the same absolute work rate, total V˙O2 is reduced at moderate simulated altitudes compared to sea level. Consequently, AOD and anaerobic contribution were higher in hypoxia, further corroborating previous findings (13,14,15). However, we extend these findings by showing for the first time that the magnitude of these differences is dependent on exercise intensity, with larger changes observed in training sessions with a greater aerobic contribution (i.e., threshold and V˙O2max), with no significant differences observed for race pace sessions between FiO2 conditions, in contrast with our hypothesis. Moreover, our data reveal that threshold and V˙O2max sessions at a simulated altitude of 2100 m, but not 1400 m, induced significant physiological differences (i.e., higher [BLa], anaerobic contribution) compared to interval exercise in normoxia. Finally, we add to previous literature showing altered total V˙O2 and AOD during single interval bouts of high-intensity exercise in hypoxia by demonstrating similar responses during repeated interval training sessions specific to endurance athletes. It has previously been suggested that athletes may not be able to sustain adequate oxygen flux during aerobic exercise at altitude (4,5); our findings confirm these assertions for certain training intensities and therefore have implications for the prescription of interval sessions for athletes completing altitude training.
Total V˙O2 during acute exercise at simulated altitude
Previous research examining V˙O2 responses during single bout, heavy intensity exercise has shown that V˙O2 and time to exhaustion are reduced at moderate altitudes when completed at equivalent work rates to normoxia (9,29,30). We sought to extend upon these findings by investigating V˙O2 responses during acute exercise in hypoxia at intensities frequently used by elite endurance athletes (16,31,32). We observed total V˙O2 during threshold and V˙O2max intensity (determined in normoxia) intervals was 4.4% and 4.7% lower at a simulated altitude of 2100 m compared with 580 m, with no differences for these two intensities between 580 m and simulated 1400 m. Furthermore, due to the hypoxia-induced reduction in V˙O2max (Fig. 1A), higher altitude-specific %V˙O2max was achieved at moderate altitudes compared with that in normoxia, confirming previous findings (30).
No differences in total V˙O2 were observed for race pace intervals at simulated altitude of either 1400 or 2100 m compared with normoxia (Fig. 4A). Similarly, in a study of physical education students completing 5 × 400 m intervals at 90% of their maximal 400 m speed at 690 and 2320 m natural altitude, no differences were observed in V˙O2 (17). Alternatively, in highly trained middle-distance runners, V˙O2 during intermittent 20-s shuttle runs (increasing in speed from 13.5 to 25 km·h−1) interspersed with 100-s recovery was significantly reduced from 18 to 25 km·h−1 at 2500 m hypobaric hypoxia compared with sea level (15). Previous research has shown that although the amplitude and time constant of the V˙O2 slow component are unaffected in hypoxia, the time constant of the primary rise in V˙O2 (i.e., onset of exercise) is slower in hypoxia, with no significant change in amplitude (33). In comparison to our study, where 45-s intervals were selected for race pace sessions, the relatively short interval length of previous research (15) may have accounted for the lower V˙O2 observed, with insufficient time to reach to amplitude of the initial V˙O2 response. Interestingly, in the study of Ogawa and colleagues (15), accumulated V˙O2 when all intervals were summated was not significantly different between normoxia and hypoxia, which is similar to our findings. However, accumulated V˙O2 during recovery intervals was significantly lower at each intensity in hypoxia, leading to total accumulated V˙O2 (run + recovery) being lower in hypoxia (15). Although V˙O2 during recovery between intervals was not measured in the current investigation, we speculate that the previously reported impairment of self-paced intermittent exercise in hypoxia (6,16,34) may be driven to an extent by a lower accumulated V˙O2 during recovery (and thus greater AOD overall), along with the lower V˙O2 during exercise intervals at certain intensities. As such, the extension of recovery interval length during exercise in hypoxia could be an important strategy to maintain V˙O2 and performance during high-intensity exercise in hypoxia (1,6); however, future investigations measuring V˙O2 during recovery intervals of varying length would be required to confirm these assertions.
Accumulated oxygen deficit and anaerobic contribution
The reduction in the rate of oxygen uptake to steady state at altitude effectively increases the anaerobic contribution to exercise at all distances (7). Accordingly, we observed the AOD at a simulated altitude of 2100 m to be 51%, 17%, and 4% higher than at 580 m for threshold, V˙O2max and race pace sessions respectively, and 14%, 12%, and 3% higher at simulated 1400 m. Previous research investigating the importance of aerobic metabolism to a single all-out running sprint over durations from 15 to 180 s showed that under hypoxic conditions (FiO2 = 13%, 3500 m), anaerobic energy release was higher than in normoxia at all durations, with the largest differences (up to 18%) being observed at sprint durations of 60 to 90 s (14). Additionally, anaerobic energy release during 40-s Wingate tests has been reported to be 9% higher at a simulated altitude of 2000 m compared with normoxia (35). Alternatively, Friedmann and colleagues (29) demonstrated no significant differences in maximal AOD (MAOD) between sea level and a simulated altitude of 2500 m during exhaustive exercise at 110% to 120% of V˙O2max, lasting 2 to 3 min, designed to induce MAOD (28). They suggested the discrepancy between their findings and those previous (14,35) was due to the protocol length selected, as Wingate tests and shorter duration sprints are insufficient in length to reach MAOD in normoxia; therefore, the capacity to increase anaerobic contribution is present in hypoxia. In the current study, the greatest increases to anaerobic contribution in hypoxia were observed in the sessions featuring the greatest aerobic contribution to exercise (i.e., threshold and V˙O2max), as the combination of duration and intensity prescribed were unlikely to elicit MAOD. Additionally, blood lactate levels were higher for these two intensities at simulated 2100 m. Meanwhile, we observed no significant changes to AOD during the race pace sessions, perhaps unsurprising given it was performed at the highest intensity, with the highest anaerobic contribution (Table 2). Similarly, Feriche and colleagues (17) reported no significant differences in AOD when athletes completed 5 × 400 m intervals at 90% maximum 400 m speed in normoxia or hypoxia.
Limitations and future directions
A limitation of the current investigation is the absence of arterial oxygen saturation (SaO2) measured during exercise—either directly or estimated via pulse oximetry (SpO2). After extensive piloting, it was determined that the movement induced by running prevented the SpO2 device being secured sufficiently well on the fingertip and thus invalidated the collection of data; unreliable results were often obtained, along with the occasional instance where an SpO2 value was not displayed, even by the end of the rest interval. Measurement of SpO2 during exercise can be unreliable and inaccurate, especially when using many forms of commercially available pulse oximetry equipment which may not perform well during exercise, given increased blood flow, movement and vibration (36). Several previous studies have quantified the desaturation of arterial blood during treadmill running (e.g. 37,38) with SpO2 measurements collected from the ear, where the movement artefact induced by running would be substantially less than at the finger, thus providing reliable and accurate measurements. Desaturation of arterial blood during exercise in hypoxia is well quantified in the literature (6,37,38), with elite athletes displaying reduced measures of SaO2 during maximal exercise in normoxia, and both submaximal and maximal exercise in hypoxia (11). Furthermore, there is a tight coupling between SaO2 and V˙O2 during exercise, as well as a significant relationship between the degree of SaO2 decline at maximal exercise, and the decline in both V˙O2max and performance during aerobically dominant events in moderate hypoxia (6,38). The responses may also vary depending on the intensity of a training session. For example, in elite female cyclists exercising at 2100 m, SaO2 was lower during longer self-paced intervals (3 × 10 min) but remained unchanged during shorter repeated sprinting (3 × 6 × 15 s), relative to the same exercise completed in normoxia (6). Our findings showed unchanged V˙O2 during race pace exercise at both simulated altitudes of 1400 and 2100 m, however V˙O2 during both threshold and V˙O2max training was lower at both altitudes compared to normoxia. We thus speculate that SaO2 measures would have mirrored these findings; however, confirmatory data monitoring SaO2 during interval training in runners are required. Based on our experience in the current study and data from previous studies, we recommend pulse oximeters attached to the ear instead of the fingertip be used to collect this noninvasively.
When interpreting and practically applying the findings of the current investigation, it is important to acknowledge the physiological and performance differences that may occur between exercise in normobaric versus hypobaric hypoxia, a topic of contention in the literature (39,40). A recent systematic review revealed a lower minute V˙E and elevated symptoms of acute mountain sickness during exposure to hypobaric hypoxia compared to normobaric hypoxia (41). Additionally, time trial performance is impaired to a greater extent when cycling in hypobaric than normobaric hypoxia, relative to sea level (42,43). However, the mechanisms for this impairment are unclear with SaO2 reported as significantly lower in hypobaric hypoxia (43) and unchanged between hypoxic conditions (42), though the latter investigation used a more severe level of hypoxia (4300 m vs 3450 m). Based on our previous findings of impaired performance during self-paced intervals at 2100 m hypobaric hypoxia (16), one may expect performance to be similarly impaired in the current investigation, however athletes were able to complete all intervals at the prescribed normoxic training intensities, likely due to an increased anaerobic contribution, particularly during threshold and V˙O2max sessions. Together, these observations suggest the physiological responses to acute exercise in hypobaric hypoxia merit further interrogation, with important implications for prescription of training during natural altitude camps.
Practical recommendations
Previous recommendations for altitude training have suggested a Live High Train Low method (5), or Live High Train Low and High paradigm, whereby high-intensity training is completed at a lower altitude, with low-intensity training remaining at the residential altitude, usually between 2000 and 2500 m (44). Based on our current and previous findings (16), we would suggest some modifications to this frequently used strategy in athletes habituated to altitude training. For threshold and maximal aerobic sessions, descending to 1400 m would be beneficial in helping to defend oxygen flux. However, if the desired outcome was to increase the anaerobic contribution to exercise, which is relevant for middle-distance performance (23), especially in the absence of an altitude-induced increase in haemoglobin mass (45), remaining at moderate altitude and increasing the length of recovery intervals to help maintain running speed would be suitable. Middle-distance race-pace sessions may be completed at moderate altitude with little change in physiological stimulus compared to sea-level training. These updated recommendations are comparable to the recently proposed Live High Train Low and High approach proposed for team sport athletes (46), involving some high-intensity exercise completed at moderate altitudes. However, the ergogenic potential of such a strategy for sea-level performance has yet to be confirmed in elite endurance athletes.
CONCLUSIONS
The results of the investigation show completing high-intensity interval running at a simulated altitude of 2100 m, but not 1400 m, is likely to induce a lower V˙O2 and greater anaerobic contribution to exercise when compared to training at 580 m, with the greatest effects observed for threshold and maximal aerobic sessions.
The authors thank the University of Canberra for their financial support. A special thanks to the athletes who participated in the investigation for their time and efforts. The authors have no conflicts of interests, and the results of the present study do not constitute endorsement by the American College of Sports Medicine. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.
REFERENCES
1. Saunders PU, Pyne DB, Gore CJ. Endurance training at altitude.
High Alt Med Biol. 2009;10(2):135–48.
2. Baumann I, Bonov P, Daniels J, Lange G. Round table: high altitude training symposium.
New Studies in Athletics. 1994;9(2):23–35.
3. Wilber RL.
Altitude Training and Athletic Performance. Human Kinetics; 2004. pp. 140–58.
4. Chapman RF, Stray-Gundersen J, Levine BD. Individual variation in response to altitude training.
J Appl Physiol. 1998;85(4):1448–56.
5. Levine BD, Stray-Gundersen J. “Living high-training low”: effect of moderate-altitude acclimatization with low-altitude training on performance.
J Appl Physiol. 1997;83(1):102–12.
6. Brosnan MJ, Martin DT, Hahn AG, Gore CJ, Hawley JA. Impaired interval exercise responses in elite female cyclists at moderate simulated altitude.
J Appl Physiol (1985). 2000;89(5):1819–24.
7. Fulco CS, Rock PB, Cymerman A. Maximal and submaximal exercise performance at altitude.
Aviat Space Environ Med. 1998;69(8):793–801.
8. Mazzeo RS. Physiological responses to exercise at altitude.
Sports Med. 2008;38(1):1–8.
9. Wehrlin JP, Hallén J. V˙O
2 max and performance with increasing altitude in endurance athletes.
Eur J Appl Physiol. 2006;96(4):404–12.
10. MacInnis MJ, Nugent SF, Macleod KE, Lohse KR. Methods to estimate VO
2 max upon acute hypoxia exposure.
Med Sci Sports Exerc. 2015;47(9):1869–76.
11. Gore CJ, Hahn AG, Scroop GC, et al. Increased arterial desaturation in trained cyclists during maximal exercise at 580 m altitude.
J Appl Physiol. 1996;80(6):2204–10.
12. Clark SA, Bourdon PC, Schmidt W, et al. The effect of acute simulated moderate altitude on power, performance and pacing strategies in well-trained cyclists.
Eur J Appl Physiol. 2007;102(1):45–55.
13. McLellan TM, Kavanagh MF, Jacobs I. The effect of hypoxia on performance during 30 s or 45 s of supramaximal exercise.
Eur J Appl Physiol Occup Physiol. 1990;60(2):155–61.
14. Weyand PG, Lee CS, Martinez-Ruiz R, Bundle MW, Bellizzi MJ, Wright S. High-speed running performance is largely unaffected by hypoxic reductions in aerobic power.
J Appl Physiol. 1999;86(6):2059–64.
15. Ogawa T, Hayashi K, Ichinose M, Wada H, Nishiyasu T. Metabolic response during intermittent graded sprint running in moderate hypobaric hypoxia in competitive middle-distance runners.
Eur J Appl Physiol. 2007;99(1):39–46.
16. Sharma AP, Saunders PU, Garvican-Lewis LA, et al. Training at 2100 m altitude affects running speed and session RPE at different intensities in elite middle-distance runners.
Int J Sports Physiol Perform. 2017;12(2 Suppl):S2-147-S2-152.
17. Feriche B, Delgado M, Calderón C, Lisbona O, et al. The effect of acute moderate hypoxia on accumulated oxygen deficit during intermittent exercise in nonacclimatized men.
J Strength Cond Res. 2007;21(2):413–8.
18. Mizuno MA, Juel CA, Bro-Rasmussen T, et al. Limb skeletal muscle adaptation in athletes after training at altitude.
J Appl Physiol. 1990;68(2):496–502.
19. 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.
20. Gore CJ, Clark SA, Saunders PU. Nonhematological mechanisms of improved sea-level performance after hypoxic exposure.
Med Sci Sports Exerc. 2007;39(9):1600–9.
21. Lundby C, Calbet JA, Robach P. The response of human skeletal muscle tissue to hypoxia.
Cell Mol Life Sci. 2009;66(22):3615–23.
22. Faiss R, Léger B, Vesin JM, et al. Significant molecular and systemic adaptations after repeated sprint training in hypoxia.
PloS One. 2013;8(2):e56522.
23. Gastin PB. Energy system interaction and relative contribution during maximal exercise.
Sports Med. 2001;31(10):725–41.
24. Carr AJ, Saunders PU, Vallance BS, Garvican-Lewis LA, Gore CJ. Increased hypoxic dose after training at low altitude with 9h per night at 3000m normobaric hypoxia.
J Sports Sci Med. 2015;14(4):776–82.
25. Poole DC, Jones AM. Measurement of the maximum oxygen uptake V˙O
2 max: V˙O
2 peak is no longer acceptable.
J Appl Physiol. 2017;122(4):997–1002.
26. Newell J, Higgins D, Madden N, et al. Software for calculating blood lactate endurance markers.
J Sports Sci. 2007;25(12):1403–9.
27. Borg G. Perceived exertion as an indicator of somatic stress.
Scand J Rehabil Med. 1970;2(2):92–8.
28. Medbo JI, Mohn AC, Tabata I, Bahr R, Vaage O, Sejersted OM. Anaerobic capacity determined by maximal accumulated O
2 deficit.
J Appl Physiol. 1988;64(1):50–60.
29. Friedmann B, Frese F, Menold E, Bärtsch P. Effects of acute moderate hypoxia on anaerobic capacity in endurance-trained runners.
Eur J Appl Physiol. 2007;101(1):67–73.
30. Black MI, Potter CR, Corbett J, Clark CC, Draper SB. Maximal oxygen uptake is achieved in hypoxia but not normoxia during an exhaustive severe intensity run.
Front Physiol. 2017;8:96.
31. Billat LV. Interval training for performance: a scientific and empirical practice. Special recommendations for middle- and long-distance running. Part I: aerobic interval training.
Sports Med. 2001;31(1):13–31.
32. Tjelta LI. The training of international level distance runners.
Int J Sports Sci Coach. 2016;11(1):122–34.
33. Engelen M, Porszasz J, Riley M, Wasserman K, Maehara K, Barstow TJ. Effects of hypoxic hypoxia on O
2 uptake and heart rate kinetics during heavy exercise.
J Appl Physiol. 1996;81(6):2500–8.
34. Deb SK, Brown DR, Gough LA, et al. Quantifying the effects of acute hypoxic exposure on exercise performance and capacity: a systematic review and meta-regression.
Eur J Sport Sci. 2018;18(2):243–56.
35. Ogura Y, Katamoto S, Uchimaru J, Takahashi K, Naito H. Effects of low and high levels of moderate hypoxia on anaerobic energy release during supramaximal cycle exercise.
Eur J Appl Physiol. 2006;98(1):41–7.
36. Yamaya Y, Bogaard HJ, Wagner PD, Niizeki K, Hopkins SR. Validity of pulse oximetry during maximal exercise in normoxia, hypoxia, and hyperoxia.
J Appl Physiol. 2002;92(1):162–8.
37. Chapman RF, Emery M, Stager JM. Degree of arterial desaturation in normoxia influences VO
2 max decline in mild hypoxia.
Med Sci Sports Exerc. 1999;31(5):658–63.
38. Chapman RF, Stager JM, Tanner DA, Stray-Gundersen J, Levine BD. Impairment of 3000-m run time at altitude is influenced by arterial oxyhemoglobin saturation.
Med Sci Sports Exerc. 2011;43(9):1649–56.
39. Millet GP, Faiss R, Pialoux V. Point: hypobaric hypoxia induces different physiological responses from normobaric hypoxia.
J Appl Physiol (1985). 2012;112(10):1783–4.
40. Mounier R, Brugniaux JV. Counterpoint: hypobaric hypoxia does not induce different responses from normobaric hypoxia.
J Appl Physiol. 2012;112(10):1784–6.
41. Coppel J, Hennis P, Gilbert-Kawai E, Grocott MP. The physiological effects of hypobaric hypoxia versus normobaric hypoxia: a systematic review of crossover trials.
Extrem Physiol Med. 2015;4:2.
42. Beidleman BA, Fulco CS, Staab JE, Andrew SP, Muza SR. Cycling performance decrement is greater in hypobaric versus normobaric hypoxia.
Extrem Physiol Med. 2014;3:8.
43. Saugy JJ, Rupp T, Faiss R, Lamon A, Bourdillon N, Millet GP. Cycling time trial is more altered in hypobaric than normobaric hypoxia.
Med Sci Sports Exerc. 2016;48(4):680–8.
44. Stray-Gundersen J, Chapman RF, Levine BD. “Living high-training low” altitude training improves sea level performance in male and female elite runners.
J Appl Physiol. 2001;91(3):1113–20.
45. Garvican LA, Pottgiesser T, Martin DT, Schumacher YO, Barras M, Gore CJ. The contribution of haemoglobin mass to increases in cycling performance induced by simulated LHTL.
Eur J Appl Physiol. 2011;111(6):1089–101.
46. Brocherie F, Millet GP, Hauser A, et al. “Live high-train low and high” hypoxic training improves team-sport performance.
Med Sci Sports Exerc. 2015;47(10):2140–9.
