Under hypoxic conditions, the reduced cellular oxygen partial pressure (PO2) results in an increased activity of the “oxy-gene” hypoxia-inducible factor 1 (HIF-1) (15). Several HIF-1 target genes have been identified, including those encoding erythropoietin, glucose transporters, glycolytic enzymes, and vascular endothelial growth factor (27), and there is some evidence for direct effects on mitochondrial function (20,29). As such, interventions that expose athletes to altitude or hypoxia for varying durations are commonly used. The traditional approach involving spending 20 h·d−1 or more for 3 weeks or more, at physical altitude, may enhance subsequent sea level endurance performance through hematologically or non hematologically mediated improvements in O2 transport and utilization (9,18). An alternative approach is for athletes to breathe a hypoxic inspirate during some of their usual exercise training, while living in normoxia, termed intermittent hypoxic training (IHT). Although IHT does not provide a sufficient exposure required for complete acclimatization (21), there have been some noteworthy adaptations reported to date.
In an early study using highly trained cyclists, Terrados et al. (30) found that 3–4 weeks of moderate and heavy intensity IHT at an altitude simulation of 2300 m resulted in significantly lower submaximal exercise blood [lactate] [BLa] and significantly enhanced capillarization, compared with a normoxic trained control group (CONT). Furthermore, cycling work capacity improved significantly more after IHT than after CONT, when tested in hypoxia. Although work capacity improved in the IHT group on average by 33% in normoxia, this was not significantly different to the 22% improvement in CONT (30). Similarly, Roels et al. (26) exposed moderately trained cyclists to 7 weeks of heavy intensity IHT at an altitude simulation of ∼3000 m. Maximal pulmonary O2 uptake (V[Combining Dot Above]O2max) significantly increased by ∼9% after IHT, compared with a mean increase of ∼5% after CONT, but this between group difference was not statistically significant. Moreover, the increased V[Combining Dot Above]O2max after IHT did not correspond to a greater endurance cycling performance improvement than after CONT.
Another study, by Dufour et al. (5), involved 18 moderately trained men completing 6 weeks of moderate intensity IHT or CONT, at the speed corresponding to the second ventilatory threshold, at the speed corresponding to the second ventilatory threshold (2). Despite the similar training load, the IHT group experienced a significant ∼5% V[Combining Dot Above]O2max increase, which did not change in CONT group, and time to exhaustion at the V[Combining Dot Above]O2max velocity also improved significantly more after IHT (+35%) compared with CONT (+10%) (5). Given that the hypoxic doses of <2 h·d−1 in Roels et al. (26) and <1 h·d−1 in Dufour et al. (5) are too small for erythropoiesis to have occurred (24), it is likely that the V[Combining Dot Above]O2max improvements were because of nonhematological adaptations (9).
The frequency of IHT sessions is important, with participants in the earlier study by Terrados et al. (30) dedicating the majority of their training to the 4–5 IHT sessions per week, compared with just 2 sessions per week in Dufour et al. (5) and Roels et al. (26). The ecological validity of the design of any such IHT intervention is paramount, in that experimental treatments aimed at enhancing athletic performance should match what is likely possible to incorporate into an athlete's training schedule.
Although participants in Roels et al. (26) were “well-trained cyclists,” with a mean baseline cycling V[Combining Dot Above]O2max of 66 ml·kg−1·min−1, and participants in Dufour et al. (5) were moderately trained runners, with a mean baseline running V[Combining Dot Above]O2max of ∼63 ml·kg−1·min−1, Terrados et al. (30) are the only authors to date to have assessed the effects of IHT in highly trained athletes (international standard cyclists with a mean baseline cycling V[Combining Dot Above]O2max of 70 ml·kg−1·min−1). Moreover, IHT participants trained in a hypobaric chamber in Terrados et al. (30), and wore face masks in Dufour et al. (5), whereas CONT participants trained in normoxic laboratories, without masks, and Roels et al. (26) did not report whether participants were blinded to the environmental treatment. A lack of blinding in these may have resulted in influential placebo or nocebo effects, which must be controlled for to ascertain the efficacy of IHT as a worthwhile intervention.
The purpose of this study was to investigate cardiopulmonary physiological adaptations resulting from IHT or CONT, in highly trained endurance runners, using a single blinded research design. We hypothesized that 8 weeks of IHT would elicit greater improvements in submaximal and maximal physiological variables, and would lead to an enhanced incremental exercise limit of tolerance (T-Lim), compared with CONT.
Experimental Approach to the Problem
After a competition phase, a group of runners completed 6 weeks of routine training to ensure stability of basic fitness, and were then randomly allocated into an IHT group (n = 9) or a CONT group (n = 9), in a single blinded manner. In weeks 1 and 10, participants completed incremental exercise tests to quantify a range of submaximal and maximal physiological variables, as well as T-Lim, in a controlled laboratory environment. In between, during weeks 2–9, participants undertook their habitual training, with 2 “anaerobic threshold” runs each week replaced by 40 minutes IHT or CONT running sessions (16 × 40-minute sessions in total over the 8 weeks) (Table 1). These anaerobic threshold runs would ordinarily be 20–60 minutes, between the speed corresponding to the lactate threshold (LTspeed) and lactate turnpoint (LTPspeed) (13,28)—see “Procedures.”
Following the approval by the University of Exeter Ethics Committee, the participating athletes were instructed as to the details of the study and then provided written informed consent. Initially, there were 18 highly trained male endurance runners who volunteered for this study. Descriptive data for the 12 participants that finished the study are detailed in Table 2. They were all part of the same training group, lived in the same accommodation, and ate in the same canteen. These participants regularly competed at a national and international level in track events ranging from 1500 to 10,000 m, as well as in a variety of cross-country races.
Incremental Step Test to Volitional Exhaustion
Tests were performed at the same time of day, on the same days of the week, in normobaric normoxia (FIO2 = 20.9%) and hypoxia (FIO2 = 16.0%, equivalent to ∼2150 m), each separated by 1 day. Participants wore a chest harness for safety, because of the high speed running, and completed a discontinuous incremental test on a treadmill (ELG; Woodway, Waukesha, WI, USA) set at a gradient of 1% to compensate for the lack of air resistance (14); this was also the case for all the IHT and CONT training sessions. Each stage lasted 3 minutes followed by a 15-second rest interval while a capillary blood sample was taken from the earlobe for [BLa] determination (Biosen C_Line; EKF, Magdeburg, Germany). The first stage was at 10 km·h−1, with subsequent increments of 1.5 km·h−1 each stage. Peripheral O2 saturation (SpO2) and heart rate (HR) were quantified continuously through fingertip pulse-oximetry (BCI-Autocorr; Smiths Medical, Waukesha) and telemetry (S610i; Polar Electro, Kempele, Finland). These variables and breath-by-breath V[Combining Dot Above]O2 (Oxycon-Pro; VIASYS, Höechberg, Germany) were averaged over the final 60 seconds of each stage, when rating of perceived exertion (RPE) was also recorded using the Borg 6–20 scale (3). Running economy (submaximal V[Combining Dot Above]O2, relative to body weight) was measured at 14.5 km·h−1, which was below the LTspeed for all participants. The running speed that elicited 4.0 mM [Bla] and the LT (LTspeed) were derived using the Lactate-E Software (22). The LTspeed was defined as the final running velocity before the first sustained increase in [Bla] above baseline (13,28). In addition, for the purpose of setting the initial training treadmill speeds, the running speed that elicited the LTP (LTPspeed) was determined by 2 independent reviewers, defined as the final running speed before the observation of a sudden and sustained rise in [Bla], at approximately 2–5 mM [Bla] (13,28). Minimum SpO2 was calculated as the lowest 5 seconds rolling average, and V[Combining Dot Above]O2peak was calculated as the highest 30 seconds rolling average. Throughout all tests, the participants were verbally encouraged to perform maximally, and the treadmill was only stopped at volitional exhaustion, when the participant placed his feet either side of the treadmill belt or lifted his legs up and became suspended from the chest harness, at which time T-Lim was recorded to the nearest second. Based on the investigators experience of testing these caliber of athletes, maximal [Bla] ([BLa]max) was assessed at 2 minutes postexhaustion.
Environmental Control and Calibrations
Ambient FIO2 and humidity were controlled by an installed S3 Hypoxic and Humidity System (Sporting Edge UK Ltd., Basingstoke, United Kingdom), switched on during all training and testing, to allow complete participant blinding. This system was accurate to within ±0.1% and was checked before and during all tests (Servomex 5200; Servomex, Crowborough, United Kingdom). During the treadmill tests, participants wore a face mask for breath-by-breath V[Combining Dot Above]O2 determination using the Oxycon-Pro, which required adapted calibrations for use in hypoxia, as previously described (12). Face masks were not worn during any of the training sessions.
Intermittent Hypoxic Training and Control Sessions
Each treadmill training session lasted 40 minutes, comprising a 5-minute moderate intensity warm-up at the LTspeed, a 30-minute heavy intensity core phase at the LTPspeed, and a 5-minute moderate intensity cool down at 1.0 km·h−1 slower than the LTspeed. All speeds were set according to the pretraining test in the specific FIO2 environment. Heart Rate, RPE, and [Bla] were recorded every 10 minutes during the 30-minute core phase, to relativize the exercise intensity between the IHT and CONT groups. The aim was to achieve a steady state [Bla], with a rise of <2.0 mM from 10 to 30 minutes. Treadmill speed was adjusted within the session after any 10-minute period using the following criteria: if RPE <13 or there was a [Bla] decrease, speed was increased by 0.5 km·h−1. If RPE >17 or there was an increase in [Bla] >2.0 mM, between 10 and 30 minutes, speed was decreased by 0.5 km·h−1. Participants were obstructed from viewing their exercising HR, [Bla], and treadmill speeds to minimize the chances of them guessing the training or testing FIO2. When asked, after the final supervised exercise session, 46% of participants were correct, and 54% were incorrect in their guess of which environmental FIO2 they had exercised in (50% of the IHT group and 43% of the CONT group were correct in their judgment, respectively), so the blinding was judged to have been successful.
The running speeds during the final 10 minutes of each of the supervised laboratory sessions were recorded and used to assess any differences in the progression between the IHT and CONT groups over the 8-week intervention. In addition, all participants wore a GPS equipped watch (Forerunner 405; Garmin Ltd., Southampton, United Kingdom) during all runs completed outside the laboratory to quantify running distances. Data were analyzed to investigate any differences in total running distances between groups. Table 1 illustrates a typical training week.
Resting Hematology and Nutrition
Venous blood samples were drawn from an antecubital vein through phlebotomy during weeks 1, 5, and 10, to screen for illness and to ensure adequate iron stores (serum ferritin concentration ([sFe]) ≥30 μg·L−1). Within 2 hours, samples were analyzed for full blood count (FBC) using an automated cell counter (ADVIA 120; Siemens AG, Erlangen, Germany) and [sFe] using a chemiluminescent microparticle immunoassay (ARCHITECT Ferritin Assay; Abbott Point of Care Inc., Birmingham, United Kingdom). Furthermore, to stress the importance of adequate dietary iron and other nutrient intake, all participants attended a 45-minute nutrition seminar by an experienced sports dietician during week 1.
Analyses were performed using PASW Statistics (v18.0; IBM SPSS, Portsmouth, United Kingdom), with the probability level of p ≤ 0.05 being accepted as statistically significant. Data are reported as mean ± SD. To assess changes pretraining to posttraining, absolute changes in each dependent variable were compared between IHT and CONT groups using repeated measures analyses of covariance with mixed measures. Pretraining values for all exercise test variables were entered as covariates, to control for any baseline differences between the IHT and CONT groups.
Of the 18 participants who started the study, 4 dropped out due to illness (IHT group n = 3, CONT group n = 1), and 2 qualified for international competitions so chose to cease their involvement (IHT group n = 1, CONT group n = 1). Baseline descriptive data is detailed in Table 2 for the 12 “finishers” who completed all 16 supervised treadmill sessions within the planned 8-week period. The mean running speed during all 16 of the laboratory training sessions was significantly slower in the IHT group compared with the CONT group (15.6 ± 0.8 vs. 17.1 ± 0.8 km·h−1, p = 0.007). There was a significant increase in the mean running speed during week 1 to week 8 when results from both groups were combined (16.1 ± 1.1 to 16.9 ± 1.0, km·h−1, p = 0.033), but there was no difference in this running speed progression between the IHT and the CONT groups (IHT 15.2 ± 0.7 to 16.0 ± 0.7 vs. CONT 16.8 ± 0.9 to 17.5 ± 0.7 km·h−1, p = 0.565). Total weekly running distances were also not different between groups (IHT 85.1 ± 5.1 km vs. CONT 84.6 ± 5.6 km; p = 0.776).
The changes in submaximal V[Combining Dot Above]O2 were significantly different after IHT compared with CONT, in both normoxia (p = 0.003) and hypoxia (p = 0.010) (Figure 1). During normoxic tests, submaximal V[Combining Dot Above]O2 increased in 5 of the 7 CONT participants (54.1 ± 2.9 to 55.2 ± 1.5 ml·kg−1·min−1, p = 0.012) and showed a tendency to decrease post-IHT, with submaximal V[Combining Dot Above]O2 decreasing in 4 out of the 5 IHT participants (52.0 ± 3.6 to 50.5 ± 2.7 ml·kg−1·min−1, p = 0.052). In hypoxia, submaximal V[Combining Dot Above]O2 did not change post-CONT (49.0 ± 1.7 to 50.3 ± 1.7 ml·kg−1·min−1, p = 0.296), but decreased in all 5 of the IHT participants (48.6 ± 3.5 to 46.0 ± 1.8 ml·kg−1·min−1, p = 0.001). Submaximal HR when tested in normoxia decreased significantly after both IHT (154 ± 13 to 148 ± 8 b·min−1; p = 0.001) and CONT (160 ± 10 to 159 ± 7 b·min−1; p = 0.021), with this difference being statistically greater after IHT (p = 0.001) (Figure 1). There were no significant submaximal HR changes within or between groups in hypoxia, and there were no significant changes within groups or differences between groups in the speed at 4.0 mM [BLa] or in submaximal SpO2 (Figure 1).
T-Lim remained unaltered after both CONT and IHT in normoxic conditions. During hypoxic tests, there was a significant T-Lim improvement after IHT (25.0 ± 1.9 to 25.6 ± 1.1 minutes; p = 0.031), but not after CONT (23.7 ± 1.7 to 23.8 ± 1.9 minutes; p = 0.836). However, T-Lim changes were not statistically different between the IHT and CONT groups when tested in either normoxia (p = 0.463) or hypoxia (p = 0.214) (Figure 2).
While V[Combining Dot Above]O2peak in normoxia increased to a greater extent after CONT than after IHT, HRmax in hypoxia decreased to a greater extent after IHT than after CONT, and SpO2 min in normoxia increased to a greater extent after IHT than after CONT (Figure 3), such changes were not significant within either group, in either normoxic or hypoxic test conditions. Similarly, while in hypoxic conditions, there was a significant decrease in [Bla]max post-IHT (13.0 ± 4.1 to 11.4 ± 2.1 mM; p = 0.007), but not post-CONT (11.0 ± 1.9 to 10.5 ± 2.3 mM; p = 0.372), and there were no significant between group differences (Figure 3).
[sFe] remained above 30 μg·L−1 in all participants who completed the study, and the medical team had no concerns regarding the FBC red or white blood cell differential variables, which did not significantly differ from baseline, and did not differ between groups.
We investigated whether 8 weeks of IHT would elicit improvements in submaximal and maximal physiological variables and incremental running T-Lim compared with CONT. Our main findings were that (a) submaximal HR reduced more after IHT, compared with CONT; (b) submaximal V[Combining Dot Above]O2 tended to increase after CONT and to decrease after IHT; and (c) although T-Lim after IHT improved in hypoxic test conditions, there were no T-Lim changes within either group in normoxic test conditions.
The greater submaximal HR reductions that we observed post-IHT (mean −4%) compared with post-CONT (mean −1%) indicate a greater cardiovascular fitness gain post-IHT. In a similar manner, Vallier et al. (31) reported that submaximal HR was on average 4% lower after highly trained triathletes performed 3 weeks of IHT. Although consistent with our results, these authors used a more extreme altitude simulation (∼4000 m), hypobaric rather than normobaric hypoxia, and did not include a CONT group, so it is impossible to judge to what extent the changes were because of the hypoxia, per se, or to enhanced cardiovascular fitness after exercise training. Moreover, Terrados et al. (29) found that constant submaximal load HR was reduced significantly more after highly trained cyclists undertook 3–4 weeks of IHT (at an altitude simulation of ∼2300 m) compared with CONT (mean −16 vs. −12%, respectively).
In this study, 5 out of the 7 CONT participants showed submaximal V[Combining Dot Above]O2 increases, which reached statistical significance in normoxia. On the contrary, of the 5 IHT participants, submaximal V[Combining Dot Above]O2 decreased in 4 of them in normoxia, and all 5 of them in hypoxia, which reached statistical significance in hypoxia. These results in conjunction with the decreased submaximal HR indicate a lower energetic cost of exercise post-IHT and an increased energetic cost of exercise post-CONT, the latter being rather surprising, as an increase in submaximal V[Combining Dot Above]O2 would not be expected after 8 weeks of training.
However, in the similarly designed study by Roels et al. (26), these authors also reported a trend for submaximal V[Combining Dot Above]O2 to decrease after IHT (n = 11, −3%, tested in hypoxia) and increase after CONT (n = 11, +2%, tested in normoxia), although these differences were not statistically significant. Furthermore, Robertson et al. (25) found that submaximal V[Combining Dot Above]O2 remained unchanged in moderately trained middle distance runners who performed 3 weeks of either IHT, or IHT while living in normobaric hypoxia (3000 m simulated altitude). This study also had a number of withdrawals, meaning that there were only 12 “finishers,” thus in a similar manner to this study, the statistical power was reduced. Finally, although the aforementioned study by Dufour et al. (5) showed impressive gains in maximal physiological variables and T-Lim after IHT (n = 9) compared with CONT (n = 9), these authors also reported no significant submaximal V[Combining Dot Above]O2 changes in either group. Given the larger sample sizes, the power of statistical analyses in Roels et al. (26) and Dufour et al. (5) would have surpassed that of this study. Together with results from Robertson et al. (25), these investigations question whether the submaximal V[Combining Dot Above]O2 changes that we observed were physiologically meaningful. Nevertheless, the attenuated submaximal HR suggests that IHT may be a useful means of improving submaximal (moderate to heavy) exercise capacity, and this area warrants additional research.
It is reasonable to suggest that any reduction in the energetic cost of exercise post-IHT would likely lead to an improved T-Lim. Indeed there was a significant T-Lim increase post-IHT, when participants were tested in hypoxia. Although the measure of T-Lim in this study can only provide a surrogate estimate of athletic performance, and changes were not statistically different between groups, the within IHT group change shows a clear trend (Figure 2). Based on this finding, and that from another of our recent IHT investigations, whereby muscle oxidative capacity improved in active men when tested in hypoxic conditions (11), we suggest that IHT elicits adaptations within skeletal muscle that result in a degree of acclimatization to subsequent hypoxia. Although an (resting) intermittent normobaric hypoxic exposure intervention has been reported not to enhance subsequent exercise tolerance at moderate altitude (7), as far as we are aware this has not been trialled using a similarly designed IHT intervention to the present study, in highly trained athletes.
Results from the similar study by Dufour et al. (5) differ to this study, as after 6 weeks' training, T-Lim in normoxia improved in all 9 IHT participants (mean + 35%; approximate range, +5 to +63%) compared with only 5 out of 9 CONT participants (mean + 10%; approximate range, −18 to +51%). Some variability is to be expected in duration-blinded time to exhaustion tests, perhaps ∼15% (16), but the changes reported by Dufour et al. are considerably greater, so it is suggested that the lack of FIO2 blinding and the resulting placebo or nocebo effects, were partially responsible. These researchers found no in vitro maximal oxidative capacity improvements after IHT compared with CONT, and concluded either that the participants had reached a mitochondrial adaptation plateau, or that IHT does not improve mitochondrial content (23).
In contrast, studies in untrained participants by Vogt et al. (33) and Geiser et al. (8) reported significantly greater increases in mitochondrial and capillary densities after IHT compared with CONT. Similarly, Terrados et al. (29) and Melissa et al. (20) reported significant citrate synthase activity increases in the hypoxic trained legs compared with normoxic trained legs of untrained participants (using the same absolute work rates), thus indicating enhanced mitochondrial function post-IHT. Results from these studies being largely different to results in moderately trained participants (23) indicate that baseline training status likely has an important impact on adaptive outcomes. In this study, the participants' pre-intervention V[Combining Dot Above]O2peak of 70.0 ± 3.5 ml·kg−1·min−1 was already high, so the scope for further maximal aerobic improvements through an 8-week intervention was likely to be limited.
Nevertheless, in another investigation using single-legged exercise, Bakkman et al. (1) had 8 untrained participants to undertake IHT and CONT, at 65% of the maximal attained work rate in the specific FIO2 that each leg was to be trained in. Maximal power output improved similarly in both legs, regardless of training condition, and in contrast to Terrados et al. (29) and Melissa et al. (20), citrate synthase activity increased significantly more in the CONT leg than in the IHT leg, in which there was no such change. These authors suggested that the lower absolute work rate in the IHT leg likely caused a reduced stimulus for mitochondrial biogenesis (1). Furthermore, Heinonen et al. (10) demonstrated that during submaximal exercise in normobaric hypoxia, elevated cardiac output provides adequately increased muscle blood flow to counteract reductions in arterial O2 content, thus total muscle O2 delivery remains largely unaltered. It is only during high intensity whole body exercise in hypoxia that muscle O2 delivery is significantly reduced (4).
On this basis, a recent study by Faiss et al. (6) found that repeated sprint cycling performance improved significantly more so after IHT compared with CONT, when equal absolute (maximal effort) workloads were used. As such, the IHT stimulus in this study, whereby the relative training intensity was essentially “clamped,” meaning that IHT participants underwent lower absolute workloads than CONT participants, may simply not have been sufficiently intense to stress O2 delivery. So the lack of observed T-Lim gains in this study, and similarly by others that used moderate to heavy intensity training (17,25,26,30), is perhaps not all that surprising. In the case of the moderate intensity, IHT study by Dufour et al. (5), again, we suggest that the performance gains were at least in part because of placebo and nocebo effects. Moreover, these authors did not report any between group statistical analyses, meaning that the effects within the IHT and CONT group were not shown to be different.
In agreement with this study, Ventura et al. (32) also reported no significant performance improvements after 6 weeks' IHT in moderately trained cyclists. However, their IHT sessions were in addition to participants' habitual training, and these authors suggested that the lack of performance change could have been a result of over-training (32). Reductions in submaximal HR, peak [Bla], and maximal exercise capacity are commonly observed during over-training (19), and it is possible that a similar effect occurred in this study. Our experimental design meant that training loads were maintained consistently over the 8 weeks, so participants were not able to take additional rest when feeling fatigued, as they may have done under normal circumstances. It is therefore likely that some participants experienced accumulated fatigue by the end of the intervention, which was confirmed anecdotally by the athletes and their coach, and is evidenced by the lack of T-Lim changes, the high proportion of withdrawals due to illness, and by trends toward a post-IHT reduction in normoxic V[Combining Dot Above]O2peak and HRmax.
It should be acknowledged that the discontinuous combined submaximal and maximal intensity exercise test may not have provided the most sensitive means of assessing maximal exercise capacity. Instead of using T-Lim as the sole surrogate estimate of athletic performance, future similar studies should include a range of exercise capacity tests, including extended submaximal protocols to establish whether any changes in submaximal HR or V[Combining Dot Above]O2 lead to greater aerobic exercise tolerance, and also using more ecologically valid assessments of maximal performance capacity.
In conclusion, 8 weeks of IHT resulted in significantly lower submaximal HR and a tendency for a reduced oxygen cost of submaximal exercise. Although incremental running time to exhaustion improved in hypoxic conditions, changes in normoxic test conditions were not apparent.
- The results of this study do not support the use of moderate intensity IHT to enhance athletic performance at sea level. One potential use for such an IHT intervention is during pre-acclimatization for subsequent training in hypoxia, although this concept requires more direct assessment, with performance being quantified using more ecologically valid methods (e.g., competitive races).
- It is recommended that those professionals who supervise IHT interventions closely monitor individual athlete fatigue and well-being, and that research is carried out into the fatigue consequent to IHT in more depth, especially if equal absolute workloads are used for IHT as they would be for normoxic training.
- Recent literature suggests that IHT using severe or supramaximal intensity exercise may provide an additional benefit on athletic performance than normoxic training, at least for repeated sprint type activities, but these claims must be further substantiated. In particular, future research should comprise rigorously designed double-blinded assessments of the effects of whole body severe or supramaximal intensity IHT in highly trained athletes, as there is a scarcity of data in this population.
The academic tuition fees related to this research were in part funded by British Swimming (www.swimming.org), the English Institute of Sport (www.eis2win.co.uk), and Sporting Edge UK Ltd (www.sportingedgeuk.co.uk).
1. 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) 190: 243–251, 2007.
2. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol (1985) 60: 2020–2027, 1986.
3. Borg G. The borg RPE scale. In: Borg's Perceived Exertion and Pain Scales. Borg G., ed. Champaign, IL: Human Kinetics, 1998. pp. 29–38.
4. Calbet JA, Radegran G, Boushel R, Saltin B. On the mechanisms that limit oxygen uptake during exercise in acute and chronic hypoxia: Role of muscle mass. J Physiol 587: 477–490, 2009.
5. Dufour SP, Ponsot E, Zoll J, Doutreleau S, Lonsdorfer-Wolf E, Geny B, Lampert E, Fluck M, Hoppeler H, Billat V, Mettauer B, Richard R, Lonsdorfer J. Exercise training in normobaric hypoxia in endurance runners. I. Improvement in aerobic performance capacity. J Appl Physiol (1985) 100: 1238–1248, 2006.
6. Faiss R, Leger B, Vesin JM, Fournier PE, Eggel Y, Deriaz O, Millet GP. Significant molecular and systemic adaptations after repeated sprint training in hypoxia. PLoS One 8: e56522, 2013.
7. Faulhaber M, Gatterer H, Haider T, Patterson C, Burtscher M. Intermittent hypoxia does not affect endurance performance at moderate altitude in well-trained athletes. J Sports Sci 28: 513–519, 2010.
8. 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 22: 579–585, 2001.
9. Gore CJ, Hopkins WG. Counterpoint: Positive effects of intermittent hypoxia (live high:train low) on exercise performance are not mediated primarily by augmented red cell volume. J Appl Physiol (1985) 99: 2055–2057, 2005.
10. Heinonen IH, Kemppainen J, Kaskinoro K, Peltonen JE, Borra R, Lindroos M, Oikonen V, Nuutila P, Knuuti J, Boushel R, Kalliokoski KK. Regulation of human skeletal muscle perfusion and its heterogeneity during exercise in moderate hypoxia. Am J Physiol Regul Integr Comp Physiol 299: R72–R79, 2010.
11. Holliss BA, Fulford J, Vanhatalo A, Pedlar CR, Jones AM. Influence of intermittent hypoxic training on muscle energetics and exercise tolerance. J Appl Physiol (1985) 114: 611–619, 2013.
12. Holliss BA, Pedlar C, Glaister M, Jones A. Validity of the Jäeger Oxycon-Pro®
expired air analyzer in normobaric hypoxia. Presented at 16th Annual Congress of the European College of Sport Science, Liverpool, United Kingdom, 6–9 July, 2011.
13. Jones AM. The physiology of the world record holder for the women's marathon. Int J Sports Sci Coach 1: 101–115, 2006.
14. Jones AM, Doust JH. A 1% treadmill grade most accurately reflects the energetic cost of outdoor running. J Sports Sci 14: 321–327, 1996.
15. Kallio PJ, Wilson WJ, O'Brien S, Makino Y, Poellinger L. Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway. J Biol Chem 274: 6519–6525, 1999.
16. Laursen PB, Francis GT, Abbiss CR, Newton MJ, Nosaka K. Reliability of time-to-exhaustion versus time-trial running tests in runners. Med Sci Sports Exerc 39: 1374–1379, 2007.
17. Lecoultre V, Boss A, Tappy L, Borrani F, Tran C, Schneiter P, Schutz Y. 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 95: 315–330, 2010.
18. Levine BD, Stray-Gundersen J. Point: Positive effects of intermittent hypoxia (live high:train low) on exercise performance are mediated primarily by augmented red cell volume. J Appl Physiol (1985) 99: 2053–2055, 2005.
19. Meeusen R, Nederhof E, Buyse L, Roelands B, de Schutter G, Piacentini MF. Diagnosing overtraining in athletes using the two-bout exercise protocol. Br J Sports Med 44: 642–648, 2010.
20. 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 29: 238–243, 1997.
21. Millet GP, Roels B, Schmitt L, Woorons X, Richalet JP. Combining hypoxic methods for peak performance. Sports Med 40: 1–25, 2010.
22. Newell J, Higgins D, Madden N, Cruickshank J, Einbeck J, McMillan K, McDonald R. Software for calculating blood lactate endurance markers. J Sports Sci 25: 1403–1409, 2007.
23. Ponsot E, Dufour SP, Zoll J, Doutrelau S, N'Guessan B, Geny B, Hoppeler H, Lampert E, Mettauer B, Ventura-Clapier R, Richard R. Exercise training in normobaric hypoxia in endurance runners. II. Improvement of mitochondrial properties in skeletal muscle. J Appl Physiol (1985) 100: 1249–1257, 2006.
24. Rasmussen P, Siebenmann C, Diaz V, Lundby C. Red cell volume expansion at altitude: A meta-analysis and Monte Carlo simulation. Med Sci Sports Exerc 45: 1767–1772, 2013.
25. 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 110: 379–387, 2010.
26. Roels B, Millet GP, Marcoux CJ, Coste O, Bentley DJ, Candau RB. Effects of hypoxic interval training on cycling performance. Med Sci Sports Exerc 37: 138–146, 2005.
27. Semenza GL. Regulation of mammalian O2 homeostasis by hypoxia-inducible factor 1. Annu Rev Cell Dev Biol 15: 551–578, 1999.
28. Smith CG, Jones AM. The relationship between critical velocity, maximal lactate steady-state velocity and lactate turnpoint velocity in runners. Eur J Appl Physiol 85: 19–26, 2001.
29. Terrados N, Jansson E, Sylven C, Kaijser L. Is hypoxia a stimulus for synthesis of oxidative enzymes and myoglobin? J Appl Physiol (1985) 68: 2369–2372, 1990.
30. Terrados N, Melichna J, Sylven 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 57: 203–209, 1988.
31. Vallier JM, Chateau P, Guezennec CY. Effects of physical training in a hypobaric chamber on the physical performance of competitive triathletes. Eur J Appl Physiol Occup Physiol 73: 471–478, 1996.
32. 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 24: 166–172, 2003.
33. 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 (1985) 91: 173–182, 2001.