Journal Logo


Reproducibility of Performance Changes to Simulated Live High/Train Low Altitude


Author Information
Medicine & Science in Sports & Exercise: February 2010 - Volume 42 - Issue 2 - p 394-401
doi: 10.1249/MSS.0b013e3181b34b57
  • Free


Many athletes use altitude training to induce physiological adaptations associated with improved performance (22). Despite more than 40 yr of research and widespread support among athletes and coaches, there is little consensus on whether altitude training can enhance sea level performance in highly trained athletes. Research findings are confounded by differing methodologies including varying duration, length, and type of altitude training and training level and status of the athletes. Nonetheless, there is evidence that natural or simulated live high/train low (LHTL) altitude may offer small performance benefits for some athletes (14,24,34), although the underlying physiological mechanism(s) remain unclear (22).

In our experience, elite endurance athletes often undertake multiple altitude exposures within and between training years to gain a competitive edge (28). However, it has not been established if an individual athlete responds in the same way to repeated bouts of altitude exposure. Small gains in performance after both natural (21,34,36) and simulated (4) LHTL have been attributed to hypoxia-induced increases in hemoglobin mass (Hbmass) and improved maximal aerobic power (V˙O2max). However, several studies have observed small ∼1% improvements in performance after simulated LHTL, with no substantial change in hematological parameters (14,24,31). With shorter daily exposures (8-12 h·d−1), other peripherally mediated mechanisms including improved running economy (31) and enhanced muscle buffering capacity (13) have been observed in the absence of hematological changes. These parameters warrant further investigation to clarify whether small changes in performance after altitude training can be explained independently of increased number of red blood cells (22).

Replication is a fundamental tenet of science, and yet no previous study has addressed the reproducibility of physiological adaptations and performance gains to repeated altitude exposure in the same athletes. Therefore, the aim of this study was to quantify the test-retest reproducibility of both physiological and performance changes in highly trained athletes after two matched 3-wk blocks of simulated LHTL.



Sixteen highly trained male and female middle-distance and distance runners (age = 30.6 ± 4.6 yr, mean ± SD) undertook 2 × 3-wk blocks of simulated LHTL altitude (n = 8) or resided near sea level (control, n = 8). Athletes were pair-matched for time trial performance and randomly assigned to either the LHTL or the control group. Both groups trained together near sea level (Canberra, ∼600-m ambient altitude) and were tested under normoxic conditions at the Australian Institute of Sport. All participants had been training and running competitively for a minimum of 5 yr (LHTL = 11 ± 4 yr, control = 14 ± 3 yr). Baseline characteristics are presented in Table 1. Written consent was obtained from subjects after they were informed of the experimental procedures and possible risks involved with participation. The study was approved by both the Ethics Committee of the Australian Institute of Sport and the Committee for Ethics in Human Research at the University of Canberra.

Baseline characteristics of the elite runners in the simulated LHTL and control groups at the start of each block (mean ± SD).

Experimental design.

The LHTL group spent 2 × 3 wk (14 h·d−1, ∼300 h) in a normobaric hypoxic five-bedroom facility (3000 m), separated by a 5-wk washout period between blocks 1 and 2 (Fig. 1).

Schematic diagram of testing during each 3-wk simulated LHTL or near sea level training block (blocks 1 and 2). Test, 4.5-km time trial, and treadmill test performed on separate days. Blood sample, venous blood collection.

Time trial.

One day before and after each 3-wk block, a 4.5-km time trial on a road course (six laps) was completed in the shortest possible time, with split times and final time recorded manually on a stopwatch (S120-4020; Seiko, Tokyo, Japan). This atypical distance was selected to prevent athletes following a predetermined pacing strategy over familiar race distances such as 3 or 5 km. Consistent with the approach of others (21,29), athletes were instructed to follow an experienced pacer for laps 1 and 2 (first 1.5 km). A paced start reduces the chance of individuals going out too fast or too slow but nevertheless allows ample opportunity for the athlete to change their pacing during the final four laps. Individualized starting pace was based on predicted finish time from recent 3- or 5-km race times. Runners were given feedback on distance completed but no temporal information on split times during the time trial. The typical error of measurement (TE) established from the control group from pretest to posttest was 1.3% (90% confidence interval = 1.0%-2.1%).

Treadmill test.

A treadmill test was used to determine V˙O2max, running economy, lactate threshold, and velocity at V˙O2max (vV˙O2max) on a custom-built motorized treadmill. The test comprised continuous incremental running for 4-5 × 4-min periods, at 0% gradient, with speeds ranging from 14 to 20 km·h−1, with a 1-min rest between each stage. Five minutes after the final submaximal stage, a maximal 60-s run (18.5-22 km·h−1) at 4% gradient was completed. After a further 5-min break, subjects recommenced running at 14-16 km·h−1, with speed increasing by 0.5 km·h−1 every 30 s for 5 min. Beyond this point, speed was maintained and gradient was increased by 0.5% every 30 s until volitional exhaustion. HR was recorded continuously (Heart Rate Monitor; Polar, Kempele, Finland), and capillary blood (5 μL) was sampled from the fingertip after each stage and 1 min after completion of the test for measurement of blood lactate concentration (Lactate Pro; ARKRAY, Kyoto, Japan). Expired ventilation samples were collected by a custom-built open-circuit indirect calorimetry system for the determination of oxygen consumption (31). The TE established from duplicate trials on a separate group of endurance-trained athletes was 2.1% (1.4%-4.1%) for V˙O2max and 2.4% (1.7%-4.6%) for submaximal V˙O2.

Running economy was derived from steady-state oxygen uptake during the last 60 s of each submaximal stage of the treadmill test to determine V˙O2 for a given velocity. Comparisons were made for economy from pooled individual data from four submaximal running speeds. Capillary blood (95 μL) was sampled from the fingertip before and 3 min after the 60-s maximal run and was analyzed on the ABL-700 series analyzer (Radiometer Medical, Copenhagen, Denmark). Blood pH, bicarbonate (HCO3), and lactate concentration were measured as an indirect means of assessing changes in buffering capacity, with less reduction in HCO3 indicative of augmented buffering capacity.


Hbmass was measured twice before, each week during, and 1 wk after each 3-wk block using a modified 2-min carbon monoxide (CO) rebreathing test (25). This method accounts for loss of CO due to exhalation and diffusion to the vascular bed. A small bolus (∼1 mL·kg−1) of CO was rebreathed with 3-3.5 L of oxygen for 2 min. Capillary blood (200 μL) was sampled from a prewarmed fingertip before and 6 and 8 min after CO inhalation to measure percent carboxyhemoglobin in quadruplicate (OSM-3 Hemoximeter; Radiometer Medical). Change in percent carboxyhemoglobin (baseline to average of 6 and 8 min) was used to calculate Hbmass (g). The TE for double-baseline Hbmass before block 1 was 2.0% (1.6%-2.6%) in this group.

Hematological parameters.

Each subject reported to the laboratory twice before blocks 1 and 2 in an overnight fasted state, and a venous blood sample was collected after 10 min of supine rest. Venous blood was collected into a 2-mL EDTA tube for red blood cell analysis and a 4-mL serum separation tube for biochemical analysis (Vacuette; Greiner Bio-One, Frickenhausan, Germany) at six time points (pre-1, pre-2, day 2, day 6, day 20, and day 27) in blocks 1 and 2 (Fig. 1).

Hemoglobin concentration, hematocrit levels, and percentage of reticulocytes were determined using the Advia 120 Hematology Analyzer (Bayer Diagnostics, Tarrytown, NY). The analyzer was calibrated against appropriate reference materials and checked daily using 3-in-1 TESTpoint quality controls. Whole blood was centrifuged at 4000 rpm for 5 min, and the serum was aliquoted for storage at −80°C until batch analysis. The concentrations of ferritin and soluble transferrin receptor (sTfR) were determined by immunoturbidimetric assay on a Hitachi 911 Automatic Analyzer (Boehringer, Mannheim, Germany). Erythropoietin (EPO) concentration was measured on an Immulite (Siemens AG, Deerfield, IL) using an enzyme-amplified chemiluminescence assay with automated bead washing (Immulite 1000 Test Unit).

All subjects were provided with daily oral iron supplement (Ferrograd C; 325 mg of dried ferrous sulfate and 562.4 mg of sodium ascorbate; Ferrogradumet; Abbott Australia, Botany, Australia) to take 1 wk prior and for the duration of each 3-wk block to ensure adequate iron stores for accelerated erythropoiesis.

Training quantification.

Subjects were instructed to maintain their normal training program throughout the study. Training data were collected from each subject during the study from a daily training log, global positioning system (GPS) monitoring (Forerunner, Garmin, KS), and HR monitoring (S-Series; Polar). Data were collated and compared for training distance (km), duration (h), and intensity (%HRmax) within and between groups. Total training distance, duration, and intensity did not differ substantially for either group from block 1 to block 2, and there was no substantial difference in training duration or training intensity between the groups. However, the LHTL group had a greater mean total training distance compared with the control group in blocks 1 (370 ± 92 vs 269 ± 62 km) and 2 (379 ± 91 vs 258 ± 60 km). Given the difference in training distance between the groups, we incorporated individual training distance as a covariate in the analysis of time trial performance (16).

Statistical analysis.

We used a contemporary statistical approach (20) because changes in performance of as little as ∼0.5% are meaningful for elite athletes (18) and because conventional statistics can be relatively insensitive to such small but important changes. Specifically, we used magnitude-based inferences about effect sizes and precision of estimation expressed as 90% confidence limits (CL) to evaluate differences within and between groups (3). The unequal-variances t statistic was used to analyze differences in the mean change for each group during blocks 1 and 2 and to compare differences in the change scores between the LHTL and control groups. All measures were log-transformed for the analyses to reduce bias arising from any nonuniformity of error and were back-transformed to obtain changes in means and SD as percents (20). The probability that the true value of the effect was practically positive, trivial, or negative accounted for the observed difference, the smallest worthwhile difference, and typical error of measurement (3). A reference value of 1% for the smallest worthwhile change in performance was calculated from 0.5 × within-athlete coefficient of variation from competition of similar distance events in elite runners (19). For measures not directly related to performance, the smallest worthwhile change was calculated from Cohen's smallest standardized effect size of 0.2 multiplied by the pretest between-athlete SD (8).

Effects that were simultaneously both >75% likely positive and <5% negative were considered substantial and beneficial. Those effects where the confidence interval overlapped the thresholds for both positive and negative were deemed unclear (17). The true individual response was used to indicate the degree of variability in response to the intervention relative to the magnitude of the mean change. The true individual response expressed as a percent was calculated as the square root of the difference in the variance in the change scores between the LHTL and control groups (15). Pearson product-moment correlations were interpreted using a scale of magnitudes (8) comprising of 0.1 (small), 0.3 (moderate), and 0.5 (large). TE for outcome measures was calculated from the SD of the change scores divided by √2 and presented as a percent.


Reproducibility of physiological and performance changes within groups.

After each 3-wk LHTL exposure, there were reproducible mean increases in V˙O2max (mean [±90% CL]: 2.1% [±2.1%] and 2.1% [±3.9%]) and Hbmass (2.8% [±2.1%] and 2.7% [±1.8%]). However, changes in the 4.5-km time trial performance were less consistent with the LHTL group substantially faster after block 1 (−1.4% [±1.1%]) but trivially slower after block 2 (0.7% [±1.3%]). The control group had only trivial changes after blocks 1 and 2 in V˙O2max (0.9% [±2.8%] and 0.7% [±3.1%]), Hbmass (1.4% [±2.7%] and −1.5% [±1.5%]) and time trial performance (0.5% [±1.5%] and −0.7% [± 0.8%]).

Overall, there was a lack of association between percent change in block 1 versus percent change in block 2 in physiological and performance measures for the LHTL group. There were moderate but unclear correlations for V˙O2max and Hbmass and only a trivial correlation for time trial performance (Fig. 2). True individual responses indicating the variability in response to the intervention were of similar magnitude to the small mean changes in blocks 1 and 2 for V˙O2max (2.6% and 3.7%), Hbmass (2.4% and 1.4%), and time trial performance (1.5% and 1.6%).

Percent change (Δ) in block 1 versus block 2 for the 4.5-km time trial performance (TT), maximal aerobic power (V˙O2max), and Hbmass for the LHTL (n = 8; filled circles) and control (n = 8; open circles) groups. The regressions are the line of best fit and 90% confidence interval (dashed line) for each group.

Time trial performance.

Compared with the control group, the LHTL group was substantially faster (−1.9% [±1.8%]) after block 1 but possibly slower (1.4% [±1.5%]) after block 2 (Fig. 3). When adjusted for training volume, the difference between the groups from pretest to posttest was unclear after block 1 (−1.1% [±2.4%]), and the LHTL group was substantially slower after block 2 (2.0% [±1.9%]). In terms of pacing, the LHTL group were faster in the final four laps in the posttest after block 1 (lap 3 = −1.5% [±1.4%], lap 4 = −1.6% [±1.5%], lap 5 = −2.1% [±1.7%], lap 6 = −2.9% [±2.5%]), but there were only trivial differences in lap times after block 2 and after both blocks for the control group.

ndividual percent change in the time trial performance. Group mean ± SD are shown inheavy black and offset slightly for clarity for LHTL (n = 8; filled circles) and control (n = 8; open circles) groups. Negative values indicate a decrease in time and improved performance. Gray shaded area indicates the range of trivial changes from baseline.

Treadmill test.

Mean change in V˙O2max for both groups was reproducible; however, other parameters from the treadmill test were more variable (Table 2). The LHTL group had no substantial improvement in running economy or vV˙O2max after either block; 4 mM lactate running speed was substantially faster only after block 1, and posttest lactate concentration was substantially lower after both blocks 1 and 2. The control group demonstrated trivial changes after both 3-wk blocks, with posttest lactate substantially higher only after block 1. Compared with the control group, the LHTL group was substantially faster at 4 mM lactate running speed after block 1 and had lower lactate posttest after blocks 1 and 2, but there were no substantial differences in the change in mean V˙O2max, vV˙O2max, or economy. After the maximal 60-s test, the difference in blood HCO3 (pre − post) was substantially lower after block 1 (−3.5% [±5.0%]) in the LHTL group, but there was little difference after block 2 or when compared with the control group.

Physiological parameters for treadmill tests 1 and 2 (block 1) and tests 3 and 4 (block 2) for the simulated LHTL and control groups.


The LHTL group exhibited a reproducible pattern of weekly increases in Hbmass in both blocks 1 and 2; in contrast, the control group showed relatively little change (Fig. 4). Hbmass measured at week 3 was substantially higher in the LHTL group compared with the control group after block 2 (4.2% [±2.1%]), but there was only a trivial difference between the groups after block 1 (1.3% [±3.2%]).

ndividual percent change in Hbmass measured before, weekly, and 1 wk after blocks 1 and 2. Pretest values for each block are the mean of the two pretests. Group mean ± SD are shown in heavy black and offset slightly for clarity for LHTL (n = 8; filled circles) and control (n = 8; open circles) groups. Gray shaded area indicates the range of trivial changes from baseline.

Hematological parameters.

There was evidence of accelerated erythropoiesis in the LHTL group with substantially higher EPO concentration at days 2 and 6 of each exposure, increased sTfR concentration at day 6 that remained elevated until day 20, and a corresponding decrease in ferritin concentration during each 3-wk LHTL exposure (Fig. 5). The control group had no substantial differences in sTfR or ferritin concentration in either block, although EPO concentration was substantially lower at days 20 and 27 in block 1. Percent reticulocytes were substantially increased in both blocks 1 and 2 for the LHTL group at days 6 and 20 and for the control group at days 6, 20, and 27.

Mean change in serum EPO, sTfR, and ferritin concentration measured before and on days 2, 6, 20, and 27 in blocks 1 and 2 in the LHTL (n = 8; filled circles) and control (n = 8; open circles) groups. Pretest values for each block are the mean of the two pretests. Data are raw mean ± SD. *Substantially different from pretest.

Compared with the control group, EPO was substantially elevated in blocks 1 and 2 at day 2 (70% [±30%] and 90% [±24%]), day 6 (37% [±29%] and 34% [±28%]), and day 20 (16% [±21%] and 32% [±28%]), and sTfR was higher at day 20 (19% [±13%] and 11% [±11%]) in the LHTL group. There were no substantial differences in ferritin between the groups.


This is the first study to demonstrate that 3-wk simulated LHTL exposure can elicit reproducible mean increases in V˙O2max (∼2%) and Hbmass (∼3%). However, these physiological enhancements did not transfer to reproducible improvements in the 4.5-km time trial running performance. There was an apparent uncoupling in the relationship between underlying improvements in physiological capacities and changes in endurance performance after altitude training in highly trained athletes. The mean changes in performance measures with hypoxic exposure were small and variable. Therefore, our results suggest that the timing of competition after altitude exposure and the management of training may be particularly important factors in realizing performance gains after altitude exposure.


After the first 3-wk block, the ∼1% performance improvement in the LHTL group compares favorably with the previously reported 1.1%-1.6% improvements in 3-km (34) and 5-km time trial performances (21,36) after LHTL. These improvements in performance corresponded with ∼3%-4% increases in V˙O2max and were attributed to increased Hbmass or red cell volume (21,34,36). However, a slower time trial performance after the second 3-wk block in the current study, despite equivalent increases in Hbmass and V˙O2max, suggests that nonerythropoietic factors also influence endurance performance (22). The failure to enhance performance after the second 3-wk block is similar to previous studies in well-trained athletes where little or no improvement was reported after simulated LHTL (6,10,13,26,27).

The lack of association between performance changes in block 1 versus block 2 demonstrates that enhancement of performance after one bout of LHTL exposure does not guarantee the same response in subsequent altitude exposures. Although a reduction in performance at the end of the second block is a concern for those athletes who regularly engage in repeated altitude exposures in a training year, it may be that the 5-wk recovery period between exposures in the current study was too short and/or that athletes were unable to produce best performances possibly because of fatigue. Further investigation of longer washout periods of several months or up to a year between repeated altitude exposures is warranted, as well as careful quantification of fatigue. The inconsistent performance outcome in this study is at odds with a previous investigation that categorized athletes as "responders" on the basis of greater performance improvement relative to the mean after a single bout of LHTL (5). In the current study, the true individual responses to the intervention for performance, as well as V˙O2max and Hbmass, were of similar magnitude to the mean response, indicating that performance is relatively labile. It furthermore suggests that, after adequate altitude exposure, attributing performance increases of ∼1%-2% solely to increased erythropoiesis may be too simplistic (22).

The variable results in time trial performance, despite similar mean improvements in V˙O2max, are consistent with the model that maximal aerobic power is only one of several factors that contribute to athletic performance. Although V˙O2max is considered a useful predictor of performance in endurance events (30,32), in some top-level athletes, it is poorly associated with performance (33). For athletes more homogenous in ability, other physiological parameters including percent of V˙O2max at a given velocity (9,32) and energy cost of running (9) account for most of the variation in running performance. According to the model of di Prampero (11), V˙O2max, percent of V˙O2max that can be maintained for the duration of the run (fractional utilization) and energy cost of running (economy) account for 70% of between-subject variation in performance. Consequently, a portion of the variability in performance in this study could be related to other factors such as fatigue, training status, and motivation. All of these factors presumably require careful management to realize enhanced performance after altitude training, with or without an increase in Hbmass.

There were no substantial improvements in running economy or vV˙O2max in the LHTL group, despite increased V˙O2max. This lack of improvement in economy after hypoxic exposure is in agreement with those studies that have also demonstrated an increase in V˙O2max (23). In contrast, those studies with no change or even a reduction in V˙O2max have observed 3%-10% improvements in economy (12). Improvements in V˙O2max were associated with lower posttest blood lactate concentration after each 3-wk block in the LHTL group. The reduction in maximal lactate and increased 4 mM running speed after block 1 in the LHTL group may indicate greater reliance on aerobic metabolism or changes in whole-body lactate metabolism after hypoxic exposure (24).

There were no clear differences in change in blood bicarbonate (HCO3) or pH, measured as a surrogate of muscle buffer capacity (βm), compared with the control group. It is possible that our indirect means of assessing buffering capacity may not have been sensitive enough to detect subtle changes after altitude training because direct measurement has previously demonstrated an enhancement in βm (13).


The reproducible small increases in Hbmass indicate that 14 h·d−1 of hypoxic exposure for 2-3 wk was sufficient to induce accelerated erythropoiesis and elicit measurable changes in Hbmass in highly trained runners. The ∼3% mean increase we observed after each block (∼300 h) is consistent with a 1% increase in Hbmass for every 100 h of hypoxic exposure (7). Although this small mean increase is somewhat lower than the ∼5%-10% increases in Hbmass (4,36) previously reported, the individual changes in Hbmass we measured lend support to the notion that some highly trained athletes can demonstrate moderate to large increases in Hbmass (∼4%-7%) after LHTL (35). The capacity for repeated Hbmass measurement with CO rebreathing (25) allows confidence in identifying small changes in Hbmass and elucidating the time course of Hbmass change in relation to hematological responses. Our results after 14 h·d−1 of LHTL confirm that the length of daily hypoxic exposure is critical for stimulating the acute increase in EPO concentration and augmented erythropoiesis. Indeed, other researchers using daily exposures shorter than 12 h have failed to observe increased red blood cell production in athletes (1,2,26,31).


It was not possible to have a period of controlled training before block 1 because of the training and competition demands on the athletes. Nevertheless, there was no substantial difference in their training load between blocks 1 and 2. A limitation of the 4.5-km time trial performance within 1 d of altitude exposure is the possibility of residual fatigue. Nonetheless, substantial improvements in time trial performance after the first 3-wk block indicate that the LHTL group did not suffer any ill effects from time trialing immediately after simulated altitude exposure. It seems that the performance response to altitude training may be mediated by prior training, and individual athletes in this study may not have adequately managed their training load with the additional stress of a second bout of altitude exposure.


A 3-wk simulated LHTL block can induce reproducible mean increases in Hbmass and V˙O2max in highly trained runners, but these enhanced physiological capacities did not transfer directly to reproducible improvements in the 4.5-km time trial performance. There was large individual variation in the change in physiological and performance measures after each block, with true individual responses of similar magnitude to the mean response. Competitive performance is dependent not only on physiological adaptations but also on a complex interaction of fitness, training status, and fatigue. All these factors require careful, individual management to improve performance after altitude training, particularly when multiple exposures are undertaken within a training year.

The authors thank the support of staff in the Department of Physiology at the Australian Institute of Sport, particularly Melissa Clark for the collection and processing of the hematological data and Dr. Daniel Green for technical assistance with testing; Kinetic Performance Technology Ltd, Canberra, for support in operating the altitude house; and financial support from the Australian Institute of Sport Applied Research Centre, Australian Sports Commission, and the Researcher Development Grants Scheme, University of Victoria. The results of the present study do not constitute endorsement by American College of Sports Medicine.


1. Ashenden MJ, Gore CJ, Dobson GP, et al. "Live high, train low" does not change the total haemoglobin mass of male endurance athletes sleeping at a simulated altitude of 3000 m for 23 nights. Eur J Appl Physiol. 1999;80:479-84.
2. Ashenden MJ, Gore CJ, Martin DT, et al. Effects of a 12-day "live high, train low" camp on reticulocyte production and haemoglobin mass in elite female road cyclists. Eur J Appl Physiol. 1999;80:472-8.
3. Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perf. 2006;1:50-7.
4. Brugniaux JV, Schmidtt L, Robach P, et al. Eighteen days of "living high, training low" stimulate erythropoiesis and enhance aerobic performance in elite middle-distance runners. J Appl Physiol. 2006;100:203-11.
5. Chapman RF, Stray-Gundersen J, Levine BD. Individual variation in response to altitude training. J Appl Physiol. 1998;85:1448-56.
6. Clark SA, Aughey RJ, Gore CJ, et al. Effects of live high, train low hypoxic exposure on lactate metabolism in trained humans. J Appl Physiol. 2004;96:517-25.
7. Clark SA, Quod MJ, Clark MA, et al. Time course of haemoglobin mass during 21 days live high:train low simulated altitude. Eur J Appl Physiol. 2009;106:399-406.
8. Cohen J. Statistical Power Analysis for the Behavioural Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988.
9. Conley DL, Krahenbuhl GS. Running economy and distance running performance of highly trained athletes. Med Sci Sports Exerc. 1980;12(5):357-60.
10. Dehnert C, Hutler M, Liu Y, et al. Erythropoiesis and performance after two weeks of living high and training low in well trained triathletes. Int J Sports Med. 2002;23:561-6.
11. di Prampero PE. The energy cost of human locomotion on land and in water. Int J Sports Med. 1986;7:55-72.
12. 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.
13. Gore CJ, Hahn AG, Aughey RJ, et al. Live high:train low increases muscle buffer capacity and submaximal cycling efficiency. Acta Physiol Scand. 2001;173:275-86.
14. Hahn AG, Gore CJ, Martin DT, et al. An evaluation of the concept of living at moderate altitude and training at sea level. Comp Biochem Physiol. 2001;128:777-89.
15. Hopkins WG. A spreadsheet for analysis of straightforward controlled trials. Sportscience [Internet]. 2003 [cited 2008 June 4];7. Available from:
16. Hopkins WG. Spreadsheets for analysis of controlled trials, with adjustment for a subject characteristic. Sportscience. 2006;10:46-50.
17. Hopkins WG. A spreadsheet for deriving a confidence interval, mechanistic inference and clinical inference from a P value. Sportscience. 2007;11:16-20.
18. Hopkins WG, Hawley JA, Burke LM. Design and analysis of research on sport performance enhancement. Med Sci Sports Exerc. 1999;31(3):472-85.
19. Hopkins WG, Hewson DJ. Variability of competitive performance of distance runners. Med Sci Sports Exerc. 2001;33(9):1588-92.
20. Hopkins WG, Marshall SW, Batterham AM, et al. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3-12.
21. Levine BD, Stray-Gundersen J. "Living high-training low": effect of moderate-altitude acclimatization with low-altitude training on performance. J Appl Physiol. 1997;83:102-12.
22. Levine BD, Stray-Gundersen J, Gore CJ, et al. Point: Counterpoint. Positive effects of intermittent hypoxia (live high:train low) on exercise performance are mediated primarily by augmented red cell volume. J Appl Physiol. 2005;99:2053-7.
23. Lundby C, Calbet JA, Sander M, et al. Exercise economy does not change after acclimatization to moderate to very high altitude. Scand J Med Sci Sports. 2007;17:281-91.
24. Nummela A, Rusko H. Acclimatization to altitude and normoxic training improve 400-m running performance at sea-level. J Sports Sci. 2000;18:411-9.
25. Prommer N, Schmidt W. Loss of CO from the intravascular bed and its impact on the optimised CO-rebreathing method. Eur J Appl Physiol. 2007;100:383-91.
26. Robach P, Schmitt L, Brugniaux JV, et al. Living high-training low: effect on erythropoiesis and maximal aerobic performance in elite Nordic skiers. Eur J Appl Physiol. 2006;97:695-705.
27. Robach P, Schmitt L, Brugniaux JV, et al. Living high-training low: effect on erythropoiesis and aerobic performance in highly-trained swimmers. Eur J Appl Physiol. 2006;96:423-33.
28. Robertson EY, Aughey RJ, Anson J, et al. Effects of simulated and real altitude exposure in elite swimmers. J Strength Cond Res. (in press).
29. Rodriguez FA, Truijens MJ, Townsend NE, et al. Performance of runners and swimmers after four weeks of intermittent hypobaric hypoxic exposure plus sea level training. J Appl Physiol. 2007;103:1523-35.
30. Saltin B, Astrand PO. Maximal oxygen uptake in athletes. J Appl Physiol. 1967;23:353-8.
31. Saunders PU, Telford RD, Pyne DB, et al. Improved running economy in elite runners after 20 days of simulated moderate-altitude exposure. J Appl Physiol. 2004;96:931-7.
32. Schabort EJ, Killian SC, St Clair Gibson A, et al. Prediction of triathlon race time from laboratory testing in national triathletes. Med Sci Sports Exerc. 2000;32(4):844-9.
33. Snell PG, Mitchell JH. The role of maximal oxygen uptake in exercise performance. Clin Chest Med. 1984;5:51-62.
34. 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;91:1113-20.
35. Wehrlin JP, Marti B. Live high-train low associated with increased haemoglobin mass as preparation for the 2003 World Championships in two native European world class runners. Br J Sports Med. 2006;40:e3.
36. Wehrlin JP, Zuest P, Hallen J, et al. Live high-train low for 24 days increases hemoglobin mass and red cell volume in elite endurance athletes. J Appl Physiol. 2006;100:1938-45.


©2010The American College of Sports Medicine