Search tips
Search criteria 


Logo of brjsmedBritish Journal of Sports MedicineVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Br J Sports Med. 2007 August; 41(8): 537–539.
Published online 2007 February 20. doi:  10.1136/bjsm.2006.033712
PMCID: PMC2465425

Game performance and intermittent hypoxic training


Live high‐train low altitude exposure simulated by hypoxic devices may improve athletic performance. In this study, intermittent normobaric hypoxia was achieved with the GO2altitude® hypoxicator to determine its effects on sea level performance in rugby players. Ten players were randomly assigned to two groups. Players in each group received 14 sessions of either hypoxic (10–15% O2) or normoxic (21% O2) exposure at rest over 14 consecutive days in a single blind fashion. Various performance measures were obtained consecutively in a single testing session pre‐ and post‐exposure. Effects of hypoxic exposure on maximum speed and sprint times were trivial (<1.0%) but unclear (90% likely range, ±5% to ±9%). In rugby simulation, hypoxic exposure produced impairments of peak power in two scrums (15%, ±8%; 9%, ±7%) and impairments of time in offensive sprints (7%, ±8%) and tackle sprints (11%, ±9%). Pending further research, rugby players would be unwise to use normobaric intermittent hypoxic exposure to prepare for games at sea level.

Keywords: hypoxia, IHT, performance, rugby, training

Intermittent hypoxic training (IHT) is a method by which athletes receive exposure to short bouts of severe hypoxia (9–12% O2), interspersed with periods of normal air.1 Recent studies reported substantial improvements in sea level endurance and anaerobic performance after IHT at rest2 or during exercise.3 These enhancements suggest that IHT may be suitable for improving performance in high intensity team sports.



Fifteen male rugby players were recruited from the Auckland Rugby Academy, of whom 10 completed the study (mean (SD) age, 23 (5) years). Subjects were professional reserve players completing their competitive season. Players were randomised into hypoxic (n = 5) or normoxic (n = 5) groups in a single blind fashion. The subjects' informed consent was obtained with the approval of the institutional research ethics committee.

Study design

Baseline performance and anthropometric data were collected one or three days before and again one or four days after 14 days of hypoxic exposure. Blood samples were collected seven and 14 days before, and two days after exposure.

Hypoxic exposure

Players breathed through hand held face masks connected to the GO2altitude® hypoxicator (Biomedtech, Victoria, Australia) for 36 minutes a day for 14 days. The hypoxic group received hypoxic air (10–15% O2) from the hypoxicator device in six six‐minute intervals separated by four minutes of ambient air.1 The hypoxicator unit was adjusted to deliver hypoxic gas to produce arterial blood saturations as previously described.4 A gradual reduction of arterial oxygen saturation was observed in each player in the hypoxic group (range 100% to 76% O2). The normoxic group completed an identical breathing task but was supplied with oxygen equivalent to ambient air (21% O2). The players' arterial blood oxygen saturation was continually monitored with a pulse oximeter (Nonin, Plymouth, Minnesota, USA).


Overall, the hypoxic group spent 587 (132) min in rugby specific, weight, and aerobic sessions while the control group spent 550 (134) min in the equivalent activities during the study period. Average intensity level for both groups was reported as moderate.

Performance measurements

Players were instructed to do little or no training the day before a testing session. For the 20‐m multistage test, players were required to run repeatedly over a distance of 20 m in time with a “beep” sound on a prerecorded tape until exhaustion. The test began at 11 km.h−1 and increased by 1 km.h−1 every three minutes.

The repeated sprint shuttles consisted of six sets of 70‐m shuttles (5, 10, and 20 m distances) carried out 10 minutes before and 10 minutes after the rugby simulation. Shuttles started every 30 s and sprint performance was recorded for each player by an infrared electronic timing system (Swift Performance Equipment, NSW, Australia).

The rugby simulation was a functional fitness test based on time–motion analysis studies of real match play.5 The test consisted of seven circuits containing a range of activities characteristic of a rugby game. Each circuit incorporated 10 activities which included a 20‐m sprint, offensive sprint, active recovery (walking), scrum, active recovery, defensive sprint, active recovery, tackle sprint, target accuracy, and a 30‐m sprint. Players had 30 seconds to complete each activity. Time in sprints was measured by infrared electronic timing. Counts of accuracy in target ball passing were recorded. Power output produced during the two scrums was recorded using a custom built scrum simulation machine.

Haematological measurements

Venous blood was drawn from the antecubital vein at rest. Samples were analysed at an accredited haematology laboratory (Middlemore Hospital, Auckland, New Zealand) for the determination of packed cell volume, haemoglobin, and ferritin.

What is already known on this topic

  • Live high‐train low altitude exposure simulated with hypoxic devices may improve athletic performance.
  • Recent studies reported substantial improvements in sea level endurance and anaerobic performance after IHT at rest or during exercise.

Statistical analysis

Repeated measure analysis was done using a mixed modelling procedure (Proc Mixed) in the Statistical Analysis System (SAS Institute, Cary, North Carolina, USA). The natural logarithm of each measure was used to reduce any effects in non‐uniformity of error and to obtain changes in measures and errors as percentages.6 The precision of estimates is reported as 90% confidence limits. Chances that the true effects were substantial were estimated with a spreadsheet when a value for the smallest worthwhile effect was entered.6 When the confidence interval of the estimate spanned across substantially positive and negative levels of magnitude we have labelled the outcome as unclear.


Performance measurements

Compared with the normoxic group the hypoxic group's mean time in the six 70‐m repeated sprints showed a trivial but unclear improvement before (0.5% (90% confidence interval (CI), −3.8% to 4.4%)) and after the rugby simulation (0.1% (−4.7% to 4.9%)). Maximum speed in the multistage running test was a trivial but unclear impairment of 0.1 (−5.7% to 5.9%) in the hypoxic compared with the control group.

The effect of hypoxic exposure on activities carried out in the rugby simulation is summarised in table 11.. Standard (typical) error of measurement expressed as a coefficient of variation ranged between 2.4% and 7.7%, depending on the variable.

Table thumbnail
Table 1 Effect of hypoxic exposure

What this study adds

  • This is the first published study to report the effect of IHT on sea level game performance.
  • A regime of one hour a day for 14 consecutive days did not produce a clear effect in some rugby performance measures, while in others it produced a detrimental effect.

Blood variables

Compared with the normoxic group, the hypoxic group showed trivial but unclear changes in haemoglobin (2.0% (90% CI, −2.7% to 6.7%)), haematocrit (0.9% (−4.3% to 6.1%)), and reticulocytes (0.7% (−35.3% to 36.7%)), while serum ferritin showed a clearly trivial increase (10.5% (−11.1% to 32.1%)) following exposure.


This study is the first to investigate the effect of IHT on game performance. We showed that in reserve elite rugby players a regimen of IHT for a total of one hour a day for 14 consecutive days does not produce a clear effect in certain specific rugby performance measures or blood indices, while in others it shows clear impairment.

The impairment observed in some measures may reflect the limited time allowed for recovery and possible supercompensation between hypoxic exposure and testing. It is also possible that the players in the hypoxic group were non‐responders to the hypoxic stimulus. Our results are in agreement with others.7

In future studies, researchers should opt for a larger sample size with more committed players to improve the chances of responders in the treatment group and to avoid problems of reliability. In addition, performance changes should be monitored for one to two weeks after the last day of exposure to allow enough time for recovery.


IHT offered to rugby players had an unclear effect on some performance measures and caused clear impairment in others.


New Zealand Rugby Union funded the study. Les McGrath of Unitec, New Zealand assisted with the performance tests. Robin Archer of Unitec New Zealand collected blood samples. Dr Tony Edwards of Adidas Sports Medicine provided medical advice. The results of the present study do not constitute endorsement of the GO2altitude® hypoxicator by the authors.


IHT - intermittent hypoxic training


1. Serebrovskaya T V. Intermittent hypoxia research in the former soviet union and the commonwealth of independent States: history and review of the concept and selected applications. High Alt Med Biol 2002. 3205–221.221 [PubMed]
2. Hamlin M, Hellemans J. Effect of intermittent normobaric hypoxic exposure at rest on haematological, physiological, and performance parameters in multi‐sport athletes. J Sports Sci 2006. 25431–441.441 [PubMed]
3. Dufour S, Ponsot E, Zoll J. et al Exercise training in normobaric hypoxia in endurance runners. I. Improvement in aerobic performance capacity. J Appl Physiol 2006. 1001238–1248.1248 [PubMed]
4. Hellemans J. Intermittent hypoxic training: a pilot study. Proceedings of the Second Annual International Altitude Training Symposium, Flagstaff AZ 1999. 145–154.154
5. Deutsch M U, Kearney G A, Rehrer N J. Rugby specific fitness testing (RSFT). World Congress of Science and Football IV 1999
6. Hopkins W G. Probabilities of clinical or practical significance. Sportscience [on line]. 2002;Available,
7. Julian C G, Gore C J, Wilber R L. et al Intermittent normobaric hypoxia does not alter performance or erythropoietic markers in highly trained distance runners. J Appl Physiol 2004. 961800–1807.1807 [PubMed]

Articles from British Journal of Sports Medicine are provided here courtesy of BMJ Publishing Group