Article Text

Download PDFPDF

Is the measurement of maximal oxygen intake passé?
  1. Roy J Shephard
  1. Faculty of Physical Education & Health, University of Toronto, Toronto, ON, Canada
  1. Roy J Shephard, PO Box 521, Brackendale, Canada BC V0N 1H0; royjshep{at}

Statistics from

A recent and controversial review1 suggests that the measurement of maximal oxygen intake is passé. The author concludes (p. 554) “It is now time to develop novel testing methods....That the measured VO2max is a relatively poor predictor of both the performance potential of athletes with similar athletic ability and of the changes in performance that occur with continued training should encourage both basic and applied sports scientists to reconsider the real value of this iconic test.

A number of the arguments that are advanced in this review seem to need correction or refutation. Specifically, this riposte will examine whether a maximal treadmill test is an unrealistic procedure for athletes, whether a unimodal approach to testing is appropriate in sports medicine, and whether an alternative laboratory test will be developed to categorise the performance of individual athletes. Comments will also be made on the place of maximal oxygen intake assessment in various areas of science, sports medicine and clinical medicine.


Noakes argues1 that the treadmill test is an unrealistic approach to the testing of athletes for three reasons: the duration of the exercise is not known to the subject, there is a steep and progressive increase in the intensity of exercise, and the person who is tested has no control over the ultimate intensity of effort. All of these criticisms may be true of the test protocol used in some laboratories. However, the standard recommendation is for a treadmill test lasting 9–11 minutes, and this should be explained to the subject. The appropriate treadmill slope and speed should be ascertained by preliminary submaximal testing; this allows the definitive test to commence close to maximal steady-state effort, thereby avoiding a steep ramp of intensity. Moreover, the maximal value can be determined by carrying out a series of tests on successive days, although the values obtained in this manner do not differ materially from those seen with an appropriately conducted continuous test protocol.2 Finally, protocols are available that allow the subject to regulate the speed of the treadmill himself or herself as the test proceeds.

There are thus ways of dealing with the problems cited by Noakes. However, there remain some important differences between a laboratory treadmill test and athletic performance that Noakes does not discuss. Most treadmills cannot match the speed of short-distance runners. While on a treadmill, an athlete does not encounter wind resistance3 and cannot profit from “drafting”.4 Few laboratories simulate radiant heating or cooling by cross-winds, and the running surface of the treadmill differs substantially from the usual track. However, these various limitations apply only to laboratory measurements, and do not negate the value of maximal oxygen intake testing, since it is quite practicable to measure oxygen consumption when an athlete is running on a standard track.5

Other factors affecting athletic performance include an appropriate choice of tactics (including selection of clothing), motivation of the individual, and the mechanical efficiency of movement. No laboratory test seems likely to evaluate tactical skills. The motivating power of the observer is important to the reaching of an oxygen consumption “plateau”, and laboratories that have difficulty in demonstrating this phenomenon6 probably need to upgrade their motivational skills. Mechanical efficiency can be estimated roughly from treadmill data, although other forms of ergometer provide more precise values.7


Another puzzling feature of the recent review1 is the apparent assumption that everyone will be tested on a treadmill. The International Working Party that standardised procedures for the measurement of maximal oxygen intake suggested that three possible modes of testing were suitable for ordinary, non-athletic individuals, based on a treadmill, a double step and a cycle ergometer.8 In non-athletic individuals, the largest values were obtained on the treadmill; step test readings were on average 4 per cent smaller, and cycle ergometer values 7 per cent smaller.

Intermodal differences in maximal oxygen intake are thought to reflect the differing proportions of the total muscle mass that are used in the different protocols9; intermodal differences are much larger in athletes who have trained one particular group of muscles. For athletes, it is thus critical to select a test modality that allows use of the muscles engaged in their chosen sport; for a cyclist, the natural choice would be a cycle ergometer or a racing bicycle mounted on rollers,10 for the oarsperson a rowing ergometer or the collection of gas samples during actual rowing,11 for the cross-country skier a skiing ergometer or uphill skiing,12 and for the swimmer a flume or the collection of gas samples during swimming in a pool.13 14 Plainly, for many categories of athlete, the treadmill is an inappropriate test modality.


Given the various constraints noted above, it seems unlikely that any laboratory test can be devised that will categorise athletes of similar ability with useful accuracy. Rather, athletes should be ranked based on the actual performances achieved over several recent competitions. Here, tactics, motivation and mechanical efficiency all come into play, and the mode of exercise is entirely natural. Moreover, individual performances can be determined to a small fraction of a second. This compares very favourably with the laboratory treadmill, since even careful determinations of maximal oxygen intake have an experimental error of 2–4 per cent, and superimposed upon this is an intraindividual biological variation of 10–20 per cent.15


Maximal oxygen intake tests are of little help in ranking athletes of similar ability. However, this does not imply that such measurements are passé. On the contrary, the testing of maximal oxygen intake has many appropriate and important applications. This concluding section highlights just a few such applications, in integrative biology, sports medicine, doping control, epidemiology and clinical medicine.

Integrative biology

Analysis of the oxygen conductance equation under conditions of maximal aerobic effort16 provides helpful insights into the factors limiting various types of physical performance.9 In activities that involve large muscle groups, oxygen transport is determined almost entirely by maximal cardiac output, and, taken together with determinations of heart rate, the measurement of maximal oxygen intake offers a useful method of examining cardiac stroke volume during maximal aerobic exercise. If the external work is measured, the mechanical efficiency of various types of activity can also be determined.7

Sports medicine

Grouped data on the maximal oxygen intake of top competitors provide insights into the extent of aerobic demands in various forms of sport.17 Comparisons of the grouped physiological profile with the individual’s personal data may suggest how large an emphasis the individual should place upon enhancing his or her aerobic power. Careful records of an individual’s maximal oxygen intake are also helpful in evaluating the extent of any deterioration in aerobic function following injury or overtraining, and in monitoring the recovery of aerobic function following such episodes.18

Doping control

A variety of drugs19 and manipulations such as blood transfusions20 can induce small increments of maximal oxygen intake that have an important influence on the outcome of endurance competitions. Measurements of maximal oxygen intake under carefully controlled double-blind trials are thus important in distinguishing medications that are acceptable (for instance, certain drugs used in the treatment of bronchospasm induced by exercise or cold) from those which would give athletes an unfair advantage.


It is widely agreed that the majority of the population in developed countries currently takes insufficient physical activity, with adverse consequences for many facets of health.21 Nevertheless, the determination of the physical activity of a population, whether by questionnaire or by use of accelerometers, is unreliable, and the mass testing of maximal oxygen intake provides a helpful alternative approach when assessing the extent of endurance activity within a population.2 Maximal oxygen intake data can also be used to demonstrate secular trends in habitual physical activity; for instance, repeated measurements in a Canadian circumpolar community have shown a progressive decrease of maximal oxygen intake as the community has adopted the sedentary lifestyle typical of their peers in southern Canada.22

Clinical medicine

In many clinical situations, the oxygen intake is limited by warning symptoms or signs before a “plateau” is reached. The value thus reported is termed the peak rather than the maximal oxygen intake. The peak aerobic power provides a useful indication of prognosis in patients with various types of cardiac disorder.23 Such observations also provide an optimal basis for the setting of a safe and effective intensity of training in programmes of cardiac rehabilitation.24 Clinical information can be derived from the extent of ST depression at various fractions of an individual’s maximal oxygen intake.25 Finally, determinations of maximal oxygen intake are helpful in gauging recovery following bed rest, injury, and exposure to zero gravity environments.


View Abstract


  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.