In a comprehensive review on training, Midgley and McNaughton’s first sentence state’s “The maximal oxygen uptake (VO2max) has been suggested to be the single most important physiological capacity in determining endurance running performance” (2006). Based on this notion, training for distance runners has become fixated on the concept of VO2max. Training to enhance VO2max is the subject of numerous review articles and popular coaching material. A whole theory of training has evolved based on the idea of training at the speed that corresponds with VO2max, and at certain percentages of VO2max (Daniels, 2005). Given the emphasis on this particular parameter one would assume that it must be very closely tied with performance and fatigue. It’s not.
In the following paper the limitations of VO2max will be discussed. Including the legitimacy of the variable itself, why it arose to such prominence, the efficacy of basing training paces off of it, should we even train to improve it, and how closely it ties to performance.
How the VO2max concept developed:
The ability to measure oxygen consumption first arose in the early 1920’s. It was in 1923, when A.V. Hill and his partner H. Lupton came up with the idea there being an upper limit on oxygen consumption. In an experiment which consisted of Hill running at various speeds around a grass track while measuring VO2, it was found that Hill reached a VO2max of 4.080 L/min at a speed 243m/min (Bassett, 2000). Despite increases in speed, his VO2 did not increase, leading Hill to conclude that there is a maximum limit to oxygen consumption, or in his words:
“In running the oxygen requirement increases continuously as the speed increases attaining enormous values at the highest speeds: the actual oxygen intake, however, reaches a maximum beyond which no effort can drive it… The oxygen intake may attain its maximum and remain constant merely because it cannot go any higher owing to the limitations of the circulatory and respiratory system” (Noakes, 2008, pg. 575).
These findings led to two lasting conclusions. First, that VO2max is limited by the circulatory and respiratory system. The second conclusion was the result of trying to device a laboratory test for determining VO2max, in which thirty years later, Taylor et al. decided that during a graded exercise test, a VO2max was obtained when a plateau occurred in VO2 (Noakes, 2008). However, in Taylor’s original definition, a plateau was not a true plateau but it rather consisted of a VO2 increase of less than 150ml/min from one workload to the next. These findings led to the idea that in order for a true VO2max to be reached, a plateau of the VO2 should occur.
Understanding how the VO2max test came about is important as it impacts the way we currently view and use the parameter. The fact that VO2max was first measured during exercise by one of the pioneers of Exercise Science in the 1920’s goes a long way in explaining the level of importance ascribed to it. Whenever a new parameter is discovered or introduced, a large degree of emphasis is put on that parameter in the research. The initial reaction by many scientists is to ascribe a great deal of significance to the newly discovered parameter, as if it will answer all of the questions that we have. It is almost as if it is human nature to go through this process of discovery and then exaggeration of the importance of the new finding. This can be seen in many instances in a wide degree of scientific fields. In Exercise Science, this may best be demonstrated by the rise of the anaerobic or lactate threshold during the 1970’s, 80’s, and 90’s. With the ability to portably test lactate, research was centered on ways to improve lactate threshold and the various methods to test for it. Coaches also devised ways of using lactate testing as a way of manipulating the training of their athletes. Whenever something is new, it is overemphasized, before it usually settles into its rightful place of importance over time.
Due to the very early development of the VO2max concept, a large amount of early research and study was focused on it, escalating the importance given to the parameter. In addition, theories were developed utilizing the VO2max concept very early on. The problem is that early development of the VO2max concept created a situation where there was enormous amount of data and research surrounding it, in essence creating a concept that is too large to break down. It is almost as if the field of Exercise Science was built upon the VO2max concept.
Recently, the legitimacy of VO2max both as a measurement and the acceptance of VO2max as a practical measurement of cardio-respiratory endurance has been called into question. Their contention is that VO2max is not actually a representative measure of the maximum ability to transport oxygen, but is rather controlled by a central governor. In Tim Noakes’ Central Governor Model (CGM), the CGM predicts that the body regulates exercise to prevent myocardial ischemia during exercise. This is accomplished by limiting the blood flow to the periphery which the brain accomplishes by regulating muscle recruitment (Noakes & Marino, 2009). Therefore, VO2max reflects this regulation of muscles recruitment. In essence, a central governor acts as a regulator for exercise instead of exercise being limited by some parameter.
There are several theoretical arguments for this model. Noakes and other CGM proponents point to the fact that fatigue is seldom catastrophic like would be predicted in traditional models. Instead, the body uses various feedback information and past experiences to modulate power output or in the case of running, pace. The idea of pacing being prevalent in endurance events and the fact that a finishing kick, or endspurt, occurs are given as further evidence to support this model (Noakes, 2003). Interestingly, evidence of alterations in pacing strategy and EMG, which measures muscle activation can be seen from the very beginning of performance, such as that seen when racing in warm versus cool weather, which leads credence to the anticipation model of fatigue (Noakes, 2008).
An increase in muscle activation is also seen during the last segments of races, which should not be able to occur if the muscle is “failing” due to fatigue. Noakes’ hypothesis is that at the end of a race, the body’s feedback says that it is near completion so that is can push slightly more into its capacity (Noakes, 2008). Evidence for this hypothesis can be seen in a study by Tucker et al. that found that when completing a 20-km cycling trial in normoxia versus hyperoxia, the improvement in power output in hyperoxia was proportional to the increase in iEMG that also occurred, which the authors cited as evidence that control of muscle activation was one way in which performance was regulated (2007).
Another interesting point raised in the CGM debate is the effect hypoxia has on Cardiac Output. Exercise in hypoxic conditions show a reduction in peak Cardiac Output, due to both a decrease in HR and SV (Calbet et al., 2003). According to the conventional model Cardiac Output, since it is regulated by muscle oxygen demands, should not be reduced. However in the CGM, Cardiac Output is reduced as a regulatory mechanism and is determined by the work done by the muscles (Noakes, 2004). Thus, a reduction in Cardiac Output in hypoxia is due to a decrease in muscle activation, which when supplementary oxygen is taken, Cardiac Output immediately increases to normal levels (Noakes, 2004). This immediate increase in Cardiac Output demonstrates that there is a regulatory mechanism in control and one has to question why Cardiac Output is reduced at altitude when oxygen demand by the muscles should be higher.
In regards to VO2max and how it is tested, Noakes has pointed out that in most cases the original requirement of seeing a plateau in VO2max during an incremental exercise test does not occur (Noakes, 2008). Demonstrating this lack of plateau, in a study on world class cyclists only 47% reached a plateau, prompting the authors of the study to state that their limitations might not be oxygen dependent (Noakes, 2008). It is amusing that some authors have commented that motivation may be the reason some athletes do not reach a plateau (Shephard, 2009). This could be a valid statement if the subjects were sedentary, however since the above study was with world class cyclists, it seems a bit ludicrous to suggest that motivation during a maximum test would be a problem in such athletes. In addition, in other studies, one by Hawkins et al., there have been individual variations in VO2max levels between the traditional incremental test and a supramaximal test (Noakes, 2008). While in the average of the whole group there were no differences between the tests, the fact that certain individuals showed different VO2max is interesting and shows that the traditional test does not always give the highest VO2.
Combining the fact that a plateau does not occur in many subjects and the fact that some individuals reached higher VO2max values during a supramaximal test than the standard incremental one, the use of the standard incremental VO2max should be called into question. Other studies show that knowing or not knowing when a test or trial will end significantly effects physiological parameters, which lends credence to the aforementioned idea. In a study by Baden et al. they demonstrated that Running Economy significantly changed, along with RPE, during a submaximal run based on whether the group knew they were running 20 minutes or whether they did not know, even if they ended up running 20 minutes (2005). The VO2max test is one in which participants do not have an exact finish distance or time, so it is possible that this degree of uncertainty could affect the physiological parameters measured. The study also points to the importance of feedback and anticipation and that it can affect physiological variables.
One final point on VO2max testing is why variation exists based on exercise testing mode (Basset & Boulay, 2000). A runner tested running versus another modality such as cycling will have different VO2max values. There is great individual variation too, between 0 and 13% in the aforementioned study. If we recognize that regardless of exercise the oxygen cascade from the air through delivery via Cardiac Output are central adaptations and should not be different between the exercise modes, then the change in VO2max must either happen on the muscular level or it is regulated via muscle recruitment. This would explain why elite cyclists reach higher percentages of treadmill VO2max when testing cycling VO2max compared to lower level cyclists (Basset & Boulay, 2000). Lastly, the fact that muscle mass activation seems to be the major reason for variations in VO2max among a whole variety of testing methods, shows that muscle activation may play a significant role in determining VO2max, at least to a certain point (Dalleck et al., 2004).
Considering this new theory of fatigue, and the fact that the requirement used for reaching VO2max does not occur in many subjects, the use of VO2max as a testing parameter is called into question. In addition, if VO2max is regulated, then the question arises if it accurately reflects cardio-respiratory endurance. If we accept this to be true, then using VO2max and percentages of VO2max for training might not give the training response that we think it does.
Efficacy of basing training paces off of VO2max
With the rise of VO2max research, training is based on the parameter in two ways. First, training at the speeds that elicit VO2max has become the magic training intensity which supposedly elicits the most improvements. Secondly, training at percentages of VO2max has become en vogue as a way to quantify training intensity.
In regards to training at VO2max, this arose because of a review of research that showed that the largest improvements in VO2max occurred when training at an intensity that corresponded with the parameter, irregardless of duration of the exercise (Wenger & Bell, 1986). This finding was subsequently used to demonstrate that training at VO2max was the best intensity for improving endurance in all groups of people. There are two problems with this conclusion. First, the studies findings are generalized to all groups, even though, as we will talk about later, VO2max does not improve in well trained individuals. Secondly, VO2max and endurance performance are used almost synonymously, which is not true, as discussed earlier VO2max may not even measure cardio-respiratory endurance and is certainly not the only factor in endurance performance.
Despite these concerns, training at VO2max has risen to prominence. In looking at the research, there are countless studies and reviews that focus on training at this intensity (Midgley et al., 2006). It has gone so far, that maximizing the time spent at VO2max has garnered much attention (Midgley et al., 2006). Researchers have studied the various interval training programs with the sole goal in seeing how much time at VO2max each subject spent during the training, which in itself is interesting because it shows the emphasis on the parameter instead of performance. The thought is that time spent at VO2max is the stimulus needed to improve VO2max. However, this theory has not been substantiated by research. For instance, in a study by Billat et al. after 4 weeks of training using an interval program designed to elicit time at VO2max, VO2max and, more importantly, performance did not improve (1999). In addition, even in untrained people, the original review by Wenger and Bell stated that improvements in VO2max at high intensities were not dependent on the volume of training (1986). Despite these facts, researchers continue to press on with the idea that time spent at VO2max is the key ingredient for improved endurance, even though no research backs up this theory.
Using %Vo2max to quantify intensity is an accepted practice in research and is used in many training programs, such as those prescribed by Jack Daniels and Joe Vigil (Vigil, Daniels, 2005). The problem with this approach is that each individual will have a wide range of adaptation, even if training at the same percentage of VO2max. This occurs due to differences in the individuals physiology. For instance, lactate threshold can occur at wide range of %VO2max, even in trained individuals (Brooks and Fahey, 2004). As an example, if two trained runners both performed at a fixed intensity at 80% VO2max, one can be below lactate threshold and one above. This would substantially impact the energetics of the workout, as can be seen in a study that showed there was a 40-fold range for increases in lactate levels at 70% VO2max among individuals (Vollaard et al., 2009). In a recent study by Scharhag-Rosenberger et al. they tested whether exercising at the same %VO2max resulted in similar metabolic strain. They found large individual variance in the lactate response at the fixed intensity, even if groups were matched for similar VO2max values. This led them to conclude that the use of percent VO2max values for training or research should not be used if the goal is to have similar metabolic strain by the exercisers.
In addition to lactate differences, other factors such as the individuals substrate use, fiber type, and other physiological variables will all vary considerably at a fixed percent of VO2max. This was demonstrated in a recent study by Vollaard et al. (2009). The study showed that while on average improvements were seen in a variety of endurance parameters after six weeks of endurance training, the individuality of the response was widespread with some showing even negative responses to the training, even though the training was at the same 70%VO2max intensity for all subjects (Vollaard et al., 2009). The study showed that there was a wide range of adaptation in maximal and submaximal tests including VO2 parameters, muscle enzyme activity, and metabolite levels. An interesting finding in the study is that low responders for an increased VO2max were not low responders in other parameters. The change in VO2max did not correlate with the change in performance on a time trial, which is a significant finding demonstrating that perhaps more attention should be paid to changing in performance instead of manipulating physiological parameters such as VO2max. One has to question the training recommendations based on training designed at improving parameters such as VO2max, with the assumption being that performance will improve because of it, when studies show that change in VO2max are often not linked with a change in performance. This phenomenon of varied response is not new and can be seen in a wide array of training situations, such as altitude training for example (Chapman et al., 1998).
Knowing the wide variance in adaptation that can occur when training at a fixed percent of VO2max, its use has to be called into question. In fact, the author’s of the study questioned the use of %VO2max as a way to standardize intensity and suggested standardization on parameters that more directly effect power output. These findings combined with those by Scharhag-Rosenberger et al. suggest that the use of %VO2max should be eliminated if the goal is to standardize an intensity. One has to really wonder about training programs that use %VO2 to prescribe training as what adaptations will take place are almost a crapshoot. This does not seem like a scientific way to train, as it is portrayed. In practical terms for trained distance runners, it probably makes more sense to standardize paces in relation to their recent race performances or based on percentages of goal race pace in well trained runners.
Should we train to improve VO2max?
As mentioned previously, studies have shown that training at VO2max elicits the most improvement in VO2max. This has been used as reasoning for training at VO2max because, as previously discussed, VO2max is the traditional measurement for endurance. The logic is that if VO2max is increased, endurance performance increases. This may not necessarily be the case. In addition, the question arises if VO2max actually improves in well trained runners? It doesn’t.
Showing the separation of VO2max and performance, the Vollaard et al. study found that the change in VO2max was not related to the change in time trial performance (2009). Studies demonstrate improved performances without changes in VO2max (Daniels et al. 1978). Also, studies show that VO2max can improve without changes in performance, which is seen in a study by Smith et al. that showed improvements in VO2max by 5.0% without an improvement in performance over either 3,000m or 5,000m (2003). In addition, in looking at long term changes in performance in elite athletes, changes in performance occur without subsequent changes in VO2max.
In highly trained athletes, many studies have shown that VO2max does not change, even with performance improvements. In one of the only studies done on a large group (33) of elite runners, Arrese et al. tracked changes in Vo2max across three years. Performance improved by an average of 1.77% in men, and .69% in women, with VO2max remaining essentially unchanged (~76.56 vs. ~76.42 in men, and ~70.31 vs. ~70.05 in women) (Legaz Arrese et al., 2005). Similarly to the case study by Jones, this points to improved performance in elite runners without changes in VO2max. Furthermore, it has been shown that among homogenous groups, such as well trained runners, VO2max does not correlate well with performance and can not be used to distinguish what runners are faster (Legaz-Arrese et al., 2007).
Further evidence can be seen in two case studies on elite runners. In a study on a female Olympic level runner, Jones showed that while the athlete’s 3,000m time improved by 46 seconds, there VO2max decreased from 72 ml/kg/min down to 66 ml/kg/min (Jones, 1998). Another study by Jones, this one on the current women’s marathon world record holder, found that while VO2max varied some based on the time of testing, it was essentially stable at 70 mL • kg–1 • min–1 from 1992 to 2003 (Jones, 2006). The fact that Radcliffe’s Vo2max was essentially stable despite her training volume and intensity increasing substantially is intriguing. Her training increased from a modest 25-30 miles per week (and her VO2max was already 72 at the time) to 120-160 miles per week. The fact that VO2max did not change despite this massive increase in volume and intensity points to the short time course of changes in VO2max.
The rapid change in VO2max can even be seen in untrained individuals. In a study by Smith and Donnell, they evaluated the changes in VO2max over a 36 week training period (1984). VO2max substantially increased by 13.6%, but all of those gains were seen in the first 24 weeks of the study with no further increases during the final 12 weeks. Similarly in a study by Daniels et al. in untrained subjects VO2max increased during the first 4 weeks of training, but did not increase after that even with a further increase of training, despite continued improvements in performance (1978). Given the evidence that VO2max does not change in elite runners and does not correlate with performance, training focused on improving VO2max does not seem like a logical idea for well trained runners.
Vollaard et al. may have put it best when they came to the conclusion that “Moreover, we demonstrate that VO2max and aerobic performance associate with distinct and separate physiological and biochemical endpoints, suggesting that proposed models for the determinants of endurance performance may need to be revisited (2009, pg. 1483)”. There recognition that aerobic performance and VO2max are not direct equals or even well linked is a step in the right direction and needs to be acknowledged to a much greater degree. Combining these findings with Noakes’ CGM creates a situation where VO2max may not be measuring what we think it is. Adding the facts that using %VO2 to classify training results in a wide range of adaptations and changes in VO2max do not occur in trained athletes, one has to question basing entire training programs on VO2max.
The bottom line question that needs to be asked is why is so much of training focused on a variable that does not change in well trained athletes, barely changes in moderately trained, levels off after a short period of time, and does not even correlate well with performance? Does this sound like a variable that we should be basing all of our training off of?
Baden, D. A, McLean, T. L., Tucker, R., Noakes, T. D., & St Clair Gibson, A. (2005). Effect of anticipation during unknown or unexpected exercise duration on ratings of perceived exertion, affect, and physiological function. Br J Sports Med, 39(1), 742–746.
Basset, F. A., & Boulay, M. R. (2000). Specificity of treadmill and cycle ergometer tests in triathletes, runners and cyclists. Eur J Appl Physiol, 81(3), 214–221.
Bassett, D. R. & Howley, E. T. (2000). Limiting factors for maximum oxygen uptake and determinants of endurance performance. Medicine and Science in Sports and Exercise, 32, 70–84.
Billat, V. L., Fletchet, B., Petit, B., Muriaux, G., & Koralsztein, J. P. (1999). Interval training at VO2max effects on aerobic performance and overtraining markers. Med Sci Sports Exerc, 31, 156–163.
Brooks, G. A., Fahey, T. D., & Baldwin, K. (2004). Exercise Physiology: Human bioenergetics and its application. McGraw-Hill.
Calbet, J. A., Boushel, R., Radegran, G., Sondergaard, H., Wagner, P. D., & Saltin, B. (2003). Determinants of maximal oxygen uptake in severe acute hypoxia. Am J Physiol Regul Integr Comp Physiol, 284(2), 291–303.
Chapman, R., Stray-Gunderson, J., & Levine, B. D. (1998). Individual variation in response to altitude training. J. Appl. Physiol, 85(4), 1448–1456.
Dalleck, L. C., Kravitz, L., & Robergs, R. A. (2004). Maximal exercise testing using the elliptical cross-trainer and treadmill. Journal of the Exercise Physiology, 7(3), 94–101.
Daniels, J. T., Yarbrough, R. A., & Foster, C. (1978). Changes in VO2 max and running performance with training. Eur J of Appl Physiol, 39(4), 249–254.
Daniels, J. (2005). Running Formula. Champaign, IL: Human Kinetics.
Jones, AM (1998). A five year physiological case study of an Olympic runner.Br J Sports Med 32: 39–43.
Jones AM (2006). The physiology of the world record holder for the women’s marathon. Int J Sports Sci Coaching 1,101–116.
Legaz Arrese, A., Serrano Ostáriz, E., Jcasajús Mallén, J. A., & Munguía Izquierdo, D. (2005). The changes in running performance and maximal oxygen uptake after long-term training in elite athletes. J Sports Med Phys Fitness, 45(4), 435–40.
Legaz Arrese, Munguía Izquierdo, D., Nuviala Nuviala, A., Serveto-Galindo, O., Moliner Urdiales, D., & Reverter Masia, J. (2007). Average VO2max as a function of running performances on different distances. Science & Sports, 22(1), 43–49.
Midgeley, A. W., McNaughton, L. R., & Wilkinson, M. (2006). Is there an optimal training intensity for enhancing maximal oxygen uptake of distance runners?: empirical research findings, current opinions, physiological rationale and practical recommendations. Sports Med, 36(2), 117–132.
Noakes, T. D. (2003). Commentary to accompany training and bioenergetic characteristics in elite male and female Kenyan runners. Med Sci Sports Exerc, 35(2), 305–306.
Noakes, T. D., Calbet, J. A., Boushel, R., Sondergaard, H., Radegran, G., Wagner, P. D., & Saltin, B. (2004). Central regulation of skeletal muscle recruitment explains the reduced maximal cardiac output during exercise in hypoxia. Am J Physiol Regul Integr Comp Physiol, 287(4) R996-999. author reply R999–1002.
Noakes, T. D. (2008). How did A.V. Hill understand the VO2max and the “plateau phenomenon”? Still no clarity? Br J Sports Med, 42(7), 574–580.
Noakes, T. D., & Marino, F. E. (2009). Point: counterpoint: maximal oxygen uptake is/is not limited by a central nervous system governor. J Appl Physiol, 106, 338–339.
Scharhag-Rosenberger, F., Meyer, T., Gabler, N., Faude, O., & Kindermann, W. (2009). Journal of Science and Medicine in Sport, in press.
Shephard, R. J. (2009). Is it time to retire the ‘central governor’. Sports Med, 39(9), 709–721.
Smith, T. P., Coombes, J. S., & Geraghty, D. P. (2003). Optimising high-intensity treadmill training using the running speed at maximal O(2) uptake and the time for which this can be maintained. Eur J Appl Physiol, 89(3-4), 337–343.
Smith, D. A. & O’Donnel, T. V. (1984). The time course during 46 weeks’ endurance training of changes in Vo2max and anaerobic threshold as determined with a new computerized method. Clin Sci, 67(2), 229–236.
Tucker, R., Kayser, B., Rae, E., Raunch, L., Bosch, A. & Noakes, T. (2007). Hyperoxia improves 20 km cycling time trial performance by increasing muscle activation levels while perceived exertion stays the same. Eur J Appl Physiol, 101(6), 771–781.
Vigil, J. (1995). Road to the Top. Creative Designs, Inc.
Vollaard, N. B. J., Constantin-Teodosiu, D., Fredriksson, K, Rooyackers, O., Jansson, E., Greenhaff, P. L., Timmons, J. A., & Sundberg, C. J. (2009). Systematic analysis of adaptations in aerobic capacity and submaximal energy metabolism provides a unique insight into determinants of human aerobic performance. J Appl Physiol, 106, 1479–1486.
Wenger, H. A. & Bell, G. J. (1986). The interactions of intensity, frequency and duration of exercise training in altering cardiorespiratory fitness. Sports Med, 3(5), 346–356.