V˙O2max and the Plateau-Needed or Not?

Maximal oxygen uptake (V˙O2max) describes the greatest rate at which the cardiovascular system can deliver blood to the working muscles during severe dynamic exercise involving a large proportion of the muscle mass (6). The concept of V˙O2max and its dependence on cardiac output was described by Hill and Lupton in 1923 (3). Over the following decades V˙O2max was accepted as a measure of the functional capacity of the cardiovascular system (6), and is taken as a measure of cardiorespiratory fitness (1). Throughout this time investigators have had to deal with a central issue in the measurement of V˙O2max: how can one be certain that a true V˙O2max had actually been obtained? The concept of a plateau in oxygen uptake with increasing intensities of exercise (using a discontinuous format) was central to Hill and Lupton's description of V˙O2max. However, in practice, great variation exists in the percentage of subjects who achieve a plateau. This led to the development of secondary criteria for having achieved V˙O2max (e.g., postexercise blood lactate > 8 mM; respiratory exchange ratio > 1.15) (4). The fact that continuous (incremental) exercise tests have become the most common protocol complicates the issue. In Snell et al.'s paper (7), the issue of the plateau is revisited, using 52 well-trained distance runners as subjects. In this carefully done study, using classic measurement techniques, the investigators showed convincingly that the V˙O2max achieved in an incremental running test was not different from that achieved in a subsequent running test requiring a V˙O2 30% greater than that measured in the incremental test. In short, this study offers clear support for the underlying proposition of Hill and Lupton 84 years ago-that is, there really is an upper limit to oxygen uptake. However clear the results of this study, the continuing dilemma faced by many investigators, who find only a small percentage of subjects experiencing a plateau in an incremental exercise test, remains. The study by Day et al. (2) may offer some relief. They measured V˙O2 on 71 subjects during an incremental ramp-type protocol taken to the point of fatigue. At test termination only 12 of 71 subjects had demonstrated a plateau in V˙O2max (relative to a V˙O2max value extrapolated from submaximal V˙O2 measurements). In 40 of 71 there was no difference between the extrapolated value and the measured V˙O2max, and in 19 of 71 the measured value was higher than the estimated V˙O2max value. Thirty-eight of the 71 subjects subsequently completed a constant load test at a power output ~90% of the peak power attained in the incremental test. Subjects terminated the test due to fatigue in 4-10 min. The V˙O2max achieved in the constant load test was not different from that measured during the incremental test. Additional tests on six subjects were conducted at a variety of higher power outputs and confirmed the V˙O2max values. The central point in this manuscript was that even though a plateau was not observed in the vast majority of the cases in the incremental test, the V˙O2max was not different from that obtained in a follow-up constant load test. The use of a power output in the follow-up test was lower than that achieved in the incremental test, which is at odds with conventional approaches to follow-up testing such as described in Snell et al. (7). However, the oxygen demand of the peak power achieved in a ramp protocol is well above the measured V˙O2, as evidenced by the subjects fatiguing in 4-10 min at a load approximately 90% of peak power. A recent study (5) from the same group used a follow-up constant load test 5 min after the subjects had fatigued in a standard incremental ramp protocol test. The follow-up tests were done at 95% and 105% of the peak power achieved on the incremental test. For five subjects, who completed both constant load protocols, fatigue occurred at an average of 130 and 89 s, respectively. The V˙O2max achieved in each constant load test was the same as that measured at the end of the incremental ramp protocol, in which no subject had demonstrated a plateau. This study showed that one could obtain confirmation of V˙O2max within the same testing session. The issue of "being sure" about the V˙O2max value will be with us for some time. Snell et al. (7) demonstrated that a true plateau in V˙O2 can be measured, which confirms beyond any doubt the proposition of Hill and Lupton (3). In a further clarification of the issue, Day et al. (2) and Rossiter et al. (5) showed that a plateau is not a prerequisite for being able to measure a true V˙O2max. EDWARD T.