The VO 2 max plateau is not associated with the anaerobic capacity in physically active subjects

Traditionally, maximum oxygen consumption (VO 2 max) has been used to represent the maximum aerobic capacity. Presently, VO 2 max is used as an indicator of cardiorespiratory  tness, running performance predictor, to evaluate training related adaptations in healthy individuals and in patients with coronary arterial disease, detraining, mortality predictor and to evaluate sleeping disorder. Thereby, VO 2 max identification is important to evaluate the  tness levels in athletes as well as in high-risk groups. VO 2 max is measured via incremental tests (TI), usually performed to voluntary exhaustion. Although several variables have been considered in order to establish maximum e€ ort, the main characterization of VO 2 max is through stabilization in oxygen consumption (VO 2 ) during the final stages of TI. This VO 2 characterization during the  nal stages of TI has been termed VO 2 plateau.  eoretically, plateau refers to a stabilization or small increases (≤ 2.1 ml.kg.min) of VO 2 , even if loads are incremented in the  nal stages of TI. However, some tests are interrupted before reaching VO 2 max. In this case, the value obtained is termed as peak oxygen consumption (VO 2 peak). It has been suggested that plateau incidence may be related to athlete’s training state, in which athletes with higher physical condition could tolerate higher levels of pain and fatigue and higher motivation to support higher loads in the final test, and that the higher intensities could be related to increases in energy supply by the anaerobic metabolism. Provided the relevance to establish VO 2 max parameters, previous studies used one additional test to con rm if VO 2 value, obtained in a traditional TI protocol, could be considered as maximum.  e con rmation test is performed until fatigue with constant loads and intensities near to the VO 2 max. For instance, Snell et al. performed the con rmation test with two intensities (95% and 105% of the maximum power in TI) and, in both conditions, it were not observed signi cant di€ erences between VO 2 in the veri cation test and the VO 2 max achieved in TI, even in the absence Abstract