Modelling of aerobic and anaerobic energy production during exhaustive exercise on a cycle ergometer

An extension of the original hyperbolic model (Model-2) was proposed by using power output required to elicit maximal oxygen uptake (P _t). This study aimed to test this new model (Model-α) using mechanical work produced during cycle ergometry. Model α assumed that power exceeding a critical power (P _c) was met partly by the anaerobic metabolism. The parameter α was the proportion of the power exceeding P _c provided by anaerobic metabolism, while power exceeding P _t was exclusively met by anaerobic metabolism. Aerobic power was assumed to rise monoexponentially with a time constant τ. The exhaustion was assumed to be reached when the anaerobic work capacity W′ was entirely utilised. Twelve subjects performed one progressive ramp test to assess the power at ventilatory threshold (P _VT) and P _t and five constant-load exercise to exhaustion within 2–30 min, with one to estimate the maximal accumulated oxygen deficit (MAOD). Parameters from Model α were fitted with τ equal to 0, 10, 20 and 30 s. Results in goodness-of-fit was better than Model-2 whatever the value assumed for τ (P < 0.05). The value of τ did not affect much the estimates for P _c and α. P _c estimates were significantly correlated with P _c from Model-2 and with P _VT. W′ estimates, which were dependent on the value ascribed to τ, were not statistically different than MAOD. These two variables were, however, not significantly correlated. In conclusion, Model α could provide useful information on the critical power and the anaerobic contribution according to exercise intensity, whereas W′ estimates should be used with care because of the sensitivity to the assumption on aerobic power kinetics τ.