The lactate anaerobic threshold (AT) determined during an incremental test has been used generally to estimate the maximal lactate steady-state intensity (MLSSint) in several sports. Furthermore, this index could be useful to predict the time-trial cycling performance and also to prescribe training intensity to enhance aerobic capacity. The aim of this study was to compare three different AT estimations with actual MLSSint in trained cyclists. Fourteen trained cyclists participated in this study. They had previously performed a maximal incremental cycling test (35 W increments each 3 min) in a laboratory followed by three to five visits to measure the MLSSint (30-min tests). Blood lactate concentration ([La]), oxygen uptake ([Vdot]O2), and heart rate (HR) were measured during all tests. Based on the incremental test, we calculated three ATs using different proposed methods: AT1-intensity corresponds to fixed [La]; AT2-minimum equivalent of the blood lactate-power output relationship plus 1.5 mmol·L-1; AT3-power output of the stage antecedent to the second lactate increase of at least 0.5 mmol·L-1 above the previous values, where the second increase was greater than the first. The MLSSint was determined for each participant as the highest power output that could be maintained with [La] fluctuating less than 1 mmol·L-1 during the final 20 min of the steady-state tests. ANOVA with repeated measures was used to compare physiological variables in the different methods. The relationship between the MLSSint and the power output of AT1, AT2, and AT3 was analysed using Pearson product-moment correlation coefficients. In addition, we calculated the bias and limits of agreement between the three different methods with actual MLSSint. The mean±s values of power output related to MLSSint, AT1, AT2, and AT3 were 247±33 W, 258±39 W, 248±35 W, and 230±36 W, respectively. The results showed that AT3 underestimated (P <0.05) the MLSSint for most of the participants and provided lower mean values compared with AT1 and AT2. Furthermore, AT2 seems to be more accurate to estimate MLSSint than other methods here verified when we analysed the mean values, correlation coefficient (r = 0.94), and Bland-Altman limits of agreement (± 9.5%). The AT1 also provided good prediction values, although it presented with a trend to overestimate MLSSint. Therefore, considering the methods analysed in the current study and the importance of this submaximal aerobic index to flat time-trials and prolonged uphill cycling performance, the AT2 method could be used with good accuracy by coaches and athletes.
© Copyright 2012 European Journal of Sport Science. Wiley. Alle Rechte vorbehalten.