Modifications of the Dmax method in comparison to the maximal lactate steady state in young male athletes

Objective: To investigate the influence of different approaches for first-rise determination on the accuracy of Dmax as an estimate of the maximal lactate steady state (MLSS). Methods: Seventeen male cyclists and 18 male runners with different levels of endurance performance completed graded exercise tests either on a cycle ergometer or treadmill to determine Dmax, calculated by the final data point and five modifications of the first rise in blood lactate concentration. Two or more constant load tests over 30 min were performed to determine MLSS. Differences between the modifications of the first rise in blood lactate concentration as well as the corresponding Dmax variants and MLSS were tested, using one-way repeated measure ANOVA with Bonferroni post-hoc tests, and illustrated, using the Bland-Altman method. The absolute agreement was observed, using intra-class correlation coefficients, based on a single measure, absolute agreement, 2-way mixed effects model. Results: The peak power output/running velocity of the groups averaged 275 +/- 43 W and 4.3 +/- 0.4 m center dot s(-1), respectively. The mean power output/running velocity at MLSS was 229 +/- 38 W and 3.77 +/- 0.38 m center dot s(-1). For both running and cycling the original Dmax described by Cheng et al. was significantly lower than MLSS (p < 0.01). All modifications showed good agreement with MLSS (ICC >= 0.75). According to the Bland-Altman method the mean differences of the modifications compared to MLSS in cycling ranged from -7 (43) to 2 (41) W. In running the mean differences ranged from -0.12 (0.34) to -0.08 (0.35) m center dot s(-1). Conclusion: We suggest using the first rise in blood lactate concentration for calculating Dmax instead of the first data point of a lactate curve as originally described. The approach of first rise determination has no substantial influence on the accuracy of Dmax compared to MLSS in cycling and running.