ABSTRACT
Abstract Aims: To determine lactate threshold (LT) by three different methods (visual inspection, algorithmic adjustment, and Dmax) during an incremental protocol performed in the leg press 45° and to evaluate correlation and agreement among these different methods. Methods: Twenty male long-distance runners participated in this study. Firstly, participants performed the dynamic force tests in one-repetition maximum (1RM). In the next session, completed an incremental protocol consisted of progressive stages of 1 min or 20 repetitions with increments of 10, 20, 25, 30, 35, and 40% 1RM. From 40% 1RM, increments corresponding to 10% 1RM were performed until a load in which the participants could not complete the 20 repetitions. A rest interval of 2 min was observed between each stage for blood collection and adjustment of the workloads for the next stage. Results: Our results showed no significant difference in relative load (% 1RM), good correlations, and high intraclass correlation coefficients (ICC) between algorithmic adjustment and Dmax (p = 0.680, r = 0.92; ICC = 0.959), algorithmic adjustment and visual inspection (p = 0.266, r = 0.91; ICC = 0.948), and Dmax and visual inspection (p = 1.000, r = 0.88; ICC = 0.940). In addition, the Bland-Altman plot and linear regression showed agreement between algorithmic adjustment and Dmax (r2 = 0.855), algorithmic adjustment and visual inspection (r2 = 0.834), and Dmax and visual inspection (r2 = 0.781). Conclusion: The good correlation and high agreement among three methods suggest their applicability to determine LT during an incremental protocol performed in the leg press 45°. However, the best agreement found between mathematical methods suggests better accuracy.