Predicting maximal lactate steady state from lactate thresholds determined using methods based on an incremental exercise test in beagle dogs: A study using univariate and multivariate approaches.

The reliability of four lactate threshold (LT) methods to estimate the maximal lactate steady state (MLSS), defined as the highest intensity that can be maintained without plasma lactate ([La]) accumulation over time, was determined in Beagle dogs. Six male Beagle dogs performed a standardized incremental exercise test on a treadmill when plasma lactate ([La]) measurements were performed. The LTs for predicting MLSS, were determined by visual inspection (LT), using a bi-segmented linear regression model (LT), or using a polynomial function on the [La]/velocity ratio (LT) by considering the vertices of the curve and calculating the point that yields the maximal distance from a curve representing [La] as a function of velocity to the line formed by the two endpoints of the curve (LT method). The agreement was assessed using Bland-Altman plots and ordinary least products (OLP) regression among the velocities corresponding to the LTs identified using different methods (VLT, VLT, VLT, and VLT) and the velocity corresponding to the MLSS (VMLSS). A principal component (PC) analysis approach was performed to detect the degree of co-relatedness among the variables. The mean ± SD [La] at MLSS was 1.03 ± 0.24 mM. VMLSS had a lower mean bias with VLT, followed by VLT. The VLT underestimated MLSS. VLT and VLT had the lowest limits of agreement with the VMLSS. The VLT and VLT showed relatively high limits of agreement with MLSS. VLT, VLTBI, and VMLSS had more collinearity and were dominantly aligned with the second component (PC2). VLT and VLT can be used as simple methods to objectively determine aerobic fitness in Beagle dogs.