The first (VT1) and second ventilator (VT2) (anaerobic) thresholds are used to individually prescribe exercise training programs. The purpose of this research was to analyze inter- and intraobserver reliabilities of determining VT1 and VT2 in subjects with lower limb amputation (LLA) and able-bodied (AB) subjects during a peak exercise test on the arm-leg (Cruiser) ergometer. Previously published data of exercise tests on the Cruiser ergometer of subjects with LLA (n = 17) and AB subjects (n = 30) were analyzed twice by two observers. The VT1 and VT2 were determined based on ventilation plots. Differences in determining the VT1 and VT2 between the observers for the first and second analyses were analyzed. To quantify variation in measurement a variance component analysis was performed. Bland-Altmann plots were made, and limits of agreement were calculated. The number of observations in which thresholds could not be determined differed significantly between observers and analysis. Variation in VT1 between and within observers was small (0-1.6%) compared with the total variation, for both the subjects with an LLA and AB subjects. The reliability coefficient for VT1 was more than 0.75, and the limits of agreement were good. In conclusion, based on the results of this study on a population level, VT1 can be used to prescribe exercise training programs after an LLA. In the current study, the determination of VT2 was less reliable than VT1. More research is needed into the clinical application of VT1 and VT2 during a peak exercise test on the Cruiser ergometer.