New technology allows cyclists to train via power output (PO) in addition to heart rate (HR). For those athletes undertaking seasonal laboratory testing (e.g., VO2, lactate threshold), it is imperative that athletes be able to directly apply this information to their training device. We examined the reliability of a standardized laboratory ergometer (Lode Excalibur Sport) and its applicability to an electromagnetically braked ergometer (Computrainer) in 2 phases. Phase I (n = 12) examined the reliability of the Lode. Phase II (n = 14) compared the Lode to the Computrainer using a randomized, counterbalance assignment. Following warm-up, each trial started at 100 W, progressing 50 W every 3 minutes to exhaustion. Outcomes were time-to-exhaustion (TTE), peak PO (W) (POpeak), peak HR (HRpeak), and ventilatory (VT) and respiratory compensation (RCP) thresholds. We used a repeated measures analysis of variance (ANOVA), Tukey post hoc analysis, regression analysis, Bland-Altman plots, and coefficient of variation (CV) analysis for each variable. During phase I, we found no significant difference for any variable, minimal dispersion of VO2 during Bland-Altman analysis, and a low CV at each test stage (<= 5%). During phase II, significant differences and higher CV for most parameters (all data; p < 0.001) were observed for Lode versus Computrainer: TTE (21 minutes, 12 seconds +/- 3 minutes, 12 seconds vs. 19 minutes, 9 seconds +/- 2 minutes, 36 seconds; CV 5 16%), POpeak (335 +/- 57.8 W vs. 295 +/- 47.1 W, CV = 17%), as well as PO at VT (CV = 51%) and RCP (CV = 24%; p < 0.01). We conclude that coaches and cyclists may need to use some caution when directly transferring results obtained from laboratory testing to the Computrainer training device.
(C) 2005 National Strength and Conditioning Association