MEASURING FATIGUE RELATIVE TO PEAK POWER OUTPUT DURING HIGH-INTENSITYCYCLE SPRINTING

The purpose of this report was to introduce and validate a method to a ccount for variable fatigue rates during cycle sprinting by incorporat ing decline from peak power (30%) as a test termination criterion. Fif teen healthy men performed three maximal sprint tests separated by 20 min (Bouts 1 and 2) and 48 hours (Bout 3). Power curves were analyzed for peak lower; time to peak, time to fatigue (decline of 30% from pea k), total work, total test time, fatigue rate, and the fatigue index. High test-retest reliability was demonstrated for all variables (R = 0 .85-0.98). No significant differences were detected between variables (Bouts 1, 2 and 3; p > 0.01). Fatigue rates varied widely among partic ipants (range = 24.7-65.4 W.s(-1)), but the fatigue index was consiste nt (31.1 +/- 0.16; mean +/- standard error of measurement, range = 30. 1-32.5%). Our data show that variable fatigue responses among particip ants are normalized to the percentage of decline from peak power by in corporating a relative fatigue cutoff criterion. This approach might i nform mechanisms relating to short-term fatigue and recovery during co nsecutive sprint bouts.