Similar sensitivity of time to exhaustion and time-trial time to changes in endurance

Purpose: There is widespread misunderstanding about the ability of constant-power tests to quantify changes in endurance performance. We have therefore compared the sensitivity of a constant-power test with that of a time trial for the effects of arterial oxygenation on endurance performance. Methods: Eight cyclists performed three constant-power rides to exhaustion and three 5-km time trials on a cycle ergometer in conditions of normoxia, hypoxia, and hyperoxia. After logarithmic transformation of performance times, sensitivity was calculated as the mean change in time divided by the error of measurement derived from the standard deviation of change scores. Results: In normoxia, performance times were 488 +/- 77 s for the constant-power test and 454 +/- 16 s (mean +/- SD) for the time trial. The mean and standard deviation of the change in performance time from normoxia to hypoxia were much larger in the constant-power test (-45% +/- 13%) than in the time trial (5.7% +/- 1.6%); there was a similar disparity in the change from normoxia to hyperoxia (123% +/- 37% and -4.1% +/- 1.4%, respectively). However, sensitivity for the normoxia-hypoxia change in performance in the constant-power test (6.3, 90% confidence interval 4.3-11.4) was similar to that in the time trial (4.5, 3.0-8.2); sensitivities were also similar for the normoxia-hyperoxia changes (3.2, 2.1-6.0; 3.8, 2.5-6.9, respectively) P values for mean performance changes (range, 0.0002-0.000002) reflected these sensitivities. Conclusions: Time to exhaustion has sensitivity similar to that of time-trial time for the effects of arterial oxygenation and presumably other factors affecting endurance performance. Sensitivity need not be a concern when using constant-power tests to quantify changes in endurance performance.