For high-intensity cycle ergometer exercise, the tolerable duration (t) is well characterized as a hyperbolic function of power output, P : t = W'/(P-thetaF), where thetaF may be termed the "fatigue threshold." The purpose of this study was to determine the effect of oral creatine (Cr) supplementation on the curvature constant parameter (W') of the power-duration curve. A double-blind research method and a cross-over design were employed for creatine/placebo supplementation. Eight healthy male subjects (aged 18 to 22 years) each performed four or five high-intensity square-wave exercise bouts on an electrically braked cycle ergometer after 5 d of Cr monohydrate (CR: 20 g of Cr with artificial sweetener/d) or placebo (PL: 6 g of glucose/d) supplementation. Each subject performed a single high-intensity exercise trial per day for four or five successive days to determination the P-t hyperbolic relation. After 6 weeks (the washout time of Cr from the muscles), each subject performed the other condition (i.e., PL or CR) and repeated the same experimental procedure. There was no significant difference for thetaF between PL and CR conditions (PL: 214.4 +/- 23.6, CR: 207.0 +/- 19.8 W, mean +/- SD). In contrast, W' was significantly increased by the Cr supplementation (PL: 10.9 +/- 2.7, CR: 13.7 +/- 3.0 kJ; p<0.05). The results indicated that Cr and/or PCr content in muscles seems to be one of the important determinants of the curvature constant parameter (W') of the power-duration hyperbolic curve for cycle ergometry.