Effect of the lengthening of the protocol on the reliability of muscle fatigue indicators.

The aim of this study was to examine absolute and relative reliability of fatigue measures calculated from peak torque or total work during 20, 30, 40 and 50 reciprocal maximal concentric contractions performed on an isokinetic dynamometer at 180 degrees x s(-1). Eighteen moderately active men performed 50 reciprocal maximal concentric contractions on three occasions with one 7-10 days recovery between each session. Peak torque and total work were computed for each contraction and subsequently summed to compute cumulated performance after respectively 20, 30, 40 and 50 repetitions. Muscle fatigue was determined after 20, 30, 40 and 50 repetitions by the fatigue index, the percent decrease in performance and the slope. Reliability of average peak torque or average total work was similar and was not affected by the lengthening of the protocol, although a learning effect was evident for knee flexors. Reliability of fatigue measures calculated from peak torque or total work was similar, improved with the lengthening of the protocol and was better for knee extensors. Measuring average peak torque or average total work and the slope during a protocol involving 30 maximal reciprocal concentric contractions appear to represent a better compromise between reliability and physiological interpretability of the data.