Abstract
We investigated the effects of different sequences of concentric isokinetic test speeds on knee extensor muscle peak torque production, and attempted to validate the findings of previous studies suggesting that slow test speeds should precede fast test speeds. Under a sequential repeated-measures paradigm, which included a test-multiple retest design, 72 subjects (39 males, 33 females; ages 17–28 yr) were randomly assigned to one of six groups for the testing of a randomly selected extremity on a Cybex 340 dynamometer system. Each test group was subjected to a different order of slow (60 deg/sec), intermediate (180 deg/sec), and fast (300 deg/sec) speeds in a sequential manner. Testing occurred on six different days, each separated by 48 h, for each group and consisted of a standardized cardiovascular and isokinetic warm-up followed by peak torque data collection across five maximal efforts at each speed. Multifactorial analysis of variance showed significant differences (p < 0.001) between speeds, but not between sessions or the combined factors of speeds and sessions across the experiment. A post hoc Scheffe test confirmed the differences between speeds (p < 0.01). Results of intraclass correlation coefficient tests showed a high level of reliability (0.81) across both test speeds and sessions. It was concluded that isokinetic test speed order did not affect concentric peak torque measurements, that distinct differences existed between the individual test speeds, and that the procedures afforded a high degree of muscle performance measurement consistency.