Reliability of Assessing Indices of Isokinetic Leg Strength in Pubertal Soccer Players

Using 95% limits of agreement (LOA), this study examined the reproducibility of indices of isokinetic leg strength in pubertal soccer players. Following familiarization, 23 boys performed isokinetic assessments of the knee joint muscles at 1.08, 2.16, and 4.32 rad/s during concentric exercise and at 2.16 rad/s during eccentric actions on two separate occasions. Peak torque values were recorded and reciprocal and bilateral muscle group ratios were calculated. As random error was heteroscedastic, 95% ratio LOA were employed. Systematic bias for the various measurements ranged from 0.97 to 1.07 for all data. Random variation associated with eccentric assessments of the hamstrings, concentric H:Q ratio calculated at 1.08 rad/s, and H_ECC:Q_CON ratios appeared narrow enough to be of practical use (×/÷ 1.05 to 1.20). Random errors associated with H:Q ratios calculated at high concentric velocities (4.32 rad/s) and during eccentric exercise (×/÷ 1.10 to1.37), and in all bilateral muscle group ratios (×/÷ 1.15 to 1.43) were too broad, rendering them prone to misinterpretation.