Abstract
The purpose of this study was to determine relative as well as absolute reproducibility for the IsoMed 2000-dynamometer in measuring Peak Torque (PT) during concentric and eccentric maximum knee extension at two angular velocities. Thirty five physically active male subjects (mean age: 23.7 years) with no prior experience of isokinetic exercise were measured on three occasions (T1–T3), separated by 48–72 h. Detection of systematic errors was performed using repeated measures analysis of variance (ANOVA) with Bonferroni post hoc adjustments. Further statistical calculations were done between consecutive pairs of sessions (T1–T2; T2–T3). Intraclass correlation coefficients (ICC 2,1) were calculated to assess relative reproducibility, whereas absolute reproducibility was determined by the standard error of measurement (SEM). A highly significant systematic error was found for the eccentric measurements with values on the first visit being around 5% lower than corresponding values on the second and third visit. ICC values of 0.907–0.955 and 0.968–0.984 with corresponding values of SEM of 7.4–19.1 and 4.5–10.9 Nm were found for T1–T2 and T2–T3, respectively. In conclusion, high reproducibility was shown for the IsoMed 2000-dynamometer in measuring PT during concentric and eccentric knee extension. However, due to a systematic practice-based improvement found for eccentric measurements, we suggest to do a familiarisation session prior the actual testing sessions in practice.