Reliability Of Isokinetic Knee Strength Measurements Using Generalizability Theory

Abstract

Isokinetic strength test is the measurement of force-velocity or torque angle. Isokinetic strength of the knee extensor and flexor muscles is often assessed using an isokinetic dynamometer such as Cybex system. Question remains concerning the reliability of the isokinetic muscle strength measurements. PURPOSE: The purpose of this study was to establish the minimum number of trials required to assess reliable estimates of isokinetic strength measurements using Cybex system. METHODS: A total of 70 adults (mean age 19.38 ± 3.32 years; 35 males and females each) completed the isokinetic strength tests using Cybex system. Concentric muscle strengths were measured at 60 degrees per second over 4 repetitions. Following data screening, participants’ data were finally analyzed by muscle strength measurements (knee extensor and flexor muscles) and gender using the Generalizability theory. A single-facet crossed design [i.e., Participant (P) x trial (T)] was used. G-study was performed to quantify the percentage of variance associated with each facet and interaction in the model. Follow-up D-study was performed to determine the optimal combinations of number of repetition trials to achieve a desirable reliability coefficient (G ≥ .80). RESULTS: The knee extensor muscle of male and female were 158.48Nm ± 33.89Nm and 96.78Nm ± 23.60Nm, respectively. The knee flexor muscle of male and female were 83.52Nm ± 31.92Nm and 45.97Nm ± 11.94Nm, respectively. For the knee extensor muscle, the results from the G-study show that P accounted for 71% and 90% variance for male and female. T had no effect on the total variance. The number of trials to achieve G ≥ .80 for male and female were 2 and 1, respectively. For the knee flexor muscle, the G-study results show that P accounted for 94% and 77% variance for male and female. T had little effect on the total variance ( < 1%). The number of trials to achieve G ≥ .80 for male and female were 1 and 2, respectively. CONCLUSIONS: In general, the minimum of 2 trials of monitoring is sufficient to achieve the desired reliability level of .80 for both knee extensor and flexor muscle measurements. Our findings regarding the optimal measurement conditions can be of benefit for researchers or practitioners interested in measuring isokinetic knee strengths using Cybex system.