Variability of Knee Adduction Moment During Gait

Abstract

Gait analysis is often performed in order to determine the effect of an injury or a treatment. For example, Munderman et al. (2004) showed a 23% increase in knee adduction moment in subjects with severe osteoarthritis relative to those with less severe disease. However, it is not known if this increased moment is normalized following total knee arthroplasty (TKA). In order to determine this effectively it is necessary to know the measurement variability between repeated tests. Not only will this assist in developing a study, but this type of statistical power analysis is increasingly required for publication. PURPOSE: The purpose of this study was to determine the variability of knee adduction moment. METHODS: Kinetic and kinematic data were recorded from 10 healthy subjects (age range 25–45 years) on two separate days as they walked across a force plate at their preferred walking speed. Motion data were recorded at 60 Hz using a 5-camera system (Qualisys). Ground reaction forces (GRF) were recorded from a force plate (Kistler) at 1000 Hz. Using an inverse dynamics model, we calculated knee adduction moment; the two peaks in knee adduction moment (normalized and expressed as percent body mass times height, BM*ht) were analyzed. Paired t-tests and Pearson product-moment correlations were used to compare the adductor moment peaks from the two different days. RESULTS: The mean of the first adduction moment peak (over both days) was 4.33 ± 1.17 with a mean difference of −0.16 ± 0.65 between days. The mean second peak was 3.66 ± 1.03 with a mean difference of-0.22 ±0.79 between days. Paired t-tests showed that the two adduction moment peaks were not different between days (P = 0.448 and P = 0.399). Additionally, the two moment peaks were significantly correlated from one day to the next (R = 0.856, P= 0.002 and R= 0.770, P= 0.009). CONCLUSION: Based on the mean and standard deviation of the difference in knee adduction moment between days, a 10% change in adduction moment could be detected with a sample of 18 patients, while a 15% change could be detected with a sample of 9 patients (P <0.05, 80% power). This information may be useful for designing studies to examine changes in knee adduction moment with progression of osteoarthritis or treatment with total knee arthroplasty.