THE LOADING RESPONSE PHASE OF THE GAIT CYCLE IS IMPORTANT TO KNEE OSTEOARTHRITIS

Abstract

The pathogenesis of knee osteoarthritis is complex and involves many correlated factors that can be measured with gait analysis. Important biomechanical factors may lie in the interrelationships between variables. This study demonstrated the use of a multidimensional gait data analysis technique that simultaneously considered multiple time varying and constant measures. The gait patterns of normal and knee osteoarthritic subjects were successfully separated with a misclassification error rate of < 6%. One of the most discriminatory features identified an important knee osteoarthritis difference during the loading response phase of the gait cycle.

The objective of this study was to detect biomechanical factors of knee osteoarthritis with a multidimensional gait data analysis technique.

A multidimensional gait data analysis technique detected a very discriminatory feature that described a knee osteoarthritis difference during the loading response phase of the gait cycle.

The combination of variables involved in the loading response feature may be important to the onset and development of knee osteoarthritis.

Discriminatory gait features associated with knee osteoarthritis were identified with a misclassification error rate of < 6%. In a very discriminatory feature, the loading response phase of the gait cycle was completely isolated as important. Body mass index (BMI) was the greatest contributing factor to the loading response feature.

Three-dimensional gait analysis was performed on fifty elderly patients with severe knee osteoarthritis and sixty-three elderly asymptomatic subjects. Three components of knee joint angles, moments and forces were calculated. Body mass index (BMI), radiographic measures and stride characteristics were also measured. A multivariate statistical technique extracted important features from the data and a discrimination procedure defined the optimal separation between the two groups.

The importance of loading response had been hypothesized previously, and this study quantitatively identified a very discriminatory gait pattern difference during loading response. The difference described was multidimensional. Although BMI was the largest contributing factor, there was no univariate difference in BMI between the two groups.

Funding: NSERC