Use of an activity tracker as a measurement tool in a knee osteoarthritis clinical trial (active-oa trial)

Abstract

The aim of this study was to test if ambulatory load can be systematically altered using dynamic unloading and additional load carrying without clinically relevant changes in gait kinematics. Methods: Twenty-four healthy young subjects (12 females, 12 males; age, 25.4 ± 1.8 years; height, 1.74 ± 0.11 m; body mass, 70.2 ± 13.7 kg; body mass index, 22.9 ± 2.6 kg/m 2) walked on an instrumented treadmill with an integrated pressure plate (zebris Medical GmbH, Germany) at a self-selected consistent walking speed (4.6 ± 0.5 km/h) under three loading conditions (80%, 100% and 120% body weight (BW)). BW was reduced by a dynamic pneumatic unloading system (h/p/cosmos airwalk, h/p/cosmos sports & medical GmbH, Germany) set to 80% BW. BW was increased by wearing an adjustable weight vest with weights corresponding to 20% body mass placed in equally distributed vest pockets on the front and back of the vest. Overall ground reaction force (GRF) and GRF in the heel, midfoot and forefoot regions during stance were calculated using the manufacturer software (zebris Medical GmbH). Sagittal plane kinematics of the ankle, knee and hip were measured with seven inertial sensors (RehaGait ® , Hasomed GmbH, Germany) placed on the sacrum, thigh, shank and foot. GRF and joint angles were exported for further analysis in Matlab (MathWorks, Natick, MA, USA). Differences in GRF and joint angles of 80%BW and 120%BW compared to the 100%BW condition were analyzed using paired t-tests with statistical parametric mapping (SPM) and a significance level of 0.025. The average difference in GRF and absolute joint angles of 80%BW and 120%BW compared to the 100%BW condition during the entire stance phase were computed for each subject. Only differences greater than 5 were deemed clinically relevant. Results: Step length, step time and cadence were comparable for all conditions (P>0.7). The GRF of the 80%BW and 120%BW differed significantly from the GRF of the 100%BW condition throughout stance (Fig. 1). The average difference in GRF between the 80%BW condition and the 100%BW condition across the entire stance phase was-20.0 ± 6.8% (mean ± 1SD; P<0.001). The average difference in GRF between the 120%BW condition and the 100%BW condition across the entire stance phase was 16.8 ± 3.9% (P<0.