Associations between gait and clinical parameters in patients with severe knee osteoarthritis: A multiple correspondence analysis

BackgroundGiven the complexity of the gait of patients with knee osteoarthritis, a multiple correspondence analysis may be helpful to optimise the extraction of relevant gait and clinical information. Therefore, the aims of this study are to identify the main associations with clinical and gait biomechanical parameters and to evaluate whether there are more specific knee osteoarthritis groups with different gait profiles.MethodsNinety patients with severe knee osteoarthritis and twenty-six healthy individuals participated in this study. Pain and function were assessed with the WOMAC Index; knee joint deformity was assessed by the hip-knee-ankle angle on full-limb radiography; and full body gait analysis was performed with a motion analysis system and force plates.FindingsUsing multiple correspondence analysis, two categories of gait parameters that best explain the gait variance of patients with knee osteoarthritis were highlighted. The forward displacement category is composed of the parameters speed, stride length, hip flexion and knee flexion. The frontal category is composed of the parameters thorax obliquity and knee adductor moments. Moreover, based on these parameters, four distinct gait profiles were identified: two gait profiles were associated with knee varus deformities, increased thorax obliquity and different forward displacements, while two gait profiles were associated with valgus deformities and different forward displacements.InterpretationThese gait parameters can be used to simplify the characterisation of the gait of the knee osteoarthritis population. Patients in varus profiles increase thorax obliquity on the stance limb and may reduce forward displacement. Patients in valgus profiles, however, only reduce forward displacement.