Changes in gait biomechanics are associated with long-term clinically important improvements in patient-reported outcomes

Abstract

Purpose: Aberrant gait biomechanics are consistently shown to be associated with knee osteoarthritis (OA) progression and are therefore the targets of numerous interventions. However, the ability to produce sustained long-term improvements in gait biomechanics remains elusive, and the potential effect of improvements in gait on long-term patient-reported outcomes is unclear. The purpose of this study was to investigate the 5-year change in gait biomechanics after limb realignment surgery in order to test the hypothesis that these long-term changes are associated with long-term clinically important improvements in patient-reported outcomes. Methods: Three-dimensional gait analysis, full-limb standing radiographs and Knee injury and Osteoarthritis Outcome Scores (KOOS) were evaluated before and 5-years after medial opening wedge high tibial osteotomy (HTO). The changes from baseline to 5 years were calculated and patients were categorized as having a clinically important improvement in the KOOS4 (≥10 points) or not. Predictors of clinically important change (dependent variable) were then evaluated using multivariate logistic regression. We tested two models. The first model included the 5-year change in mechanical axis angle and BMI, while controlling for baseline age, sex, disease severity (KLG) and KOOS4 (independent variables). The second model added the 5-year change in peak external knee adduction and flexion moments and gait speed as independent variables. Receiver operating characteristic (ROC) curves assessed the ability of each model (i.e. with and without gait biomechanics) to correctly discriminate between patients achieving a long-term clinically important improvement or not. The area under the curve (AUC) and 95% confidence intervals were calculated for each model. Results: 170 patients (135 males, 35 females) with medial knee OA participated (age: 46±9 years; BMI: 29.5±4.9 kg/m2; KLG: 1 = 10, 2 = 46, 3 = 64, 4 = 50). ROC curves are shown in Figure 1. The AUC was 0.67 (95%CI: 0.59-0.76; p<0.001) for the first model (i.e. without changes in gait biomechanics), and increased to 0.74 (95%CI: 0.66-0.81, p<0.001) for the second model (i.e. with changes in gait biomechanics). Conclusions: This is the first study we are aware of to show that long-term changes in gait biomechanics are significantly associated with long-term clinically important improvements in patient-reported outcomes. Moreover, assessing changes in gait biomechanics improves the ability to discriminate between patients with medial knee OA achieving a long-term clinically important improvement or not.