Lateral wedges induce a pattern of changes in ambulatory lower-limb frontal-plane kinetics with individual dose-response

Purpose: Ambulatory mechanics is a critical factor in the development of knee osteoarthritis (OA) and modifying gait has the potential to become an important therapeutic intervention for this disease. Lateral wedges (LW) inserted in patient shoes have been shown to generally reduce the peak knee adduction moment (pKAM), a variable positively associated with progression of medial compartment knee OA. However, despite decrease of the pKAM, inconsistent clinical improvements were reported in medial knee OA patients. It is possible that these inconsistent outcomes are related to the large variations in pKAM changes observed among patients wearing the same LW. Therefore, there is a need to better understand the effects of LW on lower-limb kinetics during gait as this could, for example, allow identifying a subtype of patients with higher probability to benefit from LW and provide new information on how to select a LW (e.g., tilt angle) for a particular patient. This study aimed to determine if the lower-limb kinetic response to LW is totally different among individuals or if individuals follow the same pattern of changes, but with varying dose-response. Specifically, this study tested the hypothesis that the changes in frontal-plane moments at the hip, knee and ankle joints induced by a LW are correlated in a group of healthy individuals. Methods: Twelve healthy subjects took part in this IRB-approved study after providing informed consent (5 males; 24±3 years old; 23±3 kg/mˆ2). Participants walked multiple trials with two footwear conditions: i) control shoes with neutral frontal-plane stability (ASICS, JP), and ii) standard foot-long 5 deg LW inserted in the control shoes. Joint moments were calculated using a validated biomechanical model (BioMove, CA) based on data recorded by a motion capture system (Vicon, UK) and a forceplate (Kistler, CH). The peak hip adduction moment (pHAM), peak knee adduction moment (pKAM), and peak ankle eversion moment (pAEM) during the first half of stance phase were extracted for each gait cycle and averaged per subject and footwear condition. Paired t-tests were performed to compare gait variables between footwear conditions and Pearson correlation coefficients were used to assess the relationship among variables, considering a significance level set at 5%. Results: Walking with LW significantly reduced the pKAM by 0.31±0.29 %BW*Ht (p=0.004) and increased the pAEM by 0.19±0.17 %BW*Ht (p=0.002) compared to walking with the control shoe alone. Changes in frontal-plane moments induced by the LW were correlated among joints: changes in pKAM were negatively correlated with changes in pAEM (Rˆ2=0.40; p=0.03) and positively correlated with changes in pHAM (Rˆ2=0.85; p<0.001), whereas pAEM changes were negatively correlated with pHAM changes (Rˆ2=0.45; p=0.02). Conversely, the changes in lower-limb moments were not correlated with the moments measured with the control shoe alone. Subjects walked at the same average speed of 1.45 m/s with both footwear conditions (p=0.6). Conclusions: The hypothesis was partially supported as changes in frontal-plane kinetics at the knee and hip were strongly correlated together and changes at the knee and hip were moderately correlated with changes at the ankle. The results thus highlighted a general response pattern to LW, with varying dose-response among individuals. While additional studies will be necessary to understand the mechanism behind the range of “amplitudes of response”, the finding of a response pattern is important because it indicated that frontal-plane kinetic changes mainly vary by a factor among individuals. This finding, strengthened by the absence of correlation between the changes in kinetics due to the LW and the kinetics measured without LW, provides a basis for future research testing for associations between the amplitude of response in frontal-plane kinetics and long term clinical improvements in medial knee OA patients. Better understanding these associations could help identifying patients and LW characteristics most likely to improve clinical outcomes. Interestingly, a general response pattern could not be identified with footwear condition comparisons as the LW either decreased or increased the pHAM depending on the individual amplitude of response to the LW. This observation stressed the need for global analyses of the ambulatory function and motivates future studies to determine if the response pattern in this study extends to other OA-related gait variables, such as the knee flexion moment.