Ankle and midfoot kinetics during normal gait: A multi-segment approach

Abstract

Multi-segment foot models are increasingly being used to evaluate intra and inter-segment foot kinematics such as the motion between the hindfoot/tibia (ankle) and the forefoot/hindfoot (midfoot) during walking. However, kinetic analysis have been mainly restricted to one-segment foot models and could be improved by considering a multi-segment approach. Therefore, the aims of this study were to (1) implement a kinetic analysis of the ankle and theoretical midfoot joints using the existing Oxford Foot Model (OFM) through a standard inverse dynamics approach using only marker, force plate and anthropometric data and (2) to compare OFM ankle joint kinetics to those output by the one-segment foot plugin-gait model (PIG). 10 healthy adolescents fitted with both the OFM and PIG markers performed barefoot comfortable speed walking trials over an instrumented walkway. The maximum ankle power generation was significantly reduced by approximately 40% through OFM calculations compared to PIG estimates (p<0.001). This result was not caused by a decrease in OFM computed joint moments, but by a reduction in the angular velocity between the tibia/hindfoot (OFM) compared to the tibia/foot (PIG) (p<0.001). Additionally, analysis revealed considerable midfoot loading. One-segment foot models overestimate ankle power, and may also overestimate the contribution of the triceps surae. A multi-segment approach may help quantify the important contribution of the midfoot ligaments and musculature to power generation. We therefore recommend the use of multi-segment foot models to estimate ankle and midfoot kinetics, especially when surgical decision-making is based on the results of three-dimensional gait analysis.

Introduction

The foot and ankle complex is an intricate structure made up of several independent segments. Thus, single-segment foot models are insufficient to reveal intra and inter-segment foot kinematic changes during gait (Leardini et al., 2007, Stebbins et al., 2006) and cannot isolate foot pathologies to a specific joint (MacWilliams et al., 2003). In response to these problems, multi-segment foot models (MSFM) have been developed. Fifteen separate models representing the foot and distal leg using between four and nine rigid segments based on either electromagnetic or retro-reflective optical marker data have been described (Deschamps et al., 2011). Of these, the Oxford Foot Model (OFM) (Stebbins et al., 2006) is commonly used for gait analysis in healthy and pathologic subjects (Deschamps et al., 2011), has strong reliability in adults (Wright et al., 2011) and children (Curtis et al., 2009), is freely downloadable for Vicon systems (Vicon, Oxford Metrics, UK), and is relatively simple to interpret. However, the OFM does not compute ankle and foot kinetics. Kinetics play a significant role in the identification, evaluation and treatment of gait abnormalities (Davis, 1997, Gage and Novacheck, 2001). Further, since the ankle provides the main propulsive power during gait (Rose and Gamble, 2006), a better understanding of the kinetics of this joint is crucial.

Few researchers have described MSFM kinetics (Abuzzahab et al., 1997, MacWilliams et al., 2003, Scott and Winter, 1991, Scott and Winter, 1993). Early models by Scott and Winter, 1991, Scott and Winter, 1993 described kinetics for an 8-segment foot, but only presented results for a few subjects. A four-segment model was later proposed (Abuzzahab et al., 1997), but is overly complex for routine gait analysis and is scarce in detail. Later, MacWilliams et al. (2003) implemented a MSFM (nine-segments) used in conjunction with force plate and pedographic data. This model represents the most complete kinetic analysis of the foot to date, but requires numerous surface markers, additional instrumentation, and time intensive processing of pressure data limiting its application scope to research.

An easily implemented MSFM kinetic analysis has not yet been developed. Therefore, the aims of this study were to describe a simple approach to kinetic analysis of the foot and ankle using the existing OFM and to compare OFM ankle kinetics to estimations based on a standard one-segment foot model. We hypothesised that (1) peak OFM ankle dorsiflexion would be decreased compared to a one-segment model since the relative movement of the forefoot and hindfoot are isolated from that of the ankle (Stebbins et al., 2006); (2) peak sagittal plane moment and thus power would be reduced using the OFM since single rigid foot models may overestimate the contribution of the ankle joint (MacWilliams et al., 2003), and (3) non-negligible power generation in the midfoot would occur since muscle and tendon activity are present at this location during gait (Saraswat et al., 2010) and that MacWilliams et al. (2003) previously reported maximum power generation of approximately 0.7 W/kg at the midfoot.