Joint contact forces when minimizing the external knee adduction moment by gait modification: A computer simulation study

• Relation between knee adduction moment (KAM) and joint contact forces (JCF) poorly understood
• Used musculoskeletal modeling to simulate gaits that minimized the KAM
• Minimizing KAM reduced peak medial JCF by 15% when cost function excluded energy expenditure
• Minimizing KAM reduced peak medial JCF by 27% when cost function included energy expenditure
• Gait modification may benefit from considering whole body motion and muscle activity

Background and purposeGait modification is often used to reduce the external knee adduction moment (KAM) in human walking, but the relationship between KAM reduction and changes in medial knee joint contact force (JCF) is not well established. Our purpose was to examine the limiting case of KAM-based gait modification: reducing the KAM as much as possible, and the resulting effects on JCF.MethodsWe used musculoskeletal modeling to perform three optimal control simulations: normal walking, a modified gait that reduced the KAM as much as theoretically possible (Min(KAM) simulation), and a second modified gait that minimized the KAM plus the metabolic cost of transport (Min(KAM+CoT) simulation).ResultsThe two modified gaits both reduced the peak KAM from normal walking (− 82% for Min(KAM) simulation, − 74% for Min(KAM+CoT) simulation) by increasing trunk lean, toe-out, and step width, and reducing knee flexion. Even though the Min(KAM+CoT) simulation had the larger KAM, it had a greater reduction in peak medial JCF (− 27%) than the Min(KAM) simulation (− 15%) because it reduced the KAM using less knee muscle activity. These results were qualitatively robust to a sensitivity analysis of the knee joint model, but the magnitude of changes varied by an order of magnitude.ConclusionsThe results suggest that (i) gait modification can benefit from considering whole-body motion rather than single adjustments, (ii) accurate interpretation of KAM effects on medial JCF requires consideration of muscle forces, and (iii) subject-specific knee models are needed to accurately determine the magnitude of KAM reduction effects on JCF.