Predictive Neuromechanical Simulation Explains Gait Biomechanics in Obesity

Abstract

Individuals with obesity exhibit gait adaptations including reduced early-stance knee flexion, altered muscle coordination, slower preferred walking speeds, and shorter step lengths. Although these features are well documented, the mechanisms by which obesity-related physiological changes produce these patterns and influence knee joint loading relevant to osteoarthritis (OA) remain unclear. This study used predictive neuromechanical simulation to examine how musculoskeletal changes and movement objectives interact to generate obesity-associated gait patterns and tibiofemoral loading. Predictive simulations were performed using a reflex-based neuromechanical walking model. A baseline non-obese model (1.8 m, 80 kg) was modified to represent obesity-related changes in segment mass distribution and muscle strength (1.8 m, 140 kg), including more apple-like and more pear-like body mass distributions. Control parameters were optimized to generate stable walking while minimizing muscle effort and tibiofemoral joint loading. Objective weightings were identified by matching simulated knee kinematics to experimental observations at a typical walking speed. Using the selected weightings, we compared joint kinematics, kinetics, and muscle activations, and simulations were performed across walking speeds to evaluate optimal walking speed, step length, muscle effort, and knee loading. The baseline model best matched reference knee kinematics using a muscle-effort objective alone, whereas the obese model required a combined objective penalizing both muscle effort and knee loading. This formulation reproduced key gait features, including reduced early-stance knee flexion, reduced vastii activation with increased plantarflexor activation, slower optimal walking speeds, and shorter step lengths. Variations in body mass distribution produced moderate but consistent effects on gait mechanics relative to larger effects of increased body mass. Obesity-related changes in body mass and muscle strength alone did not reproduce observed gait patterns, but incorporating an objective that penalizes knee loading generated multiple characteristic features. Predictive neuromechanical simulation provides a framework for identifying candidate mechanisms linking obesity, gait biomechanics, and knee joint loading. Author SummaryUnderstanding how and why obesity alters gait is a complex biomechanical problem involving multiple interacting factors including increased segmental mass, altered inertial properties, and reduced relative muscle strength. These factors interact in ways that are difficult to isolate through experimental observation alone. Here, we used computer simulations to examine how musculoskeletal changes and movement objectives interact to generate obesity-associated gait patterns and knee loading. We found that physiological changes alone did not reproduce observed gait features, whereas incorporating an objective that penalizes knee loading generated multiple characteristic features simultaneously, including reduced early-stance knee flexion, altered muscle coordination, slower optimal walking speeds, and shorter step lengths. These findings suggest that obesity-associated gait reflects coordination strategies that regulate knee loading under increased body mass.