gms | German Medical Science

Objectives: Achilles tendon rupture affects an increasing number of people every year. Common post-operative treatment protocols rely on the use of controlled ankle motion (CAM) walker boots, which are designed to immobilize the ankle and reduce the loading of the Achilles tendon. The major limitation of such protocol is that it is difficult to predict whether the boot provides optimal stimuli. Structural changes in tendon tissues are driven by mechanical stimuli (i.e., strains and forces). These stimuli determine the tissue's ability to develop, maintain integrity, or heal following injury. Stimuli above optimal levels would cause tissue damage, such as lengthening or re-rupture, whereas stimuli below would lead to atrophy and hinder recovery. In this study, we examined how a CAM walker boot affected the Achilles tendon loading during walking.

Methods: We present a non-invasive in silico method to estimate the Achilles tendon loading. Three healthy participants were instructed to walk ten meters on a level surface under four experimental conditions: 1. wearing indoor sports shoes, 2. wearing a CAM walker boot at 0° plantarflexion, 3. wearing a CAM walker boot at 15° plantarflexion, and 4. wearing a CAM walker boot at 30° plantarflexion. The boot was worn on the right foot, and an indoor sports shoe was worn on the contralateral foot. Kinematic data, ground reaction forces, and electromyographic signals from lower limb muscles, including the triceps surae, were recorded and used to estimate the Achilles tendon force. Using open source biomechanical software (OpenSim), we created a custom musculoskeletal model for each participant. Customization included scaling and adjusting the model's geometries and inertial properties, such that, during forward dynamics, the model tracked the experimental kinematic data with minimal residual forces and moments at the pelvis. The customized model was used to estimate individual muscle forces, including the Achilles tendon force, using a combination of static optimization and proportional-derivative control.

Results and conclusion: We established a simulation pipeline to estimate the effects of plantarflexion angles on lower limb kinematics and kinetics. As expected, the peak Achilles tendon force decreased with increased plantarflexion in the CAM walker boot in all studied participants. There was a significant difference between normal gait and walking with 30° plantarflexion (mean ± standard deviation: 2881 ± 320 N vs. 1448 ± 493 N, p=.0062). This simulation approach offers new possibilities of predicting the Achilles tendon force during rehabilitation after a tendon rupture. Since in vitro assessments indicate very low limit for safe loading of Achilles tendon sutures (approximately 200 N), further investigation with this non-invasive method may clarify how to optimize the Achilles tendon loading after surgery.