Owen N Beck, Paolo Taboga, Alena M Grabowski

Running-specific prostheses enable athletes with lower limb amputations to run by emulating the spring-like function of biological legs. Current prosthetic stiffness and height recommendations aim to mitigate kinematic asymmetries for athletes with unilateral transtibial amputations. However, it's unclear how different prosthetic configurations influence the biomechanics and metabolic costs of running. Consequently, we investigated how prosthetic model, stiffness, and height affect the biomechanics and metabolic costs of running. Ten athletes with unilateral transtibial amputations each performed fifteen running trials at 2.5 or 3.0 m/s while we measured ground reaction forces and metabolic rates. Athletes ran using three different prosthetic models with five different stiffness category and height combinations per model. Use of an Ottobock 1E90 Sprinter prosthesis reduced metabolic cost by 4.3% and 3.4% compared to use of Freedom Innovations Catapult (fixed effect (β)=-0.177; p<0.001) and Össur Flex-Run (β=-0.139; p=0.002) prostheses, respectively. Neither prosthetic stiffness (p≥0.180) nor height (p=0.062) affected the metabolic cost of running. The metabolic cost of running was related to lower peak (β=0.649; p=0.001) and stance average (β=0.772; p=0.018) vertical ground reaction forces, prolonged ground contact times (β=-4.349; p=0.012), and decreased leg stiffness (β=0.071; p<0.001) averaged from both legs. Metabolic cost was reduced with more symmetric peak vertical ground reaction forces (β=0.007; p=0.003), but was unrelated to symmetric stride kinematics (p≥0.636). Therefore, prosthetic recommendations based on stride kinematics do not necessarily minimize the metabolic cost of running. Instead, an optimal prosthetic model, which improves overall biomechanics, minimizes the metabolic cost of running for athletes with unilateral transtibial amputations.