Owen N Beck, Paolo Taboga, Alena M Grabowski

Inspired by the spring-like action of biological legs, running-specific prostheses are designed to enable athletes with lower-limb amputations to run. Yet, manufacturer recommendations for prosthetic stiffness and height may not optimize running performance. Therefore, we investigated the effects of using different prosthetic configurations on the metabolic cost and biomechanics of running. Five athletes with bilateral transtibial amputations each performed fifteen trials on a force-measuring treadmill at 2.5 or 3.0 m/s. Athletes ran using each of three different prosthetic models (Freedom Catapult FX6, Össur Flex-Run, and Ottobock 1E90 Sprinter) with five combinations of stiffness categories (manufacturer recommended and ± 1) and heights (International Paralympic Committee's maximum competition height and ± 2 cm) while we measured metabolic rates and ground reaction forces. Overall, prosthetic stiffness (fixed effect (β)=0.036; p=0.008) but not height (p≥0.089) affected the net metabolic cost of transport; less stiff prostheses reduced metabolic cost. While controlling for prosthetic stiffness (kN/m), using the Flex-Run (β=-0.139; p=0.044) and 1E90 Sprinter prostheses (β=-0.176; p=0.009) reduced net metabolic costs by 4.3% to 4.9% compared to using the Catapult prostheses, respectively. The metabolic cost of running improved when athletes used prosthetic configurations that decreased peak horizontal braking ground reaction forces (β=2.786; p=0.001), stride frequencies (β=0.911; p<0.001), and leg stiffness values (β=0.053; p=0.009). Remarkably, athletes did not maintain overall leg stiffness across prosthetic stiffness conditions. Rather, the in-series prosthetic stiffness governed overall leg stiffness. The metabolic cost of running in athletes with bilateral transtibial amputations is influenced by prosthetic model and stiffness, but not height.