BACKGROUND: Powered knee and ankle prostheses generate mechanical energy to replicate biological leg function, yet direct comparisons with more commonly prescribed microprocessor-controlled knee prostheses are limited by methodological variability and small sample sizes. METHODS: Twelve adults with transfemoral or knee disarticulation limb loss or difference who regularly used microprocessor knees were enrolled in a crossover study. After device fitting and an average of 10 training sessions with a powered knee and ankle prototype, participants completed a comprehensive battery of biomechanical, physiological, and functional assessments. Tests were repeated using each participant's prescribed prosthesis following a two-month washout period to reduce carryover effects. Paired statistical comparisons were used to evaluate differences across key outcome measures. RESULTS: Participants walked slightly slower and exhibited marginally reduced swing time symmetry when using the powered prosthesis. However, there were no significant differences in metabolic energy cost between devices, despite the significant increase in weight of the powered prosthesis. This may suggest that the powered leg provided sufficient assistance to compensate for its added weight, but did not yield improvements in walking speed, gait symmetry, or functional endurance. CONCLUSIONS: Use of this powered knee and ankle prosthesis resulted in comparable physiological effort and postural control to microprocessor knees, but did not improve outcomes like walking speed, symmetry, or endurance. Continued development of comparative testing protocols for prototype componentry could represents a critical step in optimizing PKA design and validation for improved clinical outcomes. Trial registration This study is registered at ClinicalTrials.gov (Identifier: NCT03204513; Date: 06/27/2017).
Embry et al. (2026) studied this question.