Feasibility study of an insole-type active assist device for ankle alignment correction during stepping in patients with knee osteoarthritis

This feasibility study aimed to examine whether an insole-type active assist device designed to dynamically adjust ankle alignment at heel contact can be safely delivered and evaluated during an on-the-spot stepping task in patients with medial knee osteoarthritis (OA). The study specifically assessed the feasibility of intervention delivery, testing procedures, and motion-capture-based outcome measurement. Six ambulatory patients with medial knee OA (Kellgren–Lawrence grade II–III) performed repeated on-the-spot stepping trials under two conditions: Active (device control enabled) and Inactive (device control disabled). The assist device tilts the heel toward eversion in response to detected ankle inversion at heel contact. Feasibility outcomes included participant recruitment and completion, safe execution of the stepping task, device activation during trials, successful acquisition and analysis of motion capture data, and occurrence of adverse events. Lateral knee thrust was quantified descriptively using a three-dimensional motion capture system to characterize measurement variability and inform future study design. All participants provided informed consent and completed the stepping protocol (6/6, 100%), with no adverse events observed. The stepping task and testing procedures were safely performed in all cases. Motion capture data were successfully acquired and analyzed for all trials (90/90, 100%). The assistive mechanism was activated in at least one stepping trial in five of six participants (83%), with activation occurring in 39 of 90 stepping trials (43%). Across conditions, lateral knee thrust values showed substantial inter-individual and condition-related variability, ranging approximately from 30 to 110 mm across participants. This study demonstrates the feasibility and safety of delivering an insole-type active assist intervention and conducting motion-capture-based evaluations during an on-the-spot stepping task in patients with medial knee OA. The observed variability in lateral knee thrust highlights important considerations for outcome selection and sample size planning, supporting progression to future adequately powered studies to evaluate clinical and biomechanical effectiveness.