Functional electrical stimulation (FES) of the peroneal nerve has long been proposed to improve ankle dorsiflexion during swing phase of gait in hemiparesis. The present study assesses the effect of a knee joint angle-based adaptive FES compared with classical FES and no use of FES on active ankle dorsiflexion during gait. The rationale behind aFES is to amplify dorsiflexor stimulation when it is most needed, to face increased plantar flexor cocontraction as gastrocnemius muscles are stretched by knee re-extension in late swing.

Three subjects with chronic hemiparesis (time since stroke, 4±3 years) underwent gait analysis at comfortable speed with shoes in three conditions: without stimulation (control), with classical FES (cFES) using trapezoidal stimulus intensity pattern and with an adaptive FES (aFES), in which stimulus intensity is indexed on the degree of knee flexion (the more re-extended the knee in swing, the higher the intensity). We compared cFES and aFES on the gain in active ankle dorsiflexion from control condition, during three periods of swing phase (T1 [0–42.5%], T2 [42.5–65%], T3 [65–100%], 10 gait cycles analyzed), and on the stimulation intensity delivered.

For the three patients pooled, the gain in active dorsiflexion from control condition was +4±2° at T1, +3±1° at T2 and +5±5° at T3 with cFES; and +2±1° at T1, +3±1° at T2 and +5±3° at T3 with aFES. Stimulation intensity was 12±10% lower with aFES compared with cFES.

Both classical and adaptive FES increase ankle dorsiflexion during swing phase in hemiparesis. These first case studies suggest higher effect of classical FES compared with adaptive FES in early swing, and similar effects of both FES systems during mid and late swing, while aFES used lower stimulation intensity.