In hemiparesis, functional electrical stimulation (FES) of the peroneal nerve aims to compensate for ankle dorsiflexor paresis in swing phase. The present prospective study compared the effects of gait training with FES vs. conventional therapy on plantar flexor overactivity.

Twenty subjects with chronic hemiparesis (6±4years post-lesion; mean±SD) were randomized into two groups: FES (45min/day of gait self-training using FES) and Control (CON, 3×45min/week of conventional physiotherapy) for 10 weeks. Outcomes at Day 1 and Week 10 included comfortable speed barefoot gait analysis with gastrocnemius medialis (GM) and soleus (SO) electromyography, yielding:, walking speed; and, at the paretic ankle: maximal passive dorsiflexion during stance, maximal active dorsiflexion during swing, velocity of active dorsiflexion over early swing, coefficient of spastic cocontraction in GM and SO over the three thirds of swing (CSC, calculated by the ratio of the RMS of the electromyogram in the period of interest over the RMS of the electromyogram of the same muscle over 100ms around its maximal voluntary contraction). Intra- and inter-group comparison used rank-ANOVAs.

No difference was observed in walking speed changes (both groups pooled; D1, 0.73±0.25m/s, W10, 0.80±0.30m/s, ns) and in passive and active dorsiflexion amplitudes (ns). However, the velocity of active dorsiflexion increased in FES while it decreased in Controls (D1 vs. W10, FES, +5±2°/s, P=0.02; CON, −4±1°/s, P=0.04). In parallel, CSC_GM in early swing tended to improve in FES only (FES, −41±23%, P=0.09; CON, −24±89%, ns; between group difference ns).

In chronic hemiparesis, FES, which focuses on agonist dorsiflexor stimulation during gait improves active dorsiflexion velocity in swing and may also be associated with reduction of plantar flexor spastic cocontraction. Further studies are required to confirm these findings and evaluate whether reciprocal inhibition toward plantar flexors may be restored by FES.