Spastic cocontraction of plantar flexors during swing phase of gait in chronic hemiparesis

Abstract

In spastic hemiparesis, quantify agonist recruitment of tibialis anterior and triceps surae spastic cocontraction during swing phase of gait. Thirty-seven subjects with chronic hemiparesis (49 ± 14 years, mean ± SD; time since lesion, 8 ± 7 years) performed a 3D gait analysis, barefoot at comfortable speed (10 optoelectronic cameras, 6 dynamometric force plates, analysis of 9 cycles), with bilateral electromyography of tibialis anterior (TA), soleus (SO) and gastrocnemius medialis (GM). Speed, step length and maximal active dorsiflexion during swing were measured. Indices of agonist recruitment, IRATA, and spastic cocontraction, ICCSO and ICCGM, were calculated over 3 periods of swing phase (T1, [0–33%]; T2, [34–66%]; T3, [67–100%]), 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 100 ms around the maximal agonist isometric activity (measured on standing position by a maximal effort against resistance). IRA and ICC were compared between the 3 periods and between paretic and non paretic sides using repeated measures MANOVA (significant effect, P < 0.05). Speed, 0.68 ± 0.26 m/s; paretic step length, 0.48 ± 0.12 m, non paretic, 0.43 ± 0.15 m; paretic maximal active dorsiflexion, −3 ± 8°, non paretic, 6 ± 4°. On the paretic side: IRATA decreased between T1 and T2 (T1, 0.69 ± 0.59; T2, 0.48 ± 0.39, P = 2E-4) and then remained unchanged at T3 (0.47 ± 0.45; vs T1, P = 1E-4; vs T2, ns), while ICCSO and ICCGM increased at T3 only (ICCSO, T2, 0.25 ± 0.21; T3, 0.57 ± 0.50; P = 5E-8; ICCGM, T2, 0.39 ± 0.40; T3, 0.59 ± 0.50; P = 2E-4). On the non-paretic side, IRATA, ICCSO and ICCGM remained unchanged (IRATA, T1, 0.32 ± 0.20; T2, 0.23 ± 0.13; T3, 0.25 ± 0.10, ns; ICCSO, T1, 0.12 ± 0.12; T2, 0.13 ± 0.12; T3, 0.25 ± 0.18 ns; ICCGM, T1, 0.17 ± 0.31; T2, 0.19 ± 0.27; T3, 0.26 ± 0.25 ns). Side xperiod effect, P = 0.04. During comfortable gait in chronic hemiparesis, there is a decrease in tibialis anterior recruitment from mid-swing phase on and an increase in soleus and gastrocnemius spastic cocontraction at late swing were found.