During rehabilitation of hemiparetic patients, physiotherapists face a discrepancy between motor skills during analytical exercises and functional abilities observed during gait. A muscle recruitment difference in the two situations could explain this gap. Among pathologic motricity manifestations observed in hemiparetic patients (Gracies 2005), exist the spastic co contraction phenomenon that presently concerned our work. Indeed, the main objective of this study was to compare co-contractions in hemiparetic subjects between an analytical exercise modality and a functional exercise modality, i.e. gait. Our hypothesis was that the co contraction is increased during gait as compared to an analytical exercise and could consequently lead to a decrease of gait ability in patients.

Muscular activity of knee flexors and extensors was recorded by surface EMG in analytic exercises and during the swing phase of gait, in a group of hemiparetic subjects (SH) and a group of healthy subjects (SS). Co-contraction index in duration and in magnitude was calculated from EMG data and allowed us the comparison between the 2 exercise modalities.

Twelve SH and 11 SS were included. ICC differed between analytical and functional modalities in SH (P<0.05).

Discussion: the observed results show a significant difference between co-contractions observed analytically and during gait in SH. These results validate the initial hypothesis since in most cases, co-contractions are increased during gait. If the conditions of muscle recruitment differ in analytic conditions and during gait, it seems unsuited to propose analytical exercises with the aim to improve gait parameters. This finding therefore suggests a more functional rehabilitation, according to the concept of task-oriented training: rehabilitate gait by gait.