Relationships between muscle activity and anterior-posterior ground reaction forces in hemiparetic walking

The objectives of this study were to further our understanding of the neuromotor mechanisms responsible for disrupted anterior-posterior (A/P) ground reaction force (GRF) patterns in hemiparetic walking by examining relationships between muscle activity and GRFs collected from 49 post-stroke patients walking at their self-selected speed. Data were analyzed within four regions (bins) of the gait cycle defined by events (first double support, first and second halves of single support, and second double support) that approximately correspond to the braking (negative A/P GRF) and propulsive (positive A/P GRF) phases of the normal gait cycle. Relationships were examined across the entire population of subjects and for subgroups based on the Brunnstrom classification for hemiparetic severity. Correlation analyses were used to identify the relationships between percent of muscle activity and net A/P GRF impulse (i.e., the time integral of the A/P GRF) within each bin by determining the corresponding Pearson's correlation coefficients. Few correlations were found between the non-paretic GRF impulses and muscle activity. However, significant correlations for the paretic leg provided insight into both direct and indirect influences of muscle activity on the propulsive (positive) and braking (negative) impulses. Medial gastrocnemius and soleus muscle activity positively correlated with paretic propulsion in late stance and pre-swing across all subjects and for subjects grouped by functional status, suggesting that paretic propulsion is strongly associated with increased plantar flexor activity. Negative correlations for paretic tibialis anterior and propulsion during pre-swing were found across all subjects and for subjects in the moderate and severe functional groups. Furthermore, paretic rectus femoris and tibialis anterior activity negatively correlated with the pre-swing propulsion for the severe group. Therefore, exaggerated muscle flexor activity may counteract the effects of the plantar flexors by offloading the leg and interfering with the limb's ability to generate the appropriate A/P GRFs. As a group, subjects did not exhibit abnormal vasti activity in early stance, which has been shown to be one of the primary muscle contributors to the braking GRF. These results suggest that other factors such as altered gait mechanics may be responsible for increased paretic braking. Future studies will incorporate foot-pelvic mechanics to further elucidate the complex relationships between these variables and their contribution to disrupted hemiparetic GRFs.