Novel hemiparetic gait characteristics composing peak paretic propulsion during overground walking

To elucidate the gait characteristics composing Pp during overground walking in patients with stroke.

Seventy-seven community-dwelling patients with stroke (mean: 64.8 years, mean months from onset: 52.7 months) underwent three-dimensional gait analysis (12 optical-infrared cameras with 100 Hz and two force-plates with 1000 Hz, MA-3000, Anima corp.) in overground with comfortable velocity (mean:58.6cm/s) for at least four strides without orthosis. We computed the gait characteristics with the emphasis on ankle joint parameters considering spasticity and severe hemiparesis. Paretic TLA, ankle plantarflexion moment and power, shank to vertical angle (SVA), hip extension angle, and knee flexion angle when peak Pp were calculated. Multiple regression demonstrated the relationship between Pp and the gait characteristics. The regression models were evaluated based on Akaike Information criteria (AIC), mean squared error (MSE), and determination coefficient (R²) by five-fold cross-validation. Python and R software were used for all the statistical analysis.

The model comprised of paretic SVA (β:0.37, p0.001), ankle plantarflexion power (β:0.31, p0.001) and knee flexion angle (β:-0.17, p0.001) when peak Pp was significantly associated with Pp (AIC:359, MSE:7.2, R²:0.76, p0.001). The conventional model comprised of TLA (β:0.52, p0.001) and ankle plantarflexion moment was significantly related to Pp (AIC:374, MSE:10.7, R²:0.65, p0.001); however, ankle plantarflexion moment was not a significant parameter (β:-0.12, p=0.700).

The paretic ankle concentrically plantarflexion with anterior inclination of the shank and knee extension was a key determinant of Pp in overground hemiplegic gait. Treadmill can help expand TLA, thus, Pp during treadmill gait is impacted by the ability to support trailing limb more posteriorly with accommodating the treadmill, whereas walking overground, hemiparesis patients need to propel center of mass by strengthening push-off. Furthermore, we may need to focus not only on TLA but also on whether the shank slants anteriorly because patients with stroke often expand TLA without shank forward tilt in overground hemiplegic gait.

Novel gait characteristics related to Pp during overground walking were rich in knowledge about rehabilitation treatment strategies. Paretic ankle plantarflexion power showed the importance of concentric movement in addition to augmentation of ankle plantarflexion moment, and SVA, which was one of the components of TLA, was informative in terms of providing a larger effect on Pp than hip extension. The acquisition of skill that is push-off while slanting shank anteriorly without knee flexion will ameliorate hemiparetic gait performance.