Leg surface electromyography patterns in children with neuro-orthopedic disorders walking on a treadmill unassisted and assisted by a robot with and without encouragement

Background: Robot-assisted gait training and treadmill training can complement conventional physical therapy in children with neuro-orthopedic movement disorders. The aim of this study was to investigate surface electromyography (sEMG) activity patterns during robot-assisted gait training (with and without motivating instructions from a therapist) and unassisted treadmill walking and to compare these with physiological sEMG patterns. Methods: Nine children with motor impairments and eight healthy children walked in various conditions: (a) on a treadmill in the driven gait orthosis Lokomat (R), (b) same condition, with additional motivational instructions from a therapist, and (c) on the treadmill without assistance. sEMG recordings were made of the tibialis anterior, gastrocnemius lateralis, vastus medialis, and biceps femoris muscles. Differences in sEMG amplitudes between the three conditions were analyzed for the duration of stance and swing phase (for each group and muscle separately) using non-parametric tests. Spearman's correlation coefficients illustrated similarity of muscle activation patterns between conditions, between groups, and with published reference trajectories. Results: The relative duration of stance and swing phase differed between patients and controls, and between driven gait orthosis conditions and treadmill walking. While sEMG amplitudes were higher when being encouraged by a therapist compared to robot-assisted gait training without instructions (0.008 <= p-value <= 0.015), muscle activation patterns were highly comparable (0.648 <= Spearman correlation coefficients <= 0.969). In general, comparisons of the sEMG patterns with published reference data of over-ground walking revealed that walking in the driven gait orthosis could induce more physiological muscle activation patterns compared to unsupported treadmill walking. Conclusions: Our results suggest that robotic-assisted gait training with therapeutic encouragement could appropriately increase muscle activity. Robotic-assisted gait training in general could induce physiological muscle activation patterns, which might indicate that this training exploits restorative rather than compensatory mechanisms.