Gait quality is improved by locomotor training in individuals with SCI regardless of training approach

BACKGROUND: While various body weight supported locomotor training (BWSLT) approaches are reported in the literature for individuals with spinal cord injury (SCI), none have evaluated outcomes in terms of gait quality. The purpose of this study was to compare changes in measures of gait quality associated with four different BWSLT approaches in individuals with chronic motor-incomplete SCI, and to identify how gait parameters differed from those of non-disabled (ND) individuals. METHODS: Data were analyzed from 51 subjects with SCI who had been randomized into one of four BWSLT groups: treadmill with manual assistance (TM), treadmill with electrical stimulation (TS), overground with electrical stimulation (OG), treadmill with locomotor robot (LR). Subjects with SCI performed a 10-meter kinematic walk test before and after 12 weeks of training. Ten ND subjects performed the test under three conditions: walking at preferred speed, at speed comparable to subjects with SCI, and with a walker at comparable speed. Six kinematic gait quality parameters were calculated including: cadence, step length, stride length, symmetry index, intralimb coordination, and timing of knee extension. RESULTS: In subjects with SCI, all training approaches were associated with improvements in gait quality. After training, subjects with SCI walked at higher cadence and had longer step and stride lengths. No significant differences were found among training groups, however there was an interaction effect indicating that step and stride length improved least in the LR group. Compared to when walking at preferred speed, gait quality of ND subjects was significantly different when walking at speeds comparable to those of the subjects with SCI (both with and without a walker). Post training, gait quality measures of subjects with SCI were more similar to those of ND subjects. CONCLUSION: BWSLT leads to improvements in gait quality (values closer to ND subjects) regardless of training approach. We hypothesize that the smaller changes in the LR group were due to the passive settings used for the robotic device. Compared to walking at preferred speed, gait quality values of ND individuals walking at a slower speed and while using a walker were more similar to those of individuals with SCI.