Metabolic responses to 4 different body weight-supported locomotor training approaches in persons with incomplete spinal cord injury.

OBJECTIVE: To describe metabolic responses accompanying 4 different locomotor training (LT) approaches. DESIGN: Single-blind, randomized controlled trial. SETTING: Rehabilitation research laboratory, academic medical center. PARTICIPANTS: Individuals (N=62) with minimal walking function due to chronic motor-incomplete spinal cord injury. INTERVENTION: Participants trained 5 days/week for 12 weeks. Groups were treadmill-based LT with manual assistance (TM), transcutaneous electrical stimulation (TS), and a driven gait orthosis (DGO) and overground (OG) LT with electrical stimulation. MAIN OUTCOME MEASURES: Oxygen uptake (V˙o2), walking velocity and economy, and substrate utilization during subject-selected "slow," "moderate," and "maximal" walking speeds. RESULTS: V˙o2 did not increase from pretraining to posttraining for DGO (.00 ± .18L/min, P=.923). Increases in the other groups depended on walking speed, ranging from .01 ± .18 m/s (P=.860) for TM (slow speed) to .20 ± .29 m/s (P=.017) for TS (maximal speed). All groups increased velocity but to varying degrees (DGO, .01 ± .18 Ln[m/s], P=.829; TM, .07 ± .29 Ln[m/s], P=.371; TS, .33 ± .45 Ln[m/s], P=.013; OG, .52 ±.61 Ln[m/s], P=.007). Changes in walking economy were marginal for DGO and TM (.01 ± .20 Ln[L/m], P=.926, and .00 ± .42 Ln[L/m], P=.981) but significant for TS and OG (.26 ± .33 Ln[L/m], P=.014, and .44 ± .62 Ln[L/m], P=.025). Many participants reached respiratory exchange ratios ≥ 1 at any speed, rendering it impossible to statistically discern differences in substrate utilization. However, after training, fewer participants reached this ceiling for each speed (slow: 9 vs 6, n=32; moderate: 12 vs 8, n=29; and maximal 15 vs 13, n=28). CONCLUSIONS: DGO and TM walking training was less effective in increasing V˙o2 and velocity across participant-selected walking speeds, while TS and OG training was more effective in improving these parameters and also walking economy. Therefore, the latter 2 approaches hold greater promise for improving clinically relevant outcomes such as enhanced endurance, functionality, or in-home/community ambulation.