Device use, locomotor training, and the presence of arm swing during treadmill walking post-spinal cord injury

OBJECTIVES: Determine the presence of walking-related arm swing following spinal cord injury (SCI), associated factors, and whether arm swing may change following locomotor training (LT). DESIGN: Observational, cross-sectional study from a convenience sample with pre-test/post-test from a sample subset. SETTING: Malcom Randall VAMC and University of Florida, Gainesville, FL. METHODS: Arm movement was assessed during treadmill stepping, pre-LT, in 30 individuals with motor incomplete SCI (iSCI, American Spinal Injury Association Impairment Scale grade C/D, as defined by the International Standards for Neurological Classifications of SCI, with neurological level of impairment at or below C4). Partial body weight support and manual-trainer assistance was provided, as needed, to achieve stepping and allow arm swing. Arm swing presence was compared based on cervical versus thoracic neurological levels of impairment and device type. Leg and arm strength and walking independence were compared between individuals with and without arm swing. Arm swing was re-evaluated post-LT in the 21/30 individuals that underwent LT. RESULTS: Of 30 individuals with iSCI, 12 demonstrated arm swing during treadmill stepping, pre-LT. Arm movement was associated with device type, lower extremity motor scores, and walking independence. Among the 21 individuals that received LT, only 5 demonstrated arm swing pre-LT. Of the 16 individuals lacking arm swing pre-LT, 8 integrated arm swing post-LT. CONCLUSION: Devices routinely used for walking post-iSCI appeared associated with arm swing. Post-LT, arm swing presence increased. Therefore, arm swing may be experience-dependent. Daily neuromuscular experiences provided to the arms may produce training effects, thereby altering arm swing expression.