RESUMEN
Walking patterns is modified during load carriage, resulting in an increased activation of lower limb muscles and energy expenditure. Negative effects of load carriage could be minimized by wearing an exoskeleton, but evidence on the effects are conflicting. The objectives of this study were to describe the influence of an adjustable, passive load-bearing exoskeleton on the metabolic cost of walking (MCW) and associated muscle activations, and to explore changes in MCW after a familiarization process. Thirteen participants walked on a treadmill with a 22.75 kg payload at six preselected speeds (from 0.67 to 1.56 m/s) under three walking conditions: 1) without exoskeleton (NoExo); 2) with exoskeleton before familiarization (ExoPre); and 3) with exoskeleton after familiarization (ExoPost). Metabolic data was normalized to walking speed to provide MCW. Multi-muscle surface electromyography (EMG) was time and amplitude normalized to the gait cycle to provide muscle activation patterns. The familiarization occurred over three weeks including exposure to the exoskeleton. Differences in MCW and muscle activations were compared using a nonparametric analysis of longitudinal data. There were statistically significant increases in MCW for all speeds in the ExoPre and ExoPost conditions compared the NoExo. The average muscle activation showed an increase during ExoPre and ExoPost for the three speeds evaluated. Post-hoc analysis showed no significant effect of the familiarization period on metabolic data. In conclusion, a first exposure to the adjustable exoskeleton increased MCW and muscle activations, but the familiarization process did not provide any benefits toward a reduction in MCW or reduction in muscle activations at all speeds evaluated.
RESUMEN
OBJECTIVE: Human gait requires complex somatosensory processing of various inputs such as proprioception. Proprioception can be altered in the presence of pain. This has been shown mostly during controlled tasks, thereby limiting the influence of external perturbations. While controlling the environment is sometimes warranted, it limits the ecological validity of the data. Using robotic orthoses to apply perturbations during movements seems a promising tool to functionally assess proprioception, where the complex somatosensory processing required in real-life situations is at play. The main objective of this study was to compare the proprioceptive threshold of healthy participants during gait in the presence and absence of an acute experimental pain. METHODS: 36 healthy participants walked on a treadmill while wearing a robotized ankle-foot orthosis (rAFO) around their right ankle. The rAFO applied torque perturbations of graded magnitudes during the swing phase of gait. Participants had to report the presence/absence of such perturbations, as a measure of proprioceptive threshold. Following initial assessment, they were randomly assigned to one of three experimental groups: Control (no stimulation), Painless (non-nociceptive stimulation) and Painful (nociceptive stimulation). Electrodes placed on the right lateral malleolus delivered an electrical stimulation during the second assessment for Painless and Painful groups. A Kruskal-Wallis was used to compare the percentage of change of the three groups between the two assessments. RESULTS: A 31.80±32.94% increase in proprioceptive threshold, representing an increase of 1.3±1.2 Nm in the detection threshold, was observed for the Painful group only (p<0.005), with an effect size of 1.6. CONCLUSION: Findings show that the presence of pain at the ankle can alter participants' proprioceptive threshold during gait. Clinical assessment of proprioception should therefore carefully consider the presence of pain when evaluating a patient's performance using clinical proprioceptive test and consider the negative effect of pain on proprioceptive threshold for test interpretation.
