Exploring the Change in Metabolic Cost of Walking before and after Familiarization with a Passive Load-Bearing Exoskeleton: A Case Series.

OCCUPATIONAL APPLICATIONSMilitary personnel are at greater risk of injuries due to frequent load carriage. Novel exoskeleton technology may have benefits for soldiers, such as reduced physical burden through load carriage support that may result in decreased metabolic cost, reduced fatigue, and lower risk of injuries during walking. However, as for most assistive devices, a familiarization period is likely necessary to obtain the full potential of the device. Our results show that the metabolic cost of walking (MWC) was initially increased significantly upon provision of the passive exoskeleton, though it returned to baseline values after a 9-day familiarization period. The exoskeleton remained effective after a three-month pause, with a MCW below baseline. These results suggest that to properly assess the assistance of an exoskeleton, a sufficient familiarization period should be mandatory.

Background: Military load carriage has been shown to alter gait patterns, resulting in an increased metabolic cost during walking (MCW). Soldiers' burden could be mitigated by wearing a passive exoskeleton, but the additional payload of the device can alter movement patterns during gait, rendering it detrimental. Integrating principles of motor learning during a familiarization period could allow users to develop adaptive motor strategies, thereby decreasing MCW.Purpose: The aim of this study was to explore the influence of a familiarization period on MCW when soldiers wear a passive, load-bearing, prototype exoskeleton (Exo).Methods: Three male soldiers walked on a treadmill with a 38 kg payload at eight speeds (1.8-6.0 km/h) under five conditions: 1) no exoskeleton (NoExo); 2) exoskeleton pre-familiarization (Exo_Pre); 3) exoskeleton post-familiarization (Exo_Post); 4) no exoskeleton follow-up (NoExo_FU); and 5) exoskeleton follow-up (Exo_FU). Each experimental trial consisted of 10 minutes of standing followed by 10 minutes of walking at a constant speed. Metabolic data were normalized to walking speed (J/kg·m) to obtain the MCW. The familiarization period consisted of 9 days of activities with the exoskeleton using a standardized protocol. Differences in MCW with and without the Exo were compared at the eight walking speeds using a nonparametric analysis of Longitudinal Data.Results: There was a statistically significant decrease in MCW after familiarization with the Exo, particularly during Exo_FU with a relative treatment effect of 0.11 − 0.19. There were also significant reductions in MCW during Exo_FU when compared to NoExo_FU [participant 01 = 0.37; participant 02 = 0.27; participant 03 = 0.35].Conclusions: A first exposure to the exoskeleton increased MCW. After familiarization, however, the MCW with the Exo returned to the NoExo level or below with a payload of 38 kg among three soldiers. A familiarization period of 3 hours per day over 2 weeks of familiarization may optimize the use of an exoskeleton.

Impact of online visual feedback on motor acquisition and retention when learning to reach in a force field.

When subjects learn a novel motor task, several sources of feedback (proprioceptive, visual or auditory) contribute to the performance. Over the past few years, several studies have investigated the role of visual feedback in motor learning, yet evidence remains conflicting. The aim of this study was therefore to investigate the role of online visual feedback (VFb) on the acquisition and retention stages of motor learning associated with training in a reaching task. Thirty healthy subjects made ballistic reaching movements with their dominant arm toward two targets, on 2 consecutive days using a robotized exoskeleton (KINARM). They were randomly assigned to a group with (VFb) or without (NoVFb) VFb of index position during movement. On day 1, the task was performed before (baseline) and during the application of a velocity-dependent resistive force field (adaptation). To assess retention, participants repeated the task with the force field on day 2. Motor learning was characterized by: (1) the final endpoint error (movement accuracy) and (2) the initial angle (iANG) of deviation (motor planning). Even though both groups showed motor adaptation, the NoVFb-group exhibited slower learning and higher final endpoint error than the VFb-group. In some condition, subjects trained without visual feedback used more curved initial trajectories to anticipate for the perturbation. This observation suggests that learning to reach targets in a velocity-dependent resistive force field is possible even when feedback is limited. However, the absence of VFb leads to different strategies that were only apparent when reaching toward the most challenging target.

Effects of rolling resistances on handrim kinetics during the performance of wheelies among manual wheelchair users with a spinal cord injury.

STUDY DESIGN: Repeated cross-sectional study. OBJECTIVES: To compare the effects of rolling resistances (RRs) on handrim kinetic intensity at the non-dominant upper limb and on handrim kinetic symmetry during wheelies performed by manual wheelchair users (MWUs) with spinal cord injury (SCI). SETTING: Pathokinesiology Laboratory. METHODS: Sixteen individuals with SCI who were able to perform wheelies participated in this study. During a laboratory assessment, participants randomly performed wheelies on four RRs: natural high-grade composite board, 5-cm thick soft foam, 5-cm thick memory foam, and with the rear wheels blocked by wooden blocks. Four trials were conducted for each of the RRs. Participant's wheelchair was equipped with instrumented wheels to record handrim kinetics, whereas the movements of the wheelchair were recorded with a motion analysis system. RESULTS: The net mean and peak total forces, including its tangential and mediolateral components, were greater during take-off compared with the other phases of the wheelie, independently of RR. During take-off, the greatest net mean and peak total and tangential forces were reached with the wheels blocked. Symmetrical tangential and mediolateral force intensities were applied at the dominant and non-dominant handrims. CONCLUSION: Wheelies performed on low or moderate density foam generate similar forces at the handrim than on a natural surface and significantly less forces than with the wheels blocked. Hence, when teaching individuals with an SCI to perform a stationary wheelie, the use of low or moderate density foam represents a valuable alternative for minimizing upper limb effort and may also optimize quasi-static postural steadiness.

Rapid changes in corticospinal excitability during force field adaptation of human walking.

Force field adaptation of locomotor muscle activity is one way of studying the ability of the motor control networks in the brain and spinal cord to adapt in a flexible way to changes in the environment. Here, we investigate whether the corticospinal tract is involved in this adaptation. We measured changes in motor-evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS) in the tibialis anterior (TA) muscle before, during, and after subjects adapted to a force field applied to the ankle joint during treadmill walking. When the force field assisted dorsiflexion during the swing phase of the step cycle, subjects adapted by decreasing TA EMG activity. In contrast, when the force field resisted dorsiflexion, they increased TA EMG activity. After the force field was removed, normal EMG activity gradually returned over the next 5 min of walking. TA MEPs elicited in the early swing phase of the step cycle were smaller during adaptation to the assistive force field and larger during adaptation to the resistive force field. When elicited 5 min after the force field was removed, MEPs returned to their original values. The changes in TA MEPs were larger than what could be explained by changes in background TA EMG activity. These effects seemed specific to walking, as similar changes in TA MEP were not seen when seated subjects were tested during static dorsiflexion. These observations suggest that the corticospinal tract contributes to the adaptation of walking to an external force field.

Contribution of cutaneous inputs from the hindpaw to the control of locomotion. II. Spinal cats.

The goal of these experiments was to define the contribution of hindpaw cutaneous inputs in the expression of spinal locomotion in cats. In 3 cats, some (n = 1) or all (n = 2) cutaneous nerves were cut bilaterally at ankle level before spinalization. This denervation caused small deficits that were gradually compensated as reported in the companion study. After spinalization, the completely denervated cats never recovered plantar foot placement or weight bearing of the hindquarters despite more than 35 days of treadmill training. Although normal electromyographic rhythmic activity developed at the hip and knee, ankle flexors and extensors were abnormally coactivated during stance. In contrast, the partially denervated cat regained foot placement and weight support 15 days after spinalization. However, after completing the denervation, foot placement and weight bearing were lost as in previous cats. In a 4th cat, spinalization was performed before denervation and the cutaneous nerves were cut sequentially in the right hindlimb only. Rapid locomotor adaptation occurred after cutting the deep peroneal, saphenous, and sural nerves. Later, cutting the superficial peroneal nerve produced paw drag, which was compensated within 8 days. On cutting the last cutaneous nerve (tibial), plantar foot placement was lost despite another 71 days of training. On the one hand, these experiments show that some cutaneous inputs are necessary for appropriate plantar foot placement and weight bearing of the hindquarters during spinal locomotion and, on the other hand, that locomotor compensation to partial cutaneous denervation after spinalization reveals important adaptive capacities of the spinal cord.

Contribution of cutaneous inputs from the hindpaw to the control of locomotion. I. Intact cats.

The goal of this study was to evaluate the role of hindpaw cutaneous feedback in the control of locomotion, by cutting some (in one cat) or all (in 2 cats) cutaneous nerves bilaterally at ankle level. Kinematic and electromyographic (EMG) recordings were obtained before and for several weeks after denervation during level and incline (15 degrees up and down) treadmill walking. Ladder walking and ground reaction forces were also documented sporadically. Early after the denervation (1-3 days), cats could not walk across a ladder, although deficits were small during level treadmill walking. Increased knee flexion velocity caused a 14% reduction in swing phase duration. EMG activity was consistently increased in knee, ankle, and toe flexors, and in at least one knee or ankle extensor. The adaptive changes during walking on the incline were much reduced after denervation. Ladder walking gradually recovered within 3-7 wk. By this time, level treadmill walking kinematics had completely returned to normal, but EMG activity in flexors remained above control. Incline walking improved but did not return to normal. Mediolateral ground reaction forces during overground walking were increased by 200%. It is concluded that in intact cats, cutaneous inputs contribute more to demanding situations such as walking on a ladder or on inclines than to level walking. Active adaptive mechanisms are likely involved given that the EMG locomotor pattern never returned to control level. The companion paper shows on the other hand that when the same cats are spinalized, these cutaneous inputs become critical for foot placement during locomotion.

Adaptive locomotor plasticity in chronic spinal cats after ankle extensors neurectomy.

After lateral gastrocnemius-soleus (LGS) nerve section in intact cats, a rapid locomotor compensation involving synergistic muscles occurs and is accompanied by spinal reflex changes. Only some of these changes are maintained after acute spinalization, indicating the involvement of descending pathways in functional recovery. Here, we address whether the development of these adaptive changes is dependent on descending pathways. The left LGS nerve was cut in three chronic spinal cats. Combined kinematics and electromyographic (EMG) recordings were obtained before and for 8 d after the neurectomy. An increased yield at the ankle was present early after neurectomy and, as in nonspinal cats, was gradually reduced within 8 d. Compensation involved transient changes in step cycle structure and a longer term increase in postcontact medial gastrocnemius (MG) EMG activity. Precontact MG EMG only increased in one of three cats. In a terminal experiment, the influence of group I afferents from MG and LGS on stance duration was measured in two cats. LGS effectiveness at increasing stance duration was largely decreased in both cats. MG effectiveness was only slightly changed: increased in one cat and decreased in another. In cat 3, the plantaris nerve was cut after LGS recovery. The recovery time courses from both neurectomies were similar (p > 0.8), suggesting that this spinal compensation is likely a generalizable adaptive strategy. From a functional perspective, the spinal cord therefore must be considered capable of adaptive locomotor plasticity after motor nerve lesions. This finding is of prime importance to the understanding of functional plasticity after spinal injury.

"Torso Rotation" experiments. 4: the role of vision and the cervico-ocular reflex in compensation for a deficient VOR.

Acute, reversible changes in human vestibular function can be produced by exposure to "Torso Rotation" (TR), a method involving the overuse of certain types of simple, self-generated movements. A single session results in multiple, short-lasting aftereffects, including perceptual illusions, VOR gain reduction, gaze and postural instability, and motion sickness. With repeated exposure, motion sickness susceptibility disappears and gaze stability improves. VOR gain continues to be reduced, however. Therefore, another gaze stabilizing system must come into play. Are visual and/or neck inputs involved in this functional compensation? Six subjects participated in this 7-day experiment. Eye and head movements were measured during 2 tests: 1) voluntary "head only" shaking between 0.3 and 3.0 Hz (lights off) and 2) voluntary "head and torso" shaking, moving the upper body en bloc (neck immobilized). Measurements were obtained before and repeatedly after TR. Velocity gain (eye velocity/head velocity) was determined for each of these tests. Each day, mean velocity gain during "head only" shaking in the dark (averaged over 1.0 to 2.0 Hz) dropped significantly after TR (P < 0.01), with no long-term improvement (P > 0.9). Similar results, although more noisy, were obtained for "head and torso" shaking. As a control, EOG calibration data confirmed that gaze stability in the light did improve over the 7 days of testing. This experiment demonstrates that the reduction in gaze instability following repeated exposure to TR results from an increased use of vision. It excludes the VOR, the COR, and predictive mechanisms (including efference copy) as contributors. In addition, in the 20 minutes following TR completion, gaze stability recovered less than during previous VOR testing in the dark. These results are compatible with the motion that exposure to TR leads to a change in sensorimotor strategy involving a de-emphasis of vestibular inputs.

"Torso rotation" experiments; 1: Adaptation to motion sickness does not correlate with changes in VOR gain.

Following a 30-minute exposure to an unusual motor strategy called "Torso Rotation" (TR), the signs and symptoms of motion sickness appear along with perceptual illusions during movement, gaze and postural instability, and a significantly reduced vestibulo-ocular response (VOR) gain. With repeated exposure to TR, the motion sickness disappears and gaze instability seems to be reduced. Is this apparent improvement in gaze stability associated with a reduction of the transient change in VOR gain? Motion sickness (subjective questionnaire) and VOR gain (passive step rotations in darkness) were measured before and repeatedly after TR on 7 consecutive days. Despite a complete loss of symptoms in 3 to 4 days, the transient, daily change in VOR gain remained unaffected. Furthermore, there was no increase in the use of compensatory saccades. It is concluded that adaptation to TR-induced motion sickness is not the result of a change in VOR's sensitivity to TR.

"Torso rotation" experiments; 2: Gaze stability during voluntary head movements improves with adaptation to motion sickness.

Following a 30-minute exposure to an unusual motor strategy called "Torso Rotation" (TR), the signs and symptoms of motion sickness appear along with perceptual illusions during movement, gaze and postural instability, and a significantly reduced vestibulo-ocular response (VOR) gain. With repeated exposure to TR, motion sickness symptoms disappear and gaze instability seems reduced, but without any concomitant change in VOR gain. Is the reduction in gaze instability a perceptual illusion or a real, measurable phenomenon? Velocity gain (eye velocity/ head velocity) was evaluated during voluntary head shaking in the light over the frequency range 0.3 to 3.0 Hz. A significant improvement was seen after 3 days of testing (P < 0.01). Furthermore, the time course of improvement in gaze stability was correlated with the loss of motion sickness symptoms reported in the previous study (1). We suggest that adaptation to motion sickness could be related to an overall change in sensori-motor strategy, perhaps including a de-emphasis of a vestibular reference.

"Torso rotation" experiments; 3: Effects of acute changes in vestibular function on the control of voluntary head movements.

"Torso Rotation" (TR) produces an acute, reversible change in human vestibular function. Experiments were performed to determine if repeated exposure to this technique would result in long-term adaptive modifications. In one experiment, VOR gain was evaluated. Measurements were obtained before and 3 times after 30 minutes of TR, on 7 consecutive days. VOR gain dropped each day after TR, returning to normal within about 20 min. In a separate experiment with different subjects, eyes-open gaze stability was measured during voluntary head shaking between 3.0 and 0.3 Hz. The same test schedule was used. Analysis of gaze stability (limited to frequencies between 1.0 and 2.0 Hz) was complicated by an unexpected finding. Despite careful instructions, head displacement increased each day after TR, also returning to normal within about 20 min. Surprisingly, subjects were unaware of this change. Combining the 2 experiments, VOR gain and head amplitude were averaged across all 7 days, separately for the 4 daily tests. Head amplitude was plotted against VOR gain for these 4 averages. Amplitude was greater when VOR gain was reduced, with a remarkably high correlation (R2 = 0.996). These findings confirm that vestibular feedback plays an important role in the control of voluntary head movement. Furthermore, the data suggest that instability of the visual scene reported by subjects shaking their heads after TR resulted not only from a lower VOR gain, but also from the inadvertent use of higher head velocities.