Continuous, bilateral Achilles’ tendon vibration is not detrimental to human walk

Sensory feedback from the moving limbs contributes to the regulation of animal and human locomotion. However, the question of the specific role of the various modalities is still open. Further, functional loss of leg afferent fibres due to peripheral neuropathy does not always lead to major alteration in the gait pattern. In order to gain further insight on proprioceptive control of human gait, we applied vibratory tendon stimulation, known to recruit spindle primary afferent fibres, to both triceps surae muscles during normal floor walk. This procedure would disturb organisation and execution of walking, especially if spindles fire continuously and subjects are blindfolded. Vibration induced significant, though minor, changes in duration and length of stance and swing phase, and on speed of walking and kinematics of lower limb segments. No effect was induced on angular displacement of the ankle joint or trunk and head kinematics. This paucity of effects was at variance with the perception of the subjects, who reported illusion of leg stiffness and gait imbalance. These findings would speak for a selective gating of Ia input during locomotion and emphasise the notion that the central nervous system can cope with an unusual continuous input along the Ia fibres from a key muscle like the soleus.     

The complex role of spindle afferent input,as evidenced by the study of posture control in normal subjects and patients

This report reviews recent findings from our laboratory on the connectivity of the group II spindle afferent input and the role played by these afferents in the control of quiet and perturbed human stance. At variance with group Ia fibres, which subserve the monosynaptic stretch reflex, the group II fibres, after having entered the spinal cord, make synamptic contacts with a short chain of interneurons which impinge on homonymous motoneurons. Analysis of the short- and medium-latency responses evoked in foot and leg muscles by perturbations of upright stance under different experimental conditions has revealed a role of group II fibres in the production of the medium-latency response. The conduction velocity of group II spindle afferent fibres and their central delay have also been estimated. Furthermore, data from normal subjects and from neuropathic and hemiparetic patients are in favour of a prevailing role of the input from group II fibres in the afferent control of quiet and perturbed stance. Since Ia fibres innervate receptors more sensitive to the veolocity of muscle stretch, and II fibres innervate receptors more sensitive to the absolute value of muscle length, it is hypothesised that the major role of the latter in the reflex control of stance reflects the slow velocity and amplitude of sway during quiet upright posture. Indirect evidence supports the conclusion that, also in humans, monoaminergic descending pathways from brainstem nuclei modulate the excitability of the circuits mediating the group II input.     

The complex role of spindle afferent input, as evidenced by the study of posture control in normal subjects and patients

This report reviews recent findings from our laboratory on the connectivity of the group II spindle afferent input and the role played by these afferents in the control of quiet and perturbed human stance. At variance with group Ia fibres, which subserve the monosynaptic stretch reflex, the group II fibres, after having entered the spinal cord, make synamptic contacts with a short chain of interneurons which impinge on homonymous motoneurons. Analysis of the short- and medium-latency responses evoked in foot and leg muscles by perturbations of upright stance under different experimental conditions has revealed a role of group II fibres in the production of the medium-latency response. The conduction velocity of group II spindle afferent fibres and their central delay have also been estimated. Furthermore, data from normal subjects and from neuropathic and hemiparetic patients are in favour of a prevailing role of the input from group II fibres in the afferent control of quiet and perturbed stance. Since Ia fibres innervate receptors more sensitive to the veolocity of muscle stretch, and II fibres innervate receptors more sensitive to the absolute value of muscle length, it is hypothesised that the major role of the latter in the reflex control of stance reflects the slow velocity and amplitude of sway during quiet upright posture. Indirect evidence supports the conclusion that, also in humans, monoaminergic descending pathways from brainstem nuclei modulate the excitability of the circuits mediating the group II input.     

Quadrupedal coordination of bipedal gait: implications for movement disorders

During recent years, evidence has come up that bipedal locomotion is based on a quadrupedal limb coordination. A task-dependent neuronal coupling of upper and lower limbs allows one to involve the arms during gait but to uncouple this connection during voluntarily guided arm/hand movements. Hence, despite the evolution of a strong cortico-spinal control of hand/arm movements in humans, a quadrupedal limb coordination persists during locomotion. This has consequences for the limb coordination in movement disorders such as in Parkinson’s disease (PD) and after stroke. In patients suffering PD, the quadrupedal coordination of gait is basically preserved. The activation of upper limb muscles during locomotion is strong, similar as in age-matched healthy subjects although arm swing is reduced. This suggests a contribution of biomechanical constraints to immobility. In post-stroke subjects a close interactions between unaffected and affected sides with an impaired processing of afferent input takes place. An afferent volley applied to a leg nerve of the unaffected leg leads to a normal reflex activation of proximal arm muscles of both sides. In contrast, when the nerve of the affected leg was stimulated, neither on the affected nor in the unaffected arm muscles EMG responses appear. Muscle activation on the affected arm becomes normalized by influences of the unaffected side during locomotion. These observations have consequences for the rehabilitation of patients suffering movement disorders.     

In humans Ib facilitation depends on locomotion while suppression of Ib inhibition requires loading

The role of force feedback during gait is still a matter of debate. From work on cats, it is known that input from Golgi tendon organs from triceps surae does produce Ib facilitation during locomotion instead of autogenic inhibition. In humans, Stephens and Yang (Stephens, M.J., Yang, J.F., 1996. Short latency, non-reciprocal group I inhibition is reduced during the stance phase of walking in humans. Brain Res. 743, 24-31) found that voluntary contraction results in a reduction of Ib inhibition. During gait, they even observed Ib facilitation in a subset of subjects. This raises the question whether the crucial elements involved in these changes are either loading of the leg or locomotion. To examine this question, Ib reflexes were investigated during sitting, lying supine, lying supine with 300 N pressure applied to the foot sole, standing, and a rhythmic loading and unloading task called "reduced" gait. Ib inhibition was obtained during sitting and lying supine. This inhibition was significantly reduced or disappeared during standing and when lying supine but loaded. During the stance phase of "reduced" gait, the inhibition disappeared in eight subjects, and even a facilitation was observed in six subjects. It is concluded that the decrease in Ib inhibition from gastrocnemius to soleus occurs during a load-bearing condition and does not require locomotion. In contrast, Ib facilitation requires locomotion at least in a rudimentary form.     

Gait analysis during treadmill and overground locomotion in children and adults

Gait analysis on the treadmill and in the overground condition is used both in scientific approaches for investigating the neuronal organisation and ontogenetic development of locomotion and in a variety of clinical applications. We investigated the differences between overground and treadmill locomotion (at identical gait velocity) in 12 adults and 14 children (6–7 years old). During treadmill locomotion the step frequency increased by 7% in adults and 10% in children compared to overground walking, whereas the stride length and the stance phase of the walking cycle decreased. The swing phase, however, increased significantly by 5% in adults and remained unchanged in children. Balance-related gait parameters such as the step width and foot rotation angles increased during treadmill locomotion. The reduction of the step length was found to be stable after 10 min of treadmill walking in most subjects. With regard to the shifted phases of the walking cycle and the changed balance related gait parameters in the treadmill condition, we assume a different modulation of the central pattern generator in treadmill walking, due to a changed afferent input. Regarding the pronounced differences between overground and treadmill walking in children, it is discussed whether the systems generating and integrating different modulations of locomotion into a stable movement pattern have reached full capacity in 6–7 year old children.     

A dynamic EMG profile index to quantify muscular activation disorder in spastic paretic gait

Spasticity is a complex phenomenon that interferes with motor control. Existing clinical and physiological measures of spasticity have mainly focused on the evaluation of clonus and reflexes. Subjected to the limitation of testing in a resting position, the results may not necessarily reflect the extent of functional impairment caused by spasticity. To evaluate spasticity in a dynamic, voluntary movement such as locomotion, a task-specific approach is essential. A dynamic index, I, derived from the EMG activity obtained during treadmill walking in human subjects, is therefore proposed as a functionally relevant measurement of spasticity in locomotion. I, defined as the ratio of integrated EMG in the pre-determined 'off' window of the normalized gait cycle to that in the 'on' window, would indicate the degree of abnormal activation of locomotor muscles from their normally relaxed state as compared to the total recruitment in the active state during walking. The present study done on 5 normal and 8 spastic paraparetic subjects showed that I was homogeneously low in the normal group but abnormally high and variable in the spastic group. A case study has further demonstrated that I is sensitive to the alteration in locomotor spasticity with pharmacological intervention, and the change in I parallels the improvement in the kinematics observed. This preliminary study indicates that the proposed index appears to be a functionally relevant and dynamic measurement of spastic locomotor disorder.     

Kinematic analysis of the different types of locomotor movement in rats after deafferentation

Comparative analysis of kinematics of rat hindlimb movements during different kind of locomotion (swimming, walking, hindlimb swimming but forelimb walking) was performed. After deafferentation an increase of the locomotor rhythm frequency and a decrease of the movement amplitude of hindlimb joints were common features for all kinds of the locomotion. This indicates that the above parameters of locomotor movements were formed by afferent influences. A decrease of the movement amplitude and of the EMG activity which were evoked after deafferentation were minimal for swimming and maximal for walking. The role of afferent influences in formation of different locomotion patterns is discussed.     

Neuronal plasticity after a human spinal cord injury: positive and negative effects

In patients suffering an incomplete spinal cord injury (SCI) an improvement in walking function can be achieved by providing a functional training with an appropriate afferent input. In contrast, in immobilized incomplete and complete subjects a negative neuroplasticity leads to a neuronal dysfunction. After an SCI, neuronal centers below the level of lesion exhibit plasticity that either can be exploited by specific training paradigms or undergo a degradation of function due to the loss of appropriate input. Load- and hip-joint-related afferent inputs seem to be of crucial importance for the generation of a locomotor pattern and, consequently, the effectiveness of the locomotor training. In severely affected SCI subjects rehabilitation robots allow for a longer and more intensive training and can provide feedback information. Conversely, in severely affected chronic SCI individuals without functional training the locomotor activity in the leg muscles exhausts rapidly during assisted locomotion. This is accompanied by a shift from early to dominant late spinal reflex components. The exhaustion of locomotor activity is also observed in non-ambulatory patients with an incomplete SCI. It is assumed that in chronic SCI the patient's immobility results in a reduced input from supraspinal and peripheral sources and leads to a dominance of inhibitory drive within spinal neuronal circuitries underlying locomotor pattern and spinal reflex generation. A training with an enhancement of an appropriate proprioceptive input early after an SCI might serve as an intervention to prevent neuronal dysfunction.     

Gait disturbances related to dysfunction of the cerebral cortex and basal ganglia

This review aimed to characterize the gait disturbances in Parkinson disease (PD) and highlight how a rehabilitation program would affect the care of patients with PD. The typical PD gait is a type of hypokinetic gait characterized by reduced stride length and velocity; shortening of the swing phase; and increase in the stance phase, double-limb support duration, and cadence rate. In the advanced phase of PD, start hesitation, shuffling and festinating gait, propulsion, and freezing of gait (FOG) become remarkable. Notably, in PD, attention may influence gait control, and sensory cueing may improve the stride length. Our study on gait impairment in PD by using a three-dimensional motion analysis system revealed that the stride length and walking speed decreased, but there was no change in cadence. The decreased stride length was due to reduction in the range of movement at the leg and pelvic joints. A 4-week physical rehabilitation program for PD improved the stride length and walking speed;this was achieved by increasing the range of movement of at the leg and pelvic joints. We also assessed the effects of a rehabilitation program for patients with PD who experienced FOG. Although the lower limb function was more impaired in patients with PD and FOG than in those with PD without FOG, the rehabilitation program was effective even for patients with PD and FOG. FOG might be associated with functional impairment of the lower limb as well as dysfunction of the fronto-basal ganglia circuit. We also reported 3 cases of camptocormia (bent spine syndrome) with autonomic dysfunction and rapid eye movement (REM) sleep behavior disorders (RBD) and compared their symptoms with those reported elsewhere. We think that the pedunculopontine nuclear area may control the postural muscle tone and locomotion in PD. On the basis of the results of our rehabilitation programs, we speculate that physical modalities may modify synaptic plasticity by utilizing the cerebellar and/or afferent sensory system. These alternative systems are believed to be functionally intact in patients with PD.     

Plasticity of interneuronal networks of the functionally isolated human spinal cord

The loss of walking after human spinal cord injury has been attributed to the dominance of supraspinal over spinal mechanisms. The evidence for central pattern generation in humans is limited due to the inability to conclusively isolate the circuitry from descending and afferent input. However, studying individuals following spinal cord injury with no detectable influence on spinal networks from supraspinal centers can provide insight to their interaction with afferent input. The focus of this article is on the interaction of sensory input with human spinal networks in the generation of locomotor patterns. The functionally isolated human spinal cord has the capacity to generate locomotor patterns with appropriate afferent input. Locomotor Training is a rehabilitative strategy that has evolved from animal and humans studies focused on the neural plasticity of the spinal cord and has been successful for many people with acute and chronic incomplete spinal cord injury. However, even those individuals with clinically complete spinal cord injury that generate appropriate locomotor patterns during stepping with assistance on a treadmill with body weight support cannot sustain overground walking. This suggests that although a significant control of locomotion can occur at the level of spinal interneuronal networks the level of sustainable excitability of these circuits is still compromised. Future studies should focus on approaches to increase the central state of excitability and may include neural repair strategies, pharmacological interventions or epidural stimulation in combination with Locomotor Training.     

Spinal neuronal dysfunction after stroke

Central nervous system lesions, such as stroke or spinal cord injury (SCI), are followed by both cortical and spinal neuronal reorganization. In a severe chronic SCI a spinal neuronal dysfunction develops which is reflected in an exhaustion of leg muscle electromyographic (EMG) activity during assisted locomotion and a change in the dominance from an early to a late polysynaptic spinal reflex (SR) component. The aim of this study was to investigate the course of spinal neuronal function after a severe stroke, i.e., a unilateral deprivation of supraspinal input. In 30 hemiparetic stroke subjects locomotor and SR behavior were assessed. SR responses in the tibialis anterior muscle were evoked by non-noxious tibial nerve stimulation on both, the affected and the unaffected leg. In nine stroke subjects EMG activity of the leg muscles was recorded during assisted locomotion. In a similar way to SCI subjects, in severely affected chronic (>12 months post-incidence) stroke subjects a late SR component was prominent in the affected leg, while an early one dominated in the unaffected leg. The late SR component correlated with muscle paresis (rho=0.714) and walking ability (rho=0.493). In contrast to SCI subjects, no exhaustion of the EMG activity was observed in the affected leg muscles during prolonged assisted locomotion. It is concluded that spinal neuronal circuits undergo functional changes also after a stroke which have common as well as divergent features compared to SCI subjects. As a consequence, different rehabilitative strategies might be required.     

Steady-state fluctuations of human walking.

In steady-state walking, fluctuations in space-time behavior are observed for normal adult subjects. In the present study, the intrinsic fluctuations of gait have been analyzed when walking on a subject-driven treadmill (with adjustable inertial forces). Furthermore, these intrinsic fluctuations have been compared with those observed in natural overground locomotion which involves a real subject's displacement and thus an optical flow. Four adult subjects participated in both experimental sessions. It was found that the frequency and amplitude of the instantaneous fluctuations of leg movement were weak and of equal magnitude with or without optical flow. This was also the case for instantaneous fluctuations in displacement speed. Secondly, a low-frequency fluctuation in walking speed was observed when no optical flow information was available to the subject. This fluctuation results from the addition of a series of leg-movement fluctuations, whose values are all either positive or negative. As the optical flow provides information about the displacement speed, it allows the subject to avoid such addition, and thus plays a role in maintaining steady leg movement. Theoretical models linking space-time behavior of rhythmic movement with stiffness strongly suggest that the observed low-frequency fluctuations in speed result from fluctuations in stiffness.     

Muscle spindle activity in alternating tremor of Parkinsonism and in clonus.

Single unit activity in spindle afferent nerve fibres from the finger flexors, the anterior tibial muscle, and the calf muscles was recorded intraneurally with tungsten microelectrodes in patients with Parkinsonism with resting tremor and in spastic patients with clonus. During tremor of Parkinsonism, involving the receptor bearing muscles, the Ia afferent fibre discharge patterns were similar to those seen previously in healthy subjects during voluntary fast alternating finger or foot movements: besides the stretch discharges occurring during the relaxation phases, discharges also occurred during the contraction phases. Such contraction discharges, presumed to originate from intrafusal muscle fibre contractions, were not seen in the spastic patients during clonus. During the clonic oscillations each afferent stretch discharge was regularly followed by a stretch reflex contraction which on its falling phase elicited a new volley of impulses in the Ia afferent fibres. The findings are considered to support the notion that, like the contractions in normal voluntary alternating movements, the contractions in tremor of Parkinsonism are organized according to the principle of alpha-gamma coactivation, whereas the contractions in clonus are stretch reflexes causing pure alpha contractions.     

Selective suppression of the vestibulo-ocular reflex during human locomotion

Introduction

In lower vertebrates, gaze stabilization during locomotion is at least partially driven by a direct coupling of spinal locomotor commands with extraocular motor signals. To what extent locomotor feed-forward mechanisms contribute to gaze stabilization during human locomotion is yet unknown. In principle, the feasibility of a feed-forward regulation of gaze during locomotion should critically depend on the spatiotemporal coupling between body and head kinematics and hence the internal predictability of head movements (HMP). The present study thus investigated whether changes in eye–head coordination during human locomotion can be explained by concurrent changes in HMP.

Methods

Eye and head movements were recorded at different locomotor speeds in light and darkness to obtain the gain and phase of the horizontal and vertical angular VOR (aVOR). Potential correlations between aVOR performance and HMP were analyzed in dependence of locomotor speed and gait cycle phase.

Results

Horizontal aVOR responses persisted independent of locomotor speed. In contrast, with increasing locomotor speed vertical eye–head coordination switched from a VOR-driven compensatory mode to a synergistic behavior where head and eyes move in phase. Concurrently, vertical HMP increased with faster locomotion. Furthermore, modulations in vertical aVOR gain across the gait cycle corresponded to simultaneous alterations in vertical HMP.

Conclusion

The vertical aVOR appears to be suppressed during faster walking and running, whereas at the same time, the predictability of resultant head movements increases. This suggests that during stereotyped human locomotion, internal feed-forward commands supplement or even suppress sensory feedback to mediate gaze stabilization in the vertical plane.

     

Plantar flexor stretch reflex responses to whole body loading/unloading during human walking

Numerous animal and human studies have shown that afferent information from the periphery contributes to the control of walking. In particular, recent studies have consistently shown that load receptor input is an important element of the locomotion control mechanism. The objective of this study was to investigate the contribution of load receptor feedback to the compensatory stretch reflex response. We examined the contribution of load receptor feedback to the magnitude of the short and medium latency components of the ankle plantar flexor stretch reflex responses following an unexpected dorsiflexion perturbation during human walking. Three body load conditions were investigated: normal body load, a 30% increase in body load, and a 30% decrease in body load. Healthy subjects walked on a treadmill at approximately 3.6 km/h with the left ankle attached to a portable stretching device. Dorsiflexion perturbations (8 degrees; 350-425 degrees/s) were generated during the late stance phase of gate (approximately 400 ms following heel contact). Electromyographic activity was recorded from the soleus, tibialis anterior, medial gastrocnemius, rectus femoris, and biceps femoris muscles using bipolar surface electrodes. Stretch reflex responses were observed in the soleus and gastrocnemius muscles for all of the body load conditions; however, increasing or decreasing the body load did not affect the timing and magnitude of the responses. This study provides evidence that load receptor input does not contribute strongly to the corrective response of the stretch reflex in the plantar flexor muscles during walking.     

Locomotor‐like movements evoked by leg muscle vibration in humans

We attempted to elicit automatic stepping in healthy humans using appropriate afferent stimulation. It was found that continuous leg muscle vibration produced rhythmic locomotor-like stepping movements of the suspended leg, persisting up to the end of stimulation and sometimes outlasting it by a few cycles. Air-stepping elicited by vibration did not differ from the intentional stepping under the same conditions, and involved movements in hip and knee joints with reciprocal electromyogram (EMG) bursts in corresponding flexor and extensor muscles. The phase shift between evoked hip and knee movements could be positive or negative, corresponding to ‘backward’ or ‘forward’ locomotion. Such an essential feature of natural human locomotion as alternating movements of two legs, was also present in vibratory-evoked leg movements under appropriate conditions. It is suggested that vibration evokes locomotor-like movements because vibratory-induced afferent input sets into active state the central structures responsible for stepping generation.     

Leg muscle reflexes mediated by cutaneous A-beta fibres are normal during gait in reflex sympathetic dystrophy.

OBJECTIVES: Reflex sympathetic dystrophy (RSD) is, from the onset, characterized by various neurological deficits such as an alteration of sensation and a decrease in muscle strength. We investigated if afferent A-beta fibre-mediated reflexes are changed in lower extremities affected by acute RSD.METHODS: The involvement of these fibres was determined by analyzing reflex responses from the tibialis anterior (TA) and biceps femoris (BF) muscles after electrical stimulation of the sural nerve. The reflexes were studied during walking on a treadmill to investigate whether the abnormalities in gait of the patients were related either to abnormal amplitudes or deficient phase-dependent modulation of reflexes. In 5 patients with acute RSD of the leg and 5 healthy volunteers these reflex responses were determined during the early and late swing phase of the step cycle.RESULTS: No significant difference was found between the RSD and the volunteers. During early swing the mean amplitude of the facilitatory P2 responses in BF and TA increased as a function of stimulus intensity (1.5, 2 and 2.5 times the perception threshold) in both groups. At end swing the same stimuli induced suppressive responses in TA. This phase-dependent reflex reversal from facilitation in early swing to suppression in late swing occurred equally in both groups. CONCLUSIONS: In the acute phase of RSD of the lower extremity there is no evidence for abnormal A-beta fibre-mediated reflexes or for defective regulation of such reflexes. This finding has implications for both the theory on RSD pathophysiology and RSD models, which are based on abnormal functioning of A-beta fibres.     

Mechanically induced ankle inversion during human walking and jumping

A new method to study sudden ankle inversions during human walking and jumping is presented. Ankle inversions of 25 degrees were elicited using a box containing a trap door. During the gait task, subjects walked at a speed of 4 km/h. At a pre-programmed delay after left heel strike, an electromagnet released the box on the treadmill. This delay enabled the subject to step on the box without having to change the walking cadence. During the jumping task, subjects jumped from a 30 cm high platform on the box in a standardised way. In both tasks 20 stimulus and 20 control trials were presented randomly. The average tilting velocity of the trap door during the stimulus trials was 403 degrees /s during the walking task and 595 degrees /s during the jumping task. For the control trials a tilting of 0 degrees was used. With this method it is possible to evoke reproducible ankle inversions causing characteristic EMG responses in six lower leg muscles.     

Locomotion speed determines gait variability in cerebellar ataxia and vestibular failure

Temporal gait variability is a critical parameter in patients with balance problems. Increased magnitude of temporal gait variability corresponds to a higher risk of falls. The purpose of this study was to investigate the influence of walking speed on temporal stride-to-stride variability in patients with cerebellar and vestibular deficits. A GAITRite system was used to analyze the gait of 40 patients with cerebellar ataxia, 22 patients with bilateral vestibular failure, and 51 healthy subjects over the entire range of the individual's speed capacity. The coefficient of variability of stride time was calculated for each walk. Temporal gait variability was increased in cerebellar patients and vestibular patients. The magnitude of this variability depended on walking speed in a disease-specific manner. In patients with cerebellar ataxia, variability was increased during slow (8.4 ± 5.3%, P < .01) and fast (7.9 ± 6.4%, P < .01) walking speed but was normal during preferred walking speed. This resulted in a speed-related U-shaped function of stride-time variability. Patients with vestibular failure had increased variability during slow walking (9.9 ± 4.3%, P < .01). During walking with medium and fast walking speed, stride time variability was normal. Minimal temporal gait variability appears to be attractive for the locomotor system in cerebellar patients because these patients preferred to walk at a velocity associated with minimal stride-time variability. In contrast to previous studies, vestibular patients accelerate rather than decelerate gait to achieve dynamic stability. This may be explained by reduced sensory integration during fast locomotion.