Effect of intrathecal clonidine on group I and group II oligosynaptic excitation in paraplegics

Music Restores Propriospinal Excitation During Stroke Locomotion

Music-based therapy for rehabilitation induces neuromodulation at the brain level and improves the functional recovery. In line with this, musical rhythmicity improves post-stroke gait. Moreover, an external distractor also helps stroke patients to improve locomotion. We raised the question whether music with irregular tempo (arrhythmic music), and its possible influence on attention would induce neuromodulation and improve the post-stroke gait. We tested music-induced neuromodulation at the level of a propriospinal reflex, known to be particularly involved in the control of stabilized locomotion; after stroke, the reflex is enhanced on the hemiparetic side. The study was conducted in 12 post-stroke patients and 12 controls. Quadriceps EMG was conditioned by electrical stimulation of the common peroneal nerve, which produces a biphasic facilitation on EMG, reflecting the level of activity of the propriospinal reflex between ankle dorsiflexors and quadriceps (CPQ reflex). The CPQ reflex was tested during treadmill locomotion at the preferred speed of each individual, in 3 conditions randomly alternated: without music vs. 2 arrhythmic music tracks, including a pleasant melody and unpleasant aleatory electronic sounds (AES); biomechanical and physiological parameters were also investigated. The CPQ reflex was significantly larger in patients during walking without sound, compared to controls. During walking with music, irrespective of the theme, there was no more difference between groups. In controls, music had no influence on the size of CPQ reflex. In patients, CPQ reflex was significantly larger during walking without sound than when listening to the melody or AES. No significant differences have been revealed concerning the biomechanical and the physiological parameters in both groups. Arrhythmic music listening modulates the spinal excitability during post-stroke walking, restoring the CPQ reflex activity to normality. The plasticity was not accompanied by any clear improvement of gait parameters, but the patients reported to prefer walking with music than without. The role of music as external focus of attention is discussed. This study has shown that music can modulate propriospinal neural network particularly involved in the gait control during the first training session. It is speculated that repetition may help to consolidate plasticity and would contribute to gait recovery after stroke.

Temporal Indices of Ankle Clonus and Relationship to Electrophysiologic and Clinical Measures in Persons With Spinal Cord Injury

BACKGROUND AND PURPOSE

Clonus arising from plantar flexor hyperreflexia is a phenomenon that is commonly observed in persons with spastic hypertonia. We assessed the temporal components of a biomechanical measure to quantify ankle clonus, and validated these in persons with spasticity due to spinal cord injury.

METHODS

In 40 individuals with chronic (>1 year) spinal cord injury, we elicited ankle clonus using a standardized mechanical perturbation (drop test). We examined reliability and construct validity of 2 components of the drop test: clonus duration (timed with a stopwatch) and number of oscillations in the first 10-second interval (measured via optical motion capture). We compared these measures to the Spinal Cord Assessment Tool for Spastic reflexes (SCATS) clonus score and H-reflex/M-wave (H/M) ratio, a clinical and electrophysiologic measure, respectively.

RESULTS

Intra- and interrater reliability of clonus duration measurement was good [intraclass correlation coefficient, ICC (2, 1) = 1.00]; test-retest reliability was good both at 1 hour [ICC (2, 2) = 0.99] and at 1 week [ICC (2, 2) = 0.99]. Clonus duration was moderately correlated with SCATS clonus score (r = 0.58). Number of oscillations had good within-session test-retest reliability [ICC (2, 1) > 0.90] and strong correlations with SCATS clonus score (r = 0.86) and soleus H/M ratio (r = 0.77).

DISCUSSION AND CONCLUSIONS

Clonus duration and number of oscillations as measured with a standardized test are reliable and valid measures of plantar flexor hyperreflexia that are accessible for clinical use. Tools for objective measurement of ankle clonus are valuable for assessing effectiveness of interventions directed at normalizing reflex activity associated with spasticity.Video Abstract available for more insights from the authors (see Supplemental Digital Content 1, http://links.lww.com/JNPT/A179).

Task-related modulation of crossed spinal inhibition between human lower limbs

Crossed reflex action mediated by muscle spindle afferent inputs has recently been revealed in humans. This raised the question of whether a complex spinal network involving commissural interneurons receiving inputs from proprioceptors and suprasegmental structures, as described in cats, persists in humans and contributes to the interlimb coordination during movement. First, we investigated the neurophysiological mechanisms underlying crossed reflex action between ankle plantar flexors and its corticospinal control from primary motor cortex. Second, we studied its modulation during motor tasks. We observed crossed inhibition in contralateral soleus motoneurons occurring with about 3 ms central latency, which is consistent with spinal transmission through oligosynaptic pathway. The early phase of inhibition was evoked with lower stimulus intensity than the late phase, suggesting mediation by group I and group II afferents, respectively. The postsynaptic origin of crossed inhibition is confirmed by the finding that both H-reflex and motor-evoked potential were reduced upon conditioning stimulation. Transcranial magnetic stimulation over ipsilateral and contralateral primary motor cortex reduced crossed inhibition, especially its late group II part. Last, late group II crossed inhibition was particularly depressed during motor tasks, especially when soleus was activated during the walking stance phase. Our results suggest that both group I and group II commissural interneurons participate in crossed reflex actions between ankle plantar flexors. Neural transmission at this level is depressed by descending inputs activated by transcranial magnetic stimulation over the primary motor cortex or during movement. The specific modulation of group II crossed inhibition suggests control from monoaminergic midbrain structures and its role for interlimb coordination during locomotion.

Spasticity

Antispastic medications that are directed to reduce clinical signs of spasticity, such as exaggerated reflexes and muscle tone, do not improve the movement disorder. Medication can even increase weakness which might interfere with functional movements, such as walking. In this chapter we address how spasticity affects mobility and how this should be taken into account in the treatment of spasticity. In clinical practice, signs of exaggerated tendon tap reflexes associated with muscle hypertonia are the consequence of spinal cord injury (SCI). They are generally thought to be responsible for spastic movement disorders. Most antispastic treatments are, therefore, directed at the reduction of reflex activity. In recent years, a discrepancy between spasticity as measured in the clinic and functional spastic movement disorder was noticed, which is primarily due to the different roles of reflexes in passive and active states, respectively. We now know that central motor lesions are associated with loss of supraspinal drive and defective use of afferent input with impaired behavior of short-latency and long-latency reflexes. These changes lead to paresis and maladaptation of the movement pattern. Secondary changes in mechanical muscle fiber, collagen tissue, and tendon properties (e.g., loss of sarcomeres, subclinical contractures) result in spastic muscle tone, which in part compensates for paresis and allows functional movements on a simpler level of organization. Antispastic drugs should primarily be applied in complete SCI. In mobile patients they can accentuate paresis and therefore should be applied with caution.

Gait disorders

This chapter deals with the neuronal mechanisms underlying impaired gait. The aim is, first, a better understanding of the underlying pathophysiology and, second, the selection of an adequate treatment. One of the first symptoms of a lesion within the central motor system perceived by patients is a movement disorder, which is most characteristic during locomotion, e.g. in patients suffering spasticity after stroke or a spinal cord injury or Parkinson disease. By the recording and analysis of electrophysiological and biomechanical signals during a movement, the significance of impaired reflex behavior or muscle tone and its contribution to the movement disorder can reliably be assessed. Adequate treatment should not be restricted to the correction of an isolated clinical sign but should be based on the mechanisms underlying the movement disorder that impairs the patient. Therapy should be directed toward functional training, which takes advantage of the plasticity of the nervous system. In the future a combination of repair and functional training will further improve the mobility of disabled patients.

[Pathophysiology of spasticity]

The term "spasticity" describes the velocity-dependent increase in tonic stretch reflexes. The symptom is commonly seen in patients with injury to the central nervous system. It is rarely isolated but, instead, part of a set of symptoms that is sometimes confusing. However, the pathophysiology of the symptom has evolved over the past three decades, and it is now considered part of a global process that includes not only spinal reflex loop modifications, but also changes in the biomechanical properties of muscle fibers. Finally, recent studies of changes in the membrane properties of motor neurons and the occurrence of plateau potential have opened new perspectives. This review aims to describe these new pathophysiological models.

Differential modulation of crossed and uncrossed reflex pathways by clonidine in adult cats following complete spinal cord injury

Clonidine, an α-noradrenergic agonist, facilitates hindlimb locomotor recovery after complete spinal transection (i.e. spinalization) in adult cats. However, the mechanisms involved in clonidine-induced functional recovery are poorly understood. Sensory feedback from the legs is critical for hindlimb locomotor recovery in spinalized mammals and clonidine could alter how spinal neurons respond to peripheral inputs in adult spinalized cats. To test this hypothesis we evaluated the effect of clonidine on the responses of hindlimb muscles, primarily in the left hindlimb, evoked by stretching the left triceps surae muscles and by stimulating the right tibial and superficial peroneal nerves in eight adult decerebrate cats that were spinalized 1 month before the terminal experiment. Cats were not trained following spinalization. Clonidine had no consistent effect on responses of ipsilateral muscles evoked by triceps surae muscle stretch. However, clonidine consistently potentiated the amplitude and duration of crossed extensor responses. Moreover, following clonidine injection, stretch and tibial nerve stimulation triggered episodes of locomotor-like activity in approximately one-third of trials. Differential effects of clonidine on crossed reflexes and on ipsilateral responses to muscle stretch indicate an action at a pre-motoneuronal site. We conclude that clonidine facilitates hindlimb locomotor recovery following spinalization in untrained cats by enhancing the excitability of central pattern generating spinal neurons that also participate in crossed extensor reflex transmission.

Locomotor-activated neurons of the cat. II. Noradrenergic innervation and colocalization with NEα 1a or NEα 2b receptors in the thoraco-lumbar spinal cord

Norepinephrine (NE) is a strong modulator and/or activator of spinal locomotor networks. Thus noradrenergic fibers likely contact neurons involved in generating locomotion. The aim of the present study was to investigate the noradrenergic innervation of functionally related, locomotor-activated neurons within the thoraco-lumbar spinal cord. This was accomplished by immunohistochemical colocalization of noradrenergic fibers using dopamine-β-hydroxylase or NEα(1A) and NEα(2B) receptors with cells expressing the c-fos gene activity-dependent marker Fos. Experiments were performed on paralyzed, precollicular-postmamillary decerebrate cats, in which locomotion was induced by electrical stimulation of the mesencephalic locomotor region. The majority of Fos labeled neurons, especially abundant in laminae VII and VIII throughout the thoraco-lumbar (T13-L7) region of locomotor animals, showed close contacts with multiple noradrenergic boutons. A small percentage (10-40%) of Fos neurons in the T7-L7 segments showed colocalization with NEα(1A) receptors. In contrast, NEα(2B) receptor immunoreactivity was observed in 70-90% of Fos cells, with no obvious rostrocaudal gradient. In comparison with results obtained from our previous study on the same animals, a significantly smaller proportion of Fos labeled neurons were innervated by noradrenergic than serotonergic fibers, with significant differences observed for laminae VII and VIII in some segments. In lamina VII of the lumbar segments, the degree of monoaminergic receptor subtype/Fos colocalization examined statistically generally fell into the following order: NEα(2B) = 5-HT(2A) ≥ 5-HT(7) = 5-HT(1A) > NEα(1A). These results suggest that noradrenergic modulation of locomotion involves NEα(1A)/NEα(2B) receptors on noradrenergic-innervated locomotor-activated neurons within laminae VII and VIII of thoraco-lumbar segments. Further study of the functional role of these receptors in locomotion is warranted.

Voluntary ankle flexor activity and adaptive coactivation gain is decreased by spasticity during subacute spinal cord injury

Although spasticity has been defined as an increase in velocity-dependent stretch reflexes and muscle hypertonia during passive movement, the measurement of flexor muscle paresis may better characterize the negative impact of this syndrome on residual motor function following incomplete spinal cord injury (iSCI). In this longitudinal study Tibialis Anterior (TA) muscle paresis produced by a loss in maximal voluntary contraction during dorsiflexion and ankle flexor muscle coactivation during ramp-and-hold controlled plantarflexion was measured in ten patients during subacute iSCI. Tibialis Anterior activity was measured at approximately two-week intervals between 3-5 months following iSCI in subjects with or without spasticity, characterized by lower-limb muscle hypertonia and/or involuntary spasms. Following iSCI, maximal voluntary contraction ankle flexor activity was lower than that recorded from healthy subjects, and was further attenuated by the presence of spasticity. Furthermore the initially high percentage value of TA coactivation increased at 75% but not at 25% maximal voluntary torque (MVT), reflected by an increase in TA coactivation gain (75%/25% MVT) from 2.5+/-0.4 to 7.5+/-1.9, well above the control level of 2.9+/-0.2. In contrast contraction-dependent TA coactivation gain decreased from 2.4+/-0.3 to 1.4+/-0.1 during spasticity. In conclusion the adaptive increase in TA coactivation gain observed in this pilot study during subacute iSCI was also sensitive to the presence of spasticity. The successful early diagnosis and treatment of spasticity would be expected to further preserve and promote adaptive motor function during subacute iSCI neurorehabilitation.

Enhanced spinal excitation from ankle flexors to knee extensors during walking in stroke patients

OBJECTIVES

It is still unclear to what an extent altered reflex activity contributes to gait deficit following stroke. Spinal group I and group II excitations from ankle dorsiflexors to knee extensors were investigated during post-stroke walking.

METHODS

Electrical stimulation was applied to the common peroneal nerve (CPN) in the early stance, and the short-latency biphasic excitation in Quadriceps motoneurones was evaluated from the Vastus Lateralis (VL) rectified and averaged (N=50) EMG activity in 14 stroke patients walking at 0.6-1.6 km/h, and 14 control subjects walking at 3.2-4.8 and at 1 km/h.

RESULTS

The second peak of the CPN-induced biphasic facilitation in VL EMG activity, which is likely mediated by group II excitatory pathways, was larger on the paretic side of the patients, as compared to their nonparetic side or control subjects, whatever their walking speed.

CONCLUSIONS

The spinal, presumed group II, excitation from ankle dorsiflexors to knee extensors is particularly enhanced during post-stroke walking probably due to plastic adaptations in the descending control.

SIGNIFICANCE

This adaptation may help to stabilize the knee in early stance when the patients have recover ankle dorsiflexor functions.

Corticospinal inhibition of transmission in propriospinal-like neurones during human walking

It is crucial for human walking that muscles acting at different joints are optimally coordinated in relation to each other. This is ensured by interaction between spinal neuronal networks, sensory feedback and supraspinal control. Here we investigated the cortical control of spinal excitation from ankle dorsiflexor afferents to quadriceps motoneurones mediated by propriospinal-like interneurones. During walking and tonic contraction of ankle dorsiflexors and knee extensors while standing [at matched electromyography (EMG) levels], the effect of common peroneal nerve (CPN) stimulation on quadriceps motoneurones was tested by assessing averaged and rectified EMG activity, H-reflexes [evoked by femoral nerve (FN) stimulation] and motor evoked potentials (MEPs) produced by transcranial magnetic stimulation (TMS). The biphasic EMG facilitation (CPQ-reflex) produced by isolated CPN stimulation was enhanced during walking, and when CPN stimulation was combined with FN or TMS, the resulting H-reflexes and MEPs were inhibited. The CPQ-reflex was also depressed when CPN stimulation was combined with subthreshold TMS. The peripheral (in CPN and FN) and corticospinal volleys may activate inhibitory non-reciprocal group I interneurones, masking spinal excitations to quadriceps motoneurones mediated by propriospinal-like interneurones. It is proposed that the enhanced CPQ-reflex produced by isolated CPN stimulation during walking cannot be fully explained by an increase in corticospinal and peripheral inputs, but is more likely caused by central facilitation of the propriospinal-like interneurones from other sources. The corticospinal control of non-reciprocal group I interneurones may be of importance for reducing reflex activity between ankle dorsiflexors and quadriceps during walking when not functionally relevant.

Speed-related spinal excitation from ankle dorsiflexors to knee extensors during human walking

Automatic adjustments of muscle activity throughout the body are required for the maintenance of balance during human walking. One mechanism that is likely to contribute to this control is the heteronymous spinal excitation between human ankle dorsiflexors and knee extensors (CPQ-reflex). Here, we investigated the CPQ-reflex at different walking speeds (1-6 km/h) and stride frequencies (0.6-1.3 Hz) in healthy human subjects to provide further evidence of its modulation, and its role in ensuring postural stability during walking. The CPQ-reflex was small or absent at walking speeds below 2-3 km/h, then increased with walking speeds about 3-4 km/h, and reached a plateau without any further change at walking speeds from 4 to 6 km/h. The reflex showed no modulation when the stride cycle was varied at constant speed (4 km/h; short steps versus long steps). These changes were unlikely to be only caused by changes in the background EMG activity and modifications in peripheral input, and likely reflected central modulation of transmission in the involved reflex pathways as well. It is suggested that the purpose of the reflex is to ensure knee stability at moderate-to-high walking speeds.

Spastic movement disorder: impaired reflex function and altered muscle mechanics

In clinical practice, signs of exaggerated tendon tap reflexes associated with muscle hypertonia are generally thought to be responsible for spastic movement disorders. Most antispastic treatments are, therefore, directed at the reduction of reflex activity. In recent years, however, researchers have noticed a discrepancy between spasticity as measured in the clinic and functional spastic movement disorders, which is primarily due to the different roles of reflexes in passive and active states, respectively. We now know that central motor lesions are associated with loss of supraspinal drive and defective use of afferent input with impaired behaviour of short-latency and long-latency reflexes. These changes lead to paresis and maladaptation of the movement pattern. Secondary changes in mechanical muscle fibre, collagen tissue, and tendon properties (eg, loss of sarcomeres, subclinical contractures) result in spastic muscle tone, which in part compensates for paresis and allows functional movements on a simpler level of organisation. Antispastic drugs can accentuate paresis and therefore should be applied with caution in mobile patients.

Effects of changing wrist positions on finger flexor hypertonia in stroke survivors

We sought to establish whether spastic hypertonia results from changes in intrinsic muscle properties or from altered stretch reflex properties. We hypothesized that finger flexor spastic hypertonia is primarily of neural origin, and that the dynamics of spastic muscle responses to stretch should therefore reflect the dynamics of muscle spindle receptor responses. In 12 stroke survivors, we recorded torque and electromyographic (EMG) responses of extrinsic finger flexors to constant-velocity rotation of the metacarpophalangeal (MCP) joints of the affected hand, over a range of initial muscle lengths. Stretch velocity was set to 6 degrees, 50 degrees, 150 degrees, or 300 degrees per second. Muscle length changes were imposed by changing wrist angle between 0 degree, 25 degrees, and 50 degrees of flexion. We found that reflex torque and EMG responses exhibited both velocity and length dependence, and there were significant interactions between velocity and length, replicating known characteristics of muscle spindle receptors. Our results support the hypothesis that finger flexor hypertonia is primarily of neural origin, and that it accurately reflects spindle receptor firing properties.

Rehabilitative Therapies after Spinal Cord Injury

We review some basic and highly relevant concepts in the effort to develop improved rehabilitative interventions for subjects with spinal cord injury (SCI). Interventions that are likely to contribute to improved sensorimotor function include (1) practice of the specific motor task that needs to be improved; and (2) combining the training with one or more interventions--such as pharmacological modulation of the excitability of spinal neural networks, implantation of selected cell types such as olfactory ensheathing glia (OEG), and/or modulation of the excitability of the spinal cord via epidural stimulation. Upon improvement of the neural control of the musculature following SCI, it will always be prudent to maximize the torque output from these activation patterns by assuring that muscle mass is maintained. Therefore, it seems quite feasible that considerable improvement in locomotor performance can be achieved by improved coordination of motor pools, as well as effective recovery of muscle mass, which will assist in the potential generation of normal forces among agonistic and antagonistic muscle groups.

Depression of involuntary activity in muscles paralyzed by spinal cord injury

Involuntary muscle contractions are common after spinal cord injury (SCI). Increased sensitivity to Ia muscle afferent input may contribute to the development of these spasms. Since tendon vibration results in a period of postactivation depression of the Ia synapse, we sought to determine whether Achilles tendon vibration (80 HZ for 2 s) altered involuntary contractions evoked by superficial peroneal nerve (SPN) stimulation (5 pulses at 300 HZ) in paralyzed leg muscles of subjects with chronic (>1 year) SCI. Responses to SPN stimulation that were conditioned by vibration were reduced in 66% of trials (by 33+/-12% in tibialis anterior and 40+/-16% in soleus). These reductions in electromyographic activity are unlikely to be mediated by changes at the Ia synapse or motoneuron because vibration did not alter the magnitude of the soleus H reflex. The electromyographic reductions may involve long-lasting neuromodulatory effects on spinal inhibitory interneurons or synapses involved in the flexor reflex pathway. Vibration-evoked depression of electromyographic activity may be clinically useful in controlling involuntary muscle contractions after SCI.

Pathophysiology of spastic paresis. II: Emergence of muscle overactivity

In the subacute and chronic stages of spastic paresis, stretch-sensitive (spastic) muscle overactivity emerges as a third fundamental mechanism of motor impairment, along with paresis and soft tissue contracture. Part II of this review primarily addresses the pathophysiology of the various forms of spastic overactivity. It is argued that muscle contracture is one of the factors that cause excessive responsiveness to stretch, which in turn aggravates contracture. Excessive responsiveness to stretch also impedes voluntary motor neuron recruitment, a concept termed stretch-sensitive paresis. None of the three mechanisms of impairment (paresis, contracture, and spastic overactivity) is symmetrically distributed between agonists and antagonists, which generates torque imbalance around joints and limb deformities. Thus, each may be best treated focally on an individual muscle-by-muscle basis. Intensive motor training of the less overactive muscles should disrupt the cycle of paresis-disuse-paresis, and concomitant use of aggressive stretch and focal weakening agents in their more overactive and shortened antagonists should break the cycle of overactivity-contracture-overactivity.

Reflex contribution of spindle group Ia and II afferent input to leg muscle spasticity as revealed by tendon vibration in hemiparesis

OBJECTIVE

Foot dorsiflexion evokes a short- (SLR) and a medium-latency EMG response (MLR) in the soleus of standing subjects. SLR is mediated by spindle group Ia, while group II fibres contribute to MLR through an oligosynaptic circuit. We studied the effects of Achilles' tendon vibration on both responses in spastic patients to disclose any abnormal excitability of these pathways.

METHODS

SLR and MLR were evoked in 11 hemiparetics and 11 normals. The vibration-induced changes in both responses were correlated to the Ashworth score of the affected leg.

RESULTS

There were no differences between normals and patients in the size of control SLR or MLR. Vibration decreased SLR to 70% in normal subjects, but increased it to 110% in patients, in both affected and unaffected leg. Vibration did not affect MLR in normals, but increased it to 165% on the affected and 120% on the unaffected side of patients. Ashworth score was solely correlated with the degree of vibration-induced increase of MLR.

CONCLUSIONS

While the lack of inhibitory effect of vibration on SLR confirms a reduced inhibitibility of the monosynaptic reflex, the increased MLR indicates a disinhibition of group II pathway in patients, connected to the loss of descending control on group II interneurones. Spastic hypertonia depends on release of group II rather than group Ia reflex pathways.

SIGNIFICANCE

These findings give a neurophysiological support for the pharmacological treatment of spastic hypertonia and suggest a method for the assessment of its effects.