Parallel facilitatory reflex pathways from the foot and hip to flexors and extensors in the injured human spinal cord

Spinal integration of sensory signals associated with hip position, muscle loading, and cutaneous sensation of the foot contributes to movement regulation. The exact interactive effects of these sensory signals under controlled dynamic conditions are unknown. The purpose of the present study was to establish the effects of combined plantar cutaneous afferent excitation and hip movement on the Hoffmann (H) and flexion reflexes in people with a spinal cord injury (SCI). The flexion and H-reflexes were elicited through stimulation of the right sural (at non-nociceptive levels) and posterior tibial nerves respectively. Reflex responses were recorded from the ipsilateral tibialis anterior (TA) (flexion reflex) and soleus (H-reflex) muscles. The plantar cutaneous afferents were stimulated at three times the perceptual threshold (200 Hz, 24-ms pulse train) at conditioning-test intervals that ranged from 3 to 90 ms. Sinusoidal movements were imposed to the right hip joint at 0.2 Hz with subjects supine. Control and conditioned reflexes were recorded as the hip moved in flexion and extension. Leg muscle activity and sagittal-plane joint torques were recorded. We found that excitation of plantar cutaneous afferents facilitated the soleus H-reflex and the long latency flexion reflex during hip extension. In contrast, the short latency flexion reflex was depressed by plantar cutaneous stimulation during hip flexion. Oscillatory joint forces were present during the transition phase of the hip movement from flexion to extension when stimuli were delivered during hip flexion. Hip-mediated input interacts with feedback from the foot sole to facilitate extensor and flexor reflex activity during the extension phase of movement. The interactive effects of these sensory signals may be a feature of impaired gait, but when they are appropriately excited, they may contribute to locomotion recovery in these patients.     

Functional organization of long ascending propriospinal pathways linking lumbo-sacral and cervical segments in the cat

In high spinal cats propriospinal pathways ascending from lumbo-sacral levels of the spinal cord can mediate strong excitatory and inhibitory changes in reflexes to different groups of motoneurones supplying muscles of the forelimb. Discharges evoked by electrical stimulation of hindlimb nerves could be evoked in 41% of experiments in the motoneurones of pectoralis major and minor. The latency of the discharge (8–18 msec) could be shortened by increasing the repetition frequency of the stimulus, the greatest reduction occurring in the range 1–4 Hz. Contralateral hindlimb nerves were less effective and the discharge generally occurred at a latency 1–2 msec longer than for ipsilateral nerves.Monosynaptic reflexes to pectoralis major and deep radial motoneurones supplying the physiological flexor muscles were strongly facilitated by hindlimb nerve stimulation, ipsilateral nerves being more effective than contralateral. Monosynaptic reflexes to latissimus dorsi showed a reciprocal pattern of conditioning, being depressed by ipsilateral and facilitated by contralateral hindlimb extensor nerves, the flexor nerves giving the reverse pattern. Monosynaptic reflexes to median and ulnar nerves supplying physiological extensor muscles were not significantly affected by hindlimb nerve conditioning.Polysynaptic reflexes to pectoralis major and deep radial motoneurones received initial strong facilitation followed by prolonged depression, ipsilateral hindlimb nerves being more effective than contralateral. In latissimus dorsi a reciprocal pattern similar to that for monosynaptic reflex testing was found. Polysynaptic reflexes to median and ulnar motoneurones received only prolonged depression.The hindlimb afferent nerves responsible for the discharge in forelimb motoneurones and for the facilitation and depression of forelimb reflexes include groups II and II muscle afferents and group II skin afferents, especially from quadriceps and sartorius muscles, and sural and superficial peroneal nerves, respectively.The ascending long propriospinal pathways are influenced bilaterally from hindlimb nerves and are located in the lower thoracic segments in the ventrolateral funiculus. The pathways mediate effects on ipsilateral and contralateral forelimb reflex systems, the ipsilateral projections being dominant. Part of the long ascending projection terminates monosynaptically on the motoneurones of pectoralis major. It is likely that group II afferents from ipsilateral quadriceps muscle activate the ascending tract monosynaptically and those from contralateral quadriceps disynaptically.The hypothesis is suggested that long propriospinal paths primarily represent intrinsic links between hindlimb and forelimb ‘motor centres’. The pattern of long ascending influences to groups of forelimb motoneurones corresponds closely to the sequences of hindlimb and forelimb stepping observed in normal cats. A functional role in stepping is therefore proposed for long ascending propriospinal pathways.     

Late prolonged discharges in the motor nerves of the hindlimbs and their relationship to locomotor rhythmicity in thalamically immobilized cats]

Effects of flexor reflex afferents stimulation were investigated on high decerebrated curarized cats. Stimulation of ipsilateral flexor reflex afferents evoked late long-lasting discharges in flexor nerves. Contralateral flexor reflex afferents stimulation evoked late discharges both in extensor and flexor nerves. Transition from late discharges to rhythmic discharges was observed. Early segmental reflexes were tonically depressed in thalamic in comparison with acute spinal cats. A similar tonic depression of segmental reflexes took place in acute spinal cats after DOPA injection. Segmental reflexes were distinctly modulated during late and rhythmic discharges. On the basis of the data available possible central mechanisms of the observed changes of segmental reflexes are discussed.     

Effect of knee joint angle on a heteronymous Ib reflex in the human lower limb

Altered efficacy, from change in receptor discharge with different positions of the knee, was investigated in a heteronymous Ib reflex of the human leg. The electrical stimulus was low threshold, to the common peroneal nerve. The divergence of the group I afferents was studied in the electromyograms (EMGs) of ipsilateral and contralateral thigh muscles. The stimulus evoked ipsilateral, short latency, excitation in the three quadriceps muscles studied and inhibition in the knee flexor semitendinosus (ST), with prior contraction of target muscles. This excitation and inhibition did not alter when studied over the range of the knee joint. The stimulus did not evoke responses in contralateral thigh muscles, contracted or relaxed. It is concluded that (1) any change in convergence from discharge of receptors, during extension of the limb, is small and sub-threshold, and (2) this spinal proprioceptive level of neural control appears to be directed primarily to the single limb.     

Convergent effects from vestibulospinal tract and primary cutaneous afferent fibers on motoneurons to proximal lower limb flexor and extensor muscles in humans

In human subjects, the latency and magnitude of cutaneomuscular test reflexes, evoked by stimulation of the sural nerve in the contralateral and ipsilateral quadricep and hamstring muscles, were investigated during static tilts in the pitch axis. The ANOVA demonstrated a highly significant, tilt-dependent modulation of the magnitude of the test reflex for the initial phase of both ipsilateral quadriceps and ipsilateral hamstrings. These results are consistent with results from a similar study on human ankle muscles and provide further evidence that the tilt-dependence of these cutaneomuscular reflexes originates from activity in the otolith receptors.     

Distal hind-limb reflex responses in cats with largely intact spinal reflex circuits

Reflex responses to single shocks to peripheral nerves were studied in the small muscles of the hind limb of the decerebrated and subsequently spinalized cat. Monosynaptic and high threshold polysynaptic reflexes of plantar muscles evoked by tibial nerve shocks were significantly more active in preparations that had extensive dissection of nerves and muscles. Monosynaptic reflexes were more active in the plantar flexor portion than in other parts of the interosseus. Compared with proximal limb muscles, this muscle showed features of reflex behavior that were neither consistently flexor nor extensor. In spinal animals the threshold of the tibialis anterior to tibial nerve shocks was often less than the threshold for other muscles, suggesting a function other than nociceptive flexion reflex, probably tactile influence in gait. Extensor digitorum brevis reflexes did not exhibit the flexor reflex pattern typified by those of the tibialis anterior muscle. The data further suggest activation of these reflexes by Ib interosseus muscle afferents and plantar cutaneous afferents and tonic inhibition from Ia receptors.     

Contribution of joint and cutaneous afferents to longer-latency reflexes in man

Electromyographic records from wrist extensors and flexors show a short latency reflex response, M1 and a longer latency response comprising of M2 and M3 peaks. M1 corresponds to the spinal stretch reflex and hence mediated by spindle afferents. In order to determine the contribution of various afferent types to M2-M3 components and simple reaction times, reflexes were elicited before and after anaesthetic blocks of palm cutaneous and wrist joint afferents in human subjects. The results show that joint and cutaneous afferents have no significant contribution to the longer latency reflexes or simple reaction times.     

Effect of continuous vibration on nociceptive flexor reflexes

The effects of continuous segmental vibration on the biceps femoris (BF) nociceptive flexor reflexes elicited by painful electrical stimulation of the sural nerve at the ankle were investigated in 25 normal subjects. During vibration of 100 Hz frequency, applied either on ipsilateral or contralateral foot skin, the nociceptive BF flexor reflexes increased in amplitude. Marked facilitation was prolonged even 20 minutes after vibration ceased. Pain sensation described by subjects did not change significantly except that radiation of pain was reduced in some cases. The results could not be explained by current views of reflex sensitisation, dishabituation, or the gate control theory.     

Electrophysiological evidence for crossed oligosynaptic trigemino-facial connections in normal man.

A crossed short latency component (R1) of the human blink reflex could be elicited in orbicularis oculi muscles to stimulation of the contralateral supraorbital nerve, when infraliminal conditioning stimuli were applied to various cutaneous afferents of the body (facial, upper and lower limbs). The crossed R1 responses appeared when the time interval between the conditioning and the test stimuli was of 30 to 40 ms, 50 to 65 ms and 95 to 110 ms for facial, upper and lower limbs afferents respectively. For the same time intervals, these conditioning volleys also exerted a facilitatory effect on the ipsilateral R1 responses. Furthermore, crossed R1 responses were also obtained during supraspinal facilitation induced by a voluntary contraction of the eyelids. These data show that crossed oligosynaptic trigemino-facial reflex connections exist in normal subjects, which become functional when adequate conditioning stimuli are available.     

Interaction of nociceptive and non-nociceptive cutaneous afferents from foot sole in common reflex pathways to tibialis anterior motoneurones in humans

In six healthy subjects, the reflex responses of the tibialis anterior muscle (TA) to stimulation of the cutaneous afferents arising from plantar foot, were studied at rest and during different levels of steady voluntary contraction of the TA. At rest, the threshold of the response and the threshold of subjective pain sensation coincided. The mean latency of this TA nociceptive response was 84.7 ms. Steady voluntary contractions of the TA, which was increased progressively from 3% to 15% of the maximum voluntary contraction, produced a significant and parallel reduction in the threshold and latency of the response: at 15%, the mean latency was about 26 ms shorter than at rest and its threshold was about half (i.e. below the pain threshold). The conduction velocity of the afferents responsible for TA response at rest was within the range of A-δ pain afferents (mean 27.4 m/s), whereas during voluntary contraction it was within the A-β fibre range (mean 45.1 m/s). This suggests that descending command makes the discharge of low-threshold, fast-conducting fibres sufficient for reflex activation of TA motoneurones (MNs). Central delay (about 4 ms) and MN recruitment order (according to the size principle) were found to be the same for both nociceptive and non-nociceptive TA reflex responses. Finally, experiments of spatial summation revealed an interaction between nociceptive and non-nociceptive inputs at a premotoneuronal level. It is therefore proposed that nociceptive and non-nociceptive cutaneous afferents arising from the foot sole use the same short-latency spinal pathway to contact TA MNs and that their relative contribution to its segmental activation is contingent upon descending command.     

Effects of ipsilateral and contralateral cervical phrenic afferents stimulation on phrenic motor unit activity in the cat

The phrenic-to-phrenic inhibitory reflex has been analyzed using recordings of activity of single C₅-phrenic motor units (PMUs). After ipsilateral or contralateral stimulation of the C₆-phrenic root, early and late PMUs exhibit a similar inhibition. After contralateral stimulation, the duration of the inhibition is smaller and the threshold stimulus is higher than the corresponding values observed after ipsilateral stimulation. The latency of the inhibition is similar for both stimulations. Hemispinalization, contralateral to the recording site, affects weakly the phrenic-to-phrenic reflex. We conclude that early and late PMUs receive a similar inhibitory input from phrenic afferents and that the inhibition observed after cervical phrenic nerve stimulation may involve spinal cord circuits.     

The critical importance of stimulus intensity in intraoperative monitoring for partial dorsal rhizotomy

During partial lumbosacral dorsal rhizotomy (PDR), intraoperative dorsal rootlet stimulation (drs) evokes motor responses, presumed to be reflexes, which are used to select rootlets for section. However, dr stimuli may also costimulate ventral root (vr) and evoke an M rather than a reflex response, the two being distinguishable only by comparison of response latencies after drs at two separate sites. In 15 consecutive spastic cerebral palsy patients undergoing PDR, we asked whether reflex and M responses were distinguishable on the basis of stimulus intensity (SI). For soleus H reflexes evoked by percutaneous tibial nerve stimulation, the SI for reflex afferents was usually subthreshold for exciting motor fibers. Similarly, for nerve roots, reflexes were evoked by drs at SIs generally less than that for M responses evoked by vr stimulation (vrs). In contrast, M responses evoked by drs required SIs that were on average 20 times greater. Finally, costimulation of contralateral vr after ipsilateral vrs occurred at SIs shown to evoke M responses after drs. We conclude that: (1) reflex and M responses evoked by drs are distinguishable on the basis of the required SI; and (2) drs employing SIs greater than that required for vrs evokes M rather than reflex responses due to costimulation of ipsilateral and contralateral vr. © 1996 John Wiley & Sons, Inc.     

The postnatal reorganization of primary afferent input and dorsal horn cell receptive fields in the rat spinal cord is an activity-dependent process

The dorsal horn of the spinal cord in the newborn rat is characterized by large cutaneous mechanoreceptive fields, a predominance of A‐fibre synaptic inputs and diffuse primary afferent A‐fibre projections, all of which are gradually reduced and refined over the first postnatal weeks. This may be partly responsible for the reduction in cutaneous flexion reflex sensitivity of rats over the postnatal period. Here we show that chronic, local exposure of the dorsal horn of the lumbar spinal cord to the NMDA antagonist MK801 from birth prevents the normal functional and structural reorganization of A‐fibre connections. Dorsal horn cells in spinal MK801‐treated animals, investigated at eight weeks of age by in vivo electrophysiological recording, had significantly larger cutaneous mechanoreceptive fields and greater A‐fibre evoked responses than vehicle controls. C‐fibre evoked responses were unaffected. Chronic MK801 also prevented the normal structural reorganization of A‐fibre terminals in the spinal cord. The postnatal withdrawal of superficially projecting A‐fibre primary afferents to deeper laminae did not occur in treated animals although C‐fibre afferent terminals and cell density in the dorsal horn were apparently unaffected. Spinal MK801‐treated animals also had significantly reduced behavioural reflex thresholds to mechanical stimulation of the hindpaw compared to naïve and vehicle‐treated animals, whereas noxious heat thresholds remained unaffected. The results indicate that the normal postnatal structural and functional development of A‐fibre sensory connectivity within the spinal cord is an activity‐dependent process requiring NMDA receptor activation.     

Axial muscle EMG responses evoked by cutaneous flank nerves in the female rat: effects of spinal transection, steroid hormones, and anesthesia.

In the lateral longissimus muscle (LL) of ovariectomized, female rats anesthetized with low surgical doses of urethane (1.0 g/kg), cutaneous reflexes with similar EMG and response patterns could be elicited from CNS-intact rats and from rats 24 h after complete thoracic spinal cord transection. The probability of eliciting a response to contralateral cutaneous nerve stimulation alone is much lower in rats with complete spinal transections compared to CNS-intact rats. For both CNS-intact and spinal-transected rats, responses to ipsilateral cutaneous nerve stimulation had a shorter latency and required significantly less current on average than responses to contralateral stimulation. The respective currents for eliciting threshold responses to ipsi- and contralateral stimulation are less for CNS-intact than spinal-transected rats. For both CNS-intact and spinal-transected rats, responses to bilateral cutaneous nerve stimulation were inconsistent in the same animal from run to run. With the variability of response at this anesthetic level, no consistent effects of progesterone (acute, i.v.) or estrogen (acute, i.v. and pretreatment, s.c.) were observed in spinal-transected rats. Intravenous progesterone reduced early, unilateral responses in CNS-intact rats anesthetized with 1.0 g of urethane/kg. For both CNS-intact and spinal-transected rats, additional anesthesia during EMG recording produced a gradual decline in response magnitude which could be recovered with a modest increase in stimulus intensity. However, spinal-transected rats appear to require less anesthesia to reduce comparable responses. The results suggest that supraspinal input is especially effective for facilitating contralateral cutaneous reflexes in back muscles, whereas it contributes more equally with afferent input and segmental circuitry to the efficacy of ipsilateral cutaneous reflexes.     

Synaptic activation of the interneurons of the thoracic portion of the spinal cord by reticulospinal fibers of the lateral funiculus

Synaptic processes evoked in various functional groups of thoracic interneurons (Th10,11) by stimulation of ipsi- and contralateral bulbar reticular formation were studied in anesthetized cats with lesions of the spinal cord that remained intact only the ipsilateral funiculus. Activation of reticulospinal fibres of the lateral funiculus with conduction velocities of 30-100 m/s evoked monosynaptic EPSPs in the following types of cells tested: monosynaptically excited by group la muscle afferents; excited by flexor reflex afferents; excited mainly by descending systems; excited by low-threshold cutaneous afferents to a less extent. All these neurons with responses to reticular stimulation were located predominantly in the central and lateral regions of Rexed's lamina VII. Most of the cells in the dorsal horn were not affected by short-latency reticulofugal influences. The only exception were 6 neurons located in the horn most dorsal laminae. Functional organization of connections between the lateral reticulospinal pathways and spinal neurons is discussed as compared to that of medial reticulo-spinal pathways as well as to the organization of "lateral" descending systems: cortico- and rubro-spinal.     

Modulation of the withdrawal reflex during hemiplegic gait: effect of stimulation site and gait phase.

OBJECTIVE: The objective of the study was to investigate the sensitivity of the nociceptive withdrawal reflex to stimulation of different locations on the sole of the foot during hemiplegic gait. METHODS: Reflexes were evoked by cutaneous electrical stimulation of 4 locations on the sole of the foot of 7 hemiplegic and 6 age-matched healthy persons. The stimuli were delivered at heel-contact, during foot-flat, at heel-off, and during mid-swing. Reflexes were recorded from muscles of the stimulated and the contralateral leg. Ankle, knee, and hip joints angles were recorded using goniometers. RESULTS: In the hemiplegic persons, the size of tibialis anterior reflexes, and the latency of soleus reflexes were site- and phase-modulated. In both groups, the tibialis anterior reflexes were significantly smaller with stimulation to the fifth metatarsophalangeal joint and the heel compared with the first metatarsophalangeal joint and the arch of the foot. The tibialis anterior reflexes evoked at heel-off and mid-swing were larger in hemiplegic persons than in healthy persons. Reflexes in the proximal and contralateral limb muscles were not site-modulated during hemiplegic gait. The kinematic response at the ankle joint was also different in the two groups during mid-swing. CONCLUSIONS: Hemiplegic and healthy middle-aged people presented different phase-modulation of the kinematic and muscle nociceptive reflex responses evoked by stimulation delivered on the sole of the foot. SIGNIFICANCE: The results have potential application in programs to rehabilitate hemiplegic gait.     

Response of brainstem trigeminal neurons to electrical stimulation of the dura

The extracellular response of medullary trigeminal neurons to electrical stimulation of the dura was studied in anesthetized cats. Fifty-six medullary trigeminal units were excited by stimulation sites near major dural vessels with an average latency of 11.0 ms. Many units also responded to infraorbital nerve shock and had cutaneous receptive fields that included the ipsilateral periorbital region. These cutaneous responses were either wide dynamic range or nociceptive specific in type. Electrical stimulation of the midbrain periaqueductal gray region suppressed the response of medullary trigeminal units to either dural stimulation or infraorbital nerve shock. Medullary trigeminal neurons that receive convergent inputs from dura and facial skin may provide a physiological substrate for the cutaneous referral of dural sensation.     

Crossed and uncrossed central effects of muscle spindle afferents from the lateral pterygoid muscle of the guinea pig

Physiological evidence is presented for the presence of stretch reflexes in the lateral pterygoid (Pt) muscle of the guinea pig. The central reflex effects of excitation of Pt stretch reflex afferents were also investigated. Passive lateral jaw displacement, which resulted in stretch of the Pt muscle on the side of jaw movement and stretch of the zygomatico-mandibularis (Z) muscle on the side contralateral to the movement, evoked increased EMG activity in these muscles. Stimulation of the trigeminal mesencephalic nucleus (mes V) evoked monosynaptic reflexes in both the Pt and Z nerves. Tonic stretch of the Pt muscle facilitated the monosynaptic reflex in the Pt nerve evoked by stimulation of mes V. Tonic vibration of the Pt muscle facilitated the mes V evoked monosynaptic reflex in the nerves to the ipsilateral Pt and contralateral Z muscles. Conversely, tonic vibration of the Z muscle facilitated the monosynaptic reflex evolved by mes V stimulation in the contralateral Pt and ipsilateral Z nerve. The results support the view that muscle spindles exist in the Pt and Z muscles and that there is a monosynaptic stretch reflex for both the Pt and Z muscles with cell bodies located in the mes V nucleus. It was also shown that the ipsilateral Pt muscle and the contralateral Z muscle act as synergists in the production of lateral jaw movements and that the organization of the stretch reflexes originating from the Pt and Z muscles support their synergistic action.     

Flexor reflexes in chronic spinal cord injury triggered by imposed ankle rotation

Hypersensitivity of the flexor reflexes to input from force-sensitive muscle afferents may contribute to the prevalence and severity of muscle spasms in patients with spinal cord injuries. In the present study, we triggered flexor reflexes with constant-velocity ankle movements into end-range dorsiflexion and plantarflexion positions in 8 individuals with spinal cord injuries. We found that all 8 subjects had coordinated increases in flexion torque at the hip and ankle following externally imposed plantarflexion movements at the ankle. In addition, end-range dorsiflexion movements also triggered flexor reflexes in 3 subjects, although greater loads were required to trigger such reflexes using dorsiflexion movements (compared to plantarflexion movements). These three-joint reflex torque patterns triggered by ankle movement were broadly comparable to flexion withdrawal responses elicited by electrocutaneous stimuli applied to a toe, although the amplitude of the torque response was generally lower. We conclude that excitation of muscle and joint-related afferents induced by end-range movements may be responsible for exaggerated flexion reflex responses in spinal cord injury.     

A functional study of uncrossed and crossed pulmonary afferent fibres in the cervical vagus nerves of the cat

The functional distribution of uncrossed and crossed pulmonary afferent fibres in the cervical vagus nerves has been studied in the anaesthetized cat using acute and chronic unilateral pneumonectomized preparations. The heart and lungs were sympathectomized routinely. The vagal afferent pathways of three pulmonary reflexes were investigated: the Hering-Breuer respiratory reflex, the lung inflation cardio-accelerator reflex, and the pulmonary chemoreflex. Inflation of the remaining lung caused temporary inhibition of inspiration. It also resulted in acceleration of the heart, but only when the background cardiac vagal tone was augmented. These respiratory and cardiac responses were abolished in most animals by ipsilateral cervical vagotomy; however, in some, a small response persisted and this was abolished by contralateral vagotomy. Stimulation of pulmonary C-fibre endings with right atrial injections of phenylbiguanide caused a reduction in respiration, bradycardia and systemic hypotension, responses which occurred with a latency of 2.9 +/- 0.15 s. They were mostly abolished by ipsilateral cervical vagotomy, but reduced responses persisted in a few animals. The residual responses were abolished by contralateral cervical vagotomy and by selective denervation of the lung. These results indicate that most afferent fibres subserving the three pulmonary reflexes studied run in the ipsilateral cervical vagus, representing the uncrossed pathway. Some afferent fibres, however, cross to the contralateral cervical vagus. Degenerative changes in cells of the contralateral nodose ganglion in chronic unilateral pneumonectomized animals support these findings.