The reliability of commonly used electrophysiology measures.

BACKGROUND: Electrophysiological measures can help understand brain function both in healthy individuals and in the context of a disease. Given the amount of information that can be extracted from these measures and their frequent use, it is essential to know more about their inherent reliability. OBJECTIVE/HYPOTHESIS: To understand the reliability of electrophysiology measures in healthy individuals. We hypothesized that measures of threshold and latency would be the most reliable and least susceptible to methodological differences between study sites. METHODS: Somatosensory evoked potentials from 112 control participants; long-latency reflexes, transcranial magnetic stimulation with resting and active motor thresholds, motor evoked potential latencies, input/output curves, and short-latency sensory afferent inhibition and facilitation from 84 controls were collected at 3 visits over 24 months at 4 Track-On HD study sites. Reliability was assessed using intra-class correlation coefficients for absolute agreement, and the effects of reliability on statistical power are demonstrated for different sample sizes and study designs. RESULTS: Measures quantifying latencies, thresholds, and evoked responses at high stimulator intensities had the highest reliability, and required the smallest sample sizes to adequately power a study. Very few between-site differences were detected. CONCLUSIONS: Reliability and susceptibility to between-site differences should be evaluated for electrophysiological measures before including them in study designs. Levels of reliability vary substantially across electrophysiological measures, though there are few between-site differences. To address this, reliability should be used in conjunction with theoretical calculations to inform sample size and ensure studies are adequately powered to detect true change in measures of interest.

Effects of afferent stimulation rate on inhibitory spinal pathways in hemiplegic spastic patients.

OBJECTIVE: It has recently been demonstrated in the cat and in healthy subjects that the effects of repetitive afferent fibre stimulation depends on the target spinal neurones. The purpose of this series of experiments was therefore to determine whether central nervous system lesions modify the behaviour of the inhibitory spinal pathways in response to repetitive activation of afferent fibres. METHODS: The H-reflex technique was used to study the effect of increasing the conditioning stimulus rate from 0.16 to 1 Hz on disynaptic inhibition and on presynaptic Ia inhibition on the affected side of 36 hemiplegic patients. RESULTS: The major finding was that, similar to results previously obtained in healthy subjects, increasing the conditioning stimulus rate in hemiplegic patients leads to an increase in the synaptic efficiency of inhibitory spinal circuits. Furthermore, a significant correlation was found between the severity of flexor carpi radialis muscle spasticity and the amount of disynaptic inhibition. CONCLUSIONS: The reinforcement of inhibitory spinal networks induced by repetitive stimulation of afferent fibres is preserved in spastic patients, whereas the mechanisms underlying this phenomena might be altered. SIGNIFICANCE: The results of these experiments open up a number of possibilities for novel spasticity therapies based on non-invasive techniques.

Effects of galvanic mastoid stimulation in seated human subjects.

The vestibular responses evoked by transmastoid galvanic stimulation (GS) in the rectified soleus electromyogram (EMG) in freely standing human subjects disappear when seated. However, a GS-induced facilitation of the soleus monosynaptic (H and tendon jerk) reflex has been described in few experiments in subjects lying prone or seated. This study addresses the issue of whether this reflex facilitation while seated is of vestibulospinal origin. GS-induced responses in the soleus (modulation of the rectified ongoing EMG and of the monosynaptic reflexes) were compared in the same normal subjects while freely standing and sitting with back and head support. The polarity-dependent biphasic responses in the free-standing position were replaced by a non-polarity-dependent twofold facilitation while seated. The effects of GS were hardly detectable in the rectified ongoing voluntary EMG activity, weak for the H reflex, but large and constant for the tendon jerk. They were subject to habituation. Anesthesia of the skin beneath the GS electrodes markedly reduced the reflex facilitation, while a similar, although weaker, facilitation of the tendon jerk was observed when GS was replaced with purely cutaneous stimulation, a tap to the tendon of the sternomastoid muscle, or an auditory click. The stimulation polarity independence of the GS-induced reflex facilitation argues strongly against a vestibular response. However, the vestibular afferent volley, insufficient to produce a vestibular reflex response while seated, could summate with the GS-induced tactile or proprioceptive volley to produce a startle-like response responsible for the reflex facilitation.

The disynaptic group I inhibition between wrist flexor and extensor muscles revisited in humans.

The present studies are designed to further characterise the interneuronal pathway mediating the disynaptic reciprocal group I inhibition between flexors and extensors at the wrist and the elbow levels in humans. In the first series of experiments, we compared the electrical threshold of the reciprocal group I inhibition at the wrist and the elbow level after a prolonged vibration aimed at raising the electrical threshold of the antagonistic activated Ia afferents. Prolonged vibration to the 'conditioning' tendon, which raised significantly the electrical threshold of the inhibition at the elbow level, did not alter it at the wrist level. These results suggest that the dominant input to the relevant interneurones is Ia in origin at the elbow level but Ib in origin at the wrist level. In the second series of experiments, using the spatial facilitation method, we compared the effects on the post-stimulus time histograms of single voluntarily activated motor units of two volleys delivered both separately and together to group I afferents in the nerves supplying the homonymous and antagonistic muscles. At the wrist, but not at the elbow level, the peak of homonymous monosynaptic group I excitation was reduced on combined stimulation, although the antagonistic IPSP was just at the threshold. Because the suppression did not involve the initial bins of the peak, it is argued that the suppression is not due to presynaptic inhibition of Ia terminals, but probably reflects convergence between the homonymous and antagonistic volleys onto the interneurones mediating the disynaptic inhibition. Taken together with the previously reported effects of recurrent inhibition on reciprocal inhibition, these results suggest that inhibition between flexors and extensors is differently organised at the elbow (reciprocal Ia inhibition) and the wrist (non-reciprocal group I inhibition) levels. It is argued that the particular connectivity at the wrist level might correspond to some functional requirements at this ball joint.

Post-activation depression in various group I spinal pathways in humans.

This investigation was designed to study the effects of post-activation depression in different spinal pathways fed by group I afferents available to investigation in human subjects. It was precipitated by a recent investigation in the cat showing that-contrary to the general assumption-post-activation depression is not a widespread phenomenon in the spinal cord. In 24 healthy subjects comparison was made between the effects of low and high-test stimulus rates on the monosynaptic Ia excitation, known to be subject to post-activation depression, and on oligosynaptic pathways fed by group I afferents. Both the amplitude of monosynaptic H reflexes and the amount of heteronymous monosynaptic Ia facilitation were significantly smaller at high than at low-test stimulus rates (1-2 s compared with 6-8 s between two consecutive stimuli). So was the amount of reciprocal Ia inhibition of tibialis anterior motoneurones. In contrast, the amount of other non-monosynaptic group I effects directed to the same motor nuclei (peroneal-induced excitation of quadriceps motoneurones, disynaptic non-reciprocal group I inhibition of flexor carpi radialis motoneurones, and D1 inhibition of flexor carpi radialis and soleus H reflexes) were enhanced at high stimulus rates. Results in humans confirm that post-activation depression depends on the type of group I afferents, and/or on the target neurones. The functional significance of the discrepancy between post-activation depression in pure Ia pathways and in other group I pathways is discussed with regard to the fusimotor-driven servo-assistance from Ia afferent discharges.

Evidence for recurrent inhibition of reciprocal inhibition from soleus to tibialis anterior in man.

Reciprocal inhibition between ankle flexors and extensors has been the subject of numerous studies in Man. They have demonstrated that this reciprocal inhibition is in all likelihood caused by a disynaptic pathway at least partly fed by Ia afferents. It is thus generally agreed that this reciprocally organized inhibition between ankle flexors and extensors in Man is similar to the reciprocal Ia inhibition described in the cat. This conclusion has, however, been challenged, when Jankowska and McCrea described in the cat a non-reciprocal group I inhibition involving interneurones co-excited by Ia and Ib afferents and mediating inhibition to both antagonistic and non-antagonistic motoneurones. The only way to distinguish between reciprocal Ia inhibition and non-reciprocal group I inhibition is to test if the inhibition is blocked by recurrent inhibition, since only Ia interneurones are inhibited by recurrent inhibition. In the present study, reciprocal inhibition from soleus to tibialis anterior was thus investigated following activation of soleus-coupled Renshaw cells in normal human subjects. It was found that reciprocal inhibition induced in tibialis anterior motoneurones by the activation of soleus group I afferents is deeply depressed by activation of soleus-coupled Renshaw cells. This finding provides the missing data to identify disynaptic inhibition between antagonistic ankle muscles as a reciprocal Ia inhibition.