- Email: [email protected]
Cutaneous afferents mediating the cutaneous silent period in the upper limbs: evidences for a role of low-threshold sensory fibres
Clinical Neurophysiology 112 (2001) 2007–2014 www.elsevier.com/locate/clinph
Cutaneous afferents mediating the cutaneous silent period in the upper limbs: evidences for a role of low-threshold sensory fibres Mariano Serrao a,*, Leoluca Parisi a, Francesco Pierelli b, Paolo Rossi a a
Istituto di Clinica delle Malattie Nervose e Mentali, II Clinica Neurologica, Universita` degli Studi di Roma ‘La Sapienza’, Viale dell’Universita` 3000185, Rome, Italy b IRCCS Neuromed, Pozzilli (IS), Italy Accepted 10 September 2001
Abstract Objectives: To evaluate the contribution of the low-threshold afferents to the production of the cutaneous silent period (CSP) in the upper limbs. Methods: The CSP was studied in 10 healthy adults and 4 patients with Friedreich’s ataxia. The following neurophysiological aspects were studied: (a) relationship between sensory threshold (ST), sensory action potential (SAP) amplitude and CSP parameters; (b) habituation and recovery cycle of the CSP at different stimulus intensities (2 £ ST and 8 £ ST); (c) pattern of responses in distal and proximal muscles at different stimulus intensities (2 £ ST and 8 £ ST). Results: (a) The CSP occurred at low intensities (1 £ ST and 2 £ ST) and increased abruptly between 3.5 £ ST and 4 £ ST (corresponding to the pain threshold). The SAP amplitude was saturated before CSP saturation. In the patients with Friedreich’s ataxia, the CSP appeared only at higher stimulus intensities (6 £ ST–8 £ ST). (b) The CSP evoked at 2 £ ST showed a fast habituation and slow recovery cycle whereas the opposite behaviour was found at 8 £ ST. (c) Low-threshold stimuli induced an inhibitory response restricted to the distal muscles. High-intensity stimulation produced an electromyographic suppression, significantly increasing from proximal to distal muscles. Conclusions: Our findings support the notion that low-threshold afferents participate in the production of the CSP in the upper limbs. The different afferents may activate different central neural networks with separate functional significance. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cutaneous silent period; Low-threshold afferents; Central neural networks
1. Introduction During sustained muscle contraction, activation of cutaneous fibres following electrical nerve stimulation produces a transient suppression of electromyographic (EMG) activity known as the cutaneous silent period (CSP) (Caccia et al., 1973; Uncini et al., 1991; Leis et al., 1994). It has been reported that, in order to evoke a CSP, a single cutaneous stimulus must be perceived as painful (Shefner and Logigian, 1993); moreover, several reports have shown that the afferents of the CSP are mainly mediated by slow conducting A-delta-type nociceptive fibres (Uncini et al., 1991; Leis et al., 1992; Shefner and Logigian, 1993). Thus, it has been suggested that the CSP evoked by high electrical stimulation of the digital nerves may be considered as part of the circuitry that mediates withdrawal flexor reflexes (Leis, 1998, Leis et al., 2000). * Corresponding author. Tel.: 139-649914815; fax: 139-64454294. E-mail address: [email protected] (M. Serrao).
It is also known that low intensity electrical stimulation of the digital nerves of the index finger produces complex inhibitory and excitatory responses on EMG activity (cutaneomuscular reflexes, CMRs) in the hand and forearm muscles (Caccia et al., 1973; Garnett and Stephens, 1980, Jenner and Stephens, 1982, Evans et al., 1989). The CMRs are thought to play a modulatory role in muscle activity during finger movements. Caccia et al. (1973), after studying the cutaneous reflexes of the hands, suggested that the I2 inhibitory response, whose latency and duration overlap the CSP, was mediated by low-threshold group 2 fibres. Since the afferents mediating the CSP have yet to be definitively established, further investigation in this direction should be undertaken. As the high stimulation intensity required to evoke the CSP activates both nociceptive high threshold fibres and a wide group of non-nociceptive low-threshold cutaneous fibres, the CSP may be produced by both high- and lowthreshold cutaneous sensory afferents. In this regard, the CSP elicited by the A-delta fibres (Group 3) may dissemble
1388-2457/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(01)00675-7
CLINPH 2001542
2008
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
a component of the CSP due to the low-threshold cutaneous sensory afferents (Group 2). In this study, we address the question of the contribution of the low-threshold afferents to the production of the CSP.
between 22 and 248C. Skin temperature was kept at 34– 368C by an automatically controlled infrared heating element. The same operator performed all the CSP measurements.
2. Methods
2.3. Relationship between sensory threshold (ST), SAP and SP
2.1. Subjects and patients Recordings were made from 10 healthy adults (7 M, 3 F), aged from 25 to 38 years, with written informed consent and local ethical committee approval. In order to understand better the role of large myelinated sensory fibres in the origin of the CSP, we studied 4 patients aged 24–35 years with Friedreich’s ataxia. All patients had a genetically confirmed diagnosis, absent deep tendon reflexes, loss of proprioceptive, vibratory, but preserved pain sensations and absent surface sensory action potentials (SAPs) in the upper limbs. 2.2. CSP measurements and data analysis The CSP was recorded during an isometric contraction of the thumb on a horizontal plane against a fixed bar, while cutaneous electrical stimuli were delivered to the index finger. Ring electrodes over the D2 interphalangeal joints (digital nerve) were used to apply 0.1, 0.5 and 1 ms constant current square wave electrical stimuli (stimulus intensity: 2– 60 mA). Voluntary EMG activity was recorded through standard surface Ag/AgCl electrodes from the abductor pollicis brevis (APB), extensor carpi (EC), flexor carpi (FC), triceps brachii (TB) and biceps brachii (BB). Sensitivity was set at 500– 1000 mV/div., with a 30–3000 Hz bandpass. Subjects had to maintain approximately 50% of the maximum voluntary isometric contraction of the target muscles with the aid of EMG acoustic and visual feedback from an oscilloscope screen displaying the force level. EMG activity from the target muscles was full-wave rectified and averaged over 10 trials for each condition recorded. The CSP was identified by a decrease in the mean rectified EMG activity lasting at least 10 ms in duration compared with a baseline level obtained during a 40 ms epoch preceding the stimulus. The CSP onset latency was determined by inspection of the rectified EMG at the point in the traces at which the average EMG amplitude dropped below 50% of pre-stimulus levels. The duration of the CSP was calculated from the stimulus artefact to the return to the EMG baseline value after the SP (S–X interval) and from the onset of the CSP latency to the point where the EMG activity amplitude returned above 50% of the pre-stimulus level. The area of the CSP (mV ms) was automatically computed between these two points. In two healthy subjects, we tested the effects of low-intensity stimuli (2 £ ST) during different levels of tonic muscular contraction (20, 40, 60, 80, 100% of the maximal force). During the study, room temperature was maintained at
The ST was measured by stimulation through the ring electrodes at one stimulus per second, while gradually changing the stimulus strength. The threshold was taken to be the stimulus voltage when the subject began to feel each stimulus distinctly, reported as a regular tapping sensation 5 times over 5 trials. The stimulus intensity was expressed in multiples of the ST perception intensity ( £ ST) up to a maximum of 12 times the threshold. Each subject’s sensory perceptual threshold (ST), determined at the beginning of each session, was approximately 3–4 mA at 0.1 ms, 2–3 mA at 0.5 ms and 2–2.5 mA at 1 ms stimulus duration. Subjects were asked to subjectively evaluate the nature of each stimulus intensity as not painful, mildly painful, moderately painful and strongly painful. By stimulating the index finger at a fixed ST multiple, we recorded the SAP from the median nerve at the wrist using bipolar electrodes. SAP amplitudes, SP latency and duration were normalised as a percentage of the corresponding maximum value. Stimulus intensity was expressed as a multiple of the ST and curves were calculated for the two functions i.e. stimulus intensity/SP duration and stimulus intensity/SAP amplitude (Inghilleri et al., 1997). 2.4. Habituation and recovery cycle of the CSP The excitability state of the neuronal substrate of CSP was studied by measuring the habituation and the recovery cycle of the EMG suppression in the APB. This experiment was conducted at low- and high-intensity levels (2 £ ST and 8 £ ST). Habituation was investigated by using a train of 20 stimuli at 3 Hz and recording the changes in the mean EMG area of the rectified and averaged EMG signal (habituation index ¼ last CSP/baseline CSP percentage). The recovery cycle was investigated by delivering paired stimuli of the same intensity at different interstimulus intervals: 500, 300, 250, 200, 150 and 100 ms. For each time interval, a series of 10 trials was repeated at 0.2 Hz frequency. To prevent fatigue and habituation, the subjects rested for 2 min at the end of each series. The recovery cycle was expressed by plotting the interstimulus interval on the X axis and the area of responses to the second shock (test) as a percentage of the area to the first shock (conditioning) on the Y axis (Cruccu et al., 1984). 2.5. Ischaemic test In order to identify the afferent fibres responsible for the observed effects, ischaemia was induced by a tourniquet
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
2009
Table 1 Relationship between CSP onset, SAP onset and pain threshold (stimulus duration 0.2 ms) Subjects
1 2 3 4 5 6 7 8 9 10
CSP onset (threshold units)
1£ 1£ 2£ 1£ 1£ 1£ 2£ 2£ 1£ 2£
SAP onset (threshold units)
2£ 1.4 £ 2£ 1.2 £ 1.5 £ 1.5 £ 1.5 £ 1.5 £ 2£ 2£
inflated above systolic blood pressure (220–250 mmHg) across the elbow proximally to the target APB muscle in 3 patients in which we were able to record an H-reflex from the FC. The CSP evoked at low (2 £ ) and high (8 £ ) stimulus intensity were evaluated at 10, 20 and 30 min. The Hreflex, SAP and cMAP from the median nerve were monitored continuously during the experiment, for approximately 30 min. Immediately after these measurements, the tourniquet was gradually released over a period of 2–3 min. The excitability parameters were remeasured 5–10 min after the starting of tourniquet deflation. 2.6. Statistical analysis Data were evaluated by analysis of variance (ANOVA) for repeated measurements. P , 0:05 were considered statistically significant. All values are reported as mean ^ SD. 3. Results 3.1. Relationship between ST, SAP and SP The relative stimulus intensities inducing the CSP by stimulation of the digital nerves are shown in Table 1. A stable reproducible CSP was identified at the perceptual threshold in 6 subjects for every stimulus duration (Fig. 1). At 2 £ ST, the CSP was detected in all subjects. In most subjects the SP was seen just before or at the point in which the SAP appeared. The pain threshold (mildly painful) ranged between 3 £ ST and 5 £ ST. Strongly painful threshold varied between 8 £ ST and 12 £ ST; at these stimulus intensities the CSP had plateaued. The effect of increasing stimulus intensity on the voluntary EMG activity of the APB and the effect on the SAP amplitude are shown in (Fig. 2). Both functions were sigmoidal but separated and differently shaped. The CSP required lower intensities to appear and increased abruptly between 3.5 £ ST and 4 £ ST. The SAP amplitude was saturated before CSP saturation (Fig. 2).
Pain threshold (threshold units) Mildly
Moderately
Strongly
4£ 3.5 £ 4£ 3£ 4.5 £ 3£ 5£ 3£ 3.5 £ 3.5 £
6£ 6£ 7£ 5£ 6.5 £ 5£ 7.5 £ 5£ 6£ 6£
10 £ 10 £ 9£ 9£ 10 £ 8£ 12 £ 8£ 9£ 8£
While the total duration of EMG inhibition increased, with increasing stimulus intensities that of the onset latency of the SP was reduced (Table 2). The S–X interval was not significantly modified by increasing stimulus intensities (Table 2). In both subjects in which the CSP at 2 £ ST was evaluated during different levels of contraction (20–100%), the duration of the SP was dependent on the intensity of the contraction. The SP was inversely related to the intensity of the contraction. The SP was present in both subjects up to 80% of the maximal force. At 100% of the maximal force, the CSP disappeared in the rectified and average EMG tracing. However, at this strength level the CSP was present in few traces of the raw EMG signal. The ST in the patients with Friedreich’s ataxia was 3– 5 mA. At 2 £ ST, no CSP was evoked in ABP muscle. The CSP appeared only at higher stimulus intensities perceived as painful (6 £ ST–8 £ ST). 3.2. Habituation and recovery cycle Repetitive stimulation at 2 £ ST disclosed, in every
Fig. 1. CSP elicited in APB muscle with stimulation of the index finger at 2 £ (upper trace) and 8 £ (lower trace) ST. Rectified and averaged EMG signal during isometric contraction at 50% of maximal strength. The effect of stimulus intensity is represented by a decrease in onset latency and an increase in the duration.
2010
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 2. Function stimulus 2 intensity/CSP duration vs. function stimulus 2 intensity/SAP amplitude in 10 healthy subjects. Stimulus intensity is expressed as a multiple of ST. Both SAP amplitude and CSP duration are expressed as a percentage of maximum values. The mean pain threshold is indicated by a vertical dashed line. Both functions were sigmoidal but separated and shaped differently. The CSP (triangles) required lower intensities to appear and increased abruptly between 3.5 £ ST and 4 £ ST. The SAP amplitude (squares) was saturated before CSP saturation. Bars, standard deviation. A schematic model of cutaneous afferents involved in CSP production, based on neurophysiological findings is given above the graph.
subject, a fast habituation of the inhibitory response in the APB (Fig. 3a). The CSP obtained in response to high intensity (8 £ ST) repetitive stimulation did not habituate (Habituation Index ¼ 1.09 ^ 0.3) (Fig. 3b). The results of the recovery cycle varied according to the threshold stimuli (Fig. 4). High intensity paired stimuli did not modify the test CSP. Low intensity paired stimuli induced a reduction of EMG inhibition at 100–150 interstimulus interval, recovering the baseline level of EMG suppression, from 200 up to 500 ms. 3.3. Pattern of responses in distal and proximal muscles after digital nerve stimulation The stimulation of the index finger evoked a different pattern of inhibitory effects in the proximal and distal muscles depending on the stimulus intensity. Low-threshold stimuli (2 £ ST) induced an inhibitory response in the ABP, EC and FC. No change in EMG activity of the BB and TB Table 2 Mean latency, duration and S–X interval of the CSP at different stimulus intensities (stimulus duration 0.2 ms) Stimulus intensity (threshold units)
Onset latency (ms) a
Duration (ms) a
S–X interval (ms)
1£ 2£ 3£ 4£ 6£ 8£
87.8 ^ 6.4 82.0 ^ 6.8 78.8 ^ 7.3 67.0 ^ 7.2 65.4 ^ 7.3 65.0 ^ 7.0
14.0 ^ 5.7 19.0 ^ 5.9 23.4 ^ 7.1 45.0 ^ 8.3 52.8 ^ 8.2 56.0 ^ 8.6
106.3 ^ 10.2 106.2 ^ 10.8 107.0 ^ 11.2 110.0 ^ 13.9 111.0 ^ 13.8 115.0 ^ 14.2
a The CSP onset latency significantly decreased and the duration significantly increased at higher stimulus intensities (ANOVA for repeated measures, P , 0:05).
was found (Table 3). High-intensity stimulation produced EMG suppression in the small hand muscles, a shorter CSP in the forearm muscles and an even shorter one in the TB (ANOVA, P , 0:01). No inhibitory response was found in the BB. 3.4. Ischaemic test The subjects in whom ischaemia was induced by the inflated tourniquet, experienced paraesthesiae in the hand during the first minutes of the ischaemic period. Touch and deep sensitivity were reduced earlier than pain– temperature sensation. Change in low-threshold CSP, SAP and H-reflex, paralleled touch and deep sensation. A 50% reduction of the low-threshold CSP duration occurred after 10 min. At 20 min of the ischaemic period the low-threshold CSP, the H-reflex and SAP were abolished (Fig. 5), whereas the cMAP and the high-threshold CSP decreased (highthreshold CSP reduction ranged between 25 and 30%). At the end of the ischaemic period (30 min) the block was virtually complete. At 5 min after release of the cuff lowthreshold CSP was fully restored in both subjects.
4. Discussion 4.1. Peripheral afferents mediating the cutaneous SP There is a significant amount of evidence which suggest that afferent impulses generating the CSP are carried by Adelta fibres (Uncini et al., 1991; Leis, 1992; Shefner and Logigian, 1993; Leis et al., 1994; Inghilleri et al., 1997). These fibres, which range in diameter from 1 to 6 mM, have a higher threshold for activation than large fibres and
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 3. (A) Raw EMG activity recorded from APB muscle after trains of stimuli delivered at 3 Hz and low stimulus intensity (2 £ ST). The CSP promptly habituates. (B) Raw EMG activity recorded from APB muscle after trains of stimuli delivered at 3 Hz and high-stimulus intensity (8 £ ST). The CSP does not show habituation.
mediate impulses resulting from nociceptive stimuli. (Uncini et al., 1991; Leis et al., 1994). However, the possibility that large-diameter afferents may help to induce the CSP cannot be dismissed. Caccia et al. (1973), after studying the cutaneous reflexes of the hands, suggested that the inhibitory response (I2), whose latency and duration overlap the CSP, was already present when the stimulus was at the perceptual threshold level and was mediated by low-threshold group 2 fibres. Recently, Syed et al. (2000) assessed the CSP in patients with Fabry’s disease to verify whether smalldiameter afferents are responsible for producing the CSP. Their findings showed minor abnormalities in the CSP parameters, thus suggesting the possibility that large-diameter fibres may further contribute in producing the CSP. In our study, the CSP evoked at low stimulus intensity, already appeared at or just above the sensory perception
2011
threshold (1 £ ST–2 £ ST). The onset latency of the lowthreshold CSP ranged between 80 and 90 ms and the duration between 25 and 12 ms. Thus, a suppression of the EMG signal appeared at lower current levels than those required to evoke the SAP, as observed by Caccia et al. (1973). When the stimulus intensity was increased (from 3 £ ST to 12 £ ST), the onset latency of the CSP decreased, whereas the duration increased and the S–X interval remained stable. The stimulus intensity/function of the CSP (Fig. 2) became steeper between 3.5 £ ST and 5 £ ST, at the very point in which the stimulus was perceived as painful. These findings suggest that the CSP evoked by lowthreshold stimuli (up to 2 £ ST–3.5 £ ST), is mediated by large-diameter sensory fibres that do not convey painful impulses. This is confirmed by the fact that low-threshold CSP was absent in patients affected by Friedreich’s ataxia with no recordable SAPs. Additionally, the CSP evoked at low stimulus intensity early decreased after the ischaemic test indicating that type 2 afferents are involved in the origin of the CSP at these stimulus intensities. The suppressive effect on EMG of cutaneous stimulation increased markedly when the stimulus was perceived as painful, which indicates that in this condition the CSP is mainly mediated by A-delta fibres, as reported by several authors (Uncini et al., 1991; Shefner and Logigian, 1993, Inghilleri et al., 1997). The study of the stimulus/intensity function for SAP amplitude and CSP duration showed an almost parallel increase in both parameters, suggesting that the contribution of group 2 fibres may not be limited to the lower stimulus intensities and is probably masked by the recruitment of the group 3 fibres at higher intensities. The observed reduction of high-threshold CSP duration during the ischaemic block at the time in which the low-threshold CSP disappeared, further supports the contribution of the type 2 fibres in the generation of the high-threshold CSP. As the S–X interval was not significantly modified by the current levels, the effect of the nociceptive fibres was to shorten the latency of the CSP. In this regard, group 2 fibres may play a role in inducing the later part of the CSP, while group 3 afferents may be responsible for the earlier part. As mentioned earlier, many authors, though not all, have failed to find marked, stable EMG suppression after low-intensity cutaneous stimulation (Uncini et al., 1991; Inghilleri et al., 1997; Leis, 1994; Leis et al., 2000). These discrepancies are difficult to explain. It may be that, as the CSP is not a wellstandardised technique, the variability in the results obtained in different laboratories is due to different stimulation, recording and measurement paradigms. For instance, Uncini et al. (1991) computed the CSP in 3 different time windows, inappropriate for identifying a short duration CSP as that evoked by low-threshold stimuli. In fact, to obtain the SP from the averaged trials, Uncini used an off-line analysis program calculating the integrated EMG activity during 4 time periods in the EMG tracing. The first period
2012
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 4. Mean recovery cycles of the CSP evoked by paired shocks at high (8 £ ST, black triangles) and low (2 £ , squares) stimulus intensity in 10 subjects. High intensity paired stimuli did not modify the CSP test. Low intensity paired stimuli lowered of EMG inhibition at a 100–150 interstimulus interval, recovering the baseline level of suppression from 200 up to 500 ms.
consisted of 100 ms of baseline activity (PRE) from the prestimulus epoch. The other 3 periods, taken from the poststimulus epoch, consisted of the SP, 50 ms before (PI) the SP and 50 ms after (PIII) the SP itself. As a confirmation for the determination of the SP by visual inspection of the rectified averaged EMG tracings, Uncini considered an SP definitely present when the SP/PI ratio was less than 0.6 (mean value of SP/PI 1 3 SD). It is obvious that the SP period measured by Uncini totally corresponds to the EMG suppression activity evoked at high intensity of stimulation, starting at 50–60 ms after the stimuli and lasting 50 ms. In this view, he found a very significant value from the SP/PI ratio (0.29 ^ 0.16). In our study, the CSP evoked at low-threshold stimuli beginning at 80–90 ms corresponds to the end of the SP period reported by Uncini. If we consider that SP evoked at low-threshold lasts 14–19 ms, there will be a drawback in the SP/PI ratio for this reflex. In fact, in this case the low-threshold CSP could correspond to both SP and PIII period reported by Uncini.
With regard to the specific aim of this paper, several technical factors have to be taken into account. First, an occlusion in the excitatory pathways between descending inputs, peripheral feedback activity (during tonic contraction) and the peripheral volley might explain the difficulty in recording inhibition of the muscle activity on the raw EMG signal, after low-intensity stimuli or during excessive muscle contraction. Thus, the visual inspection of the raw EMG signal as used by Shefner and Logigian (1993) may not be suitable for the measurement of the CSP. Moreover, when few trials are averaged and full-wave rectified during intense muscular contraction, as done by Inghilleri et al. (1997), low-level afferent inputs may collide with an intense excitatory output and consequently fail to show significant inhibition on the EMG activity. This is supported by our findings of an inverse relationship between CSP duration
Table 3 Mean latency, duration and S–X interval of the CSP elicited by index finger stimulation in different muscles of the upper limbs at different stimulus intensities (stimulus duration 0.2 ms) a Muscle
BB TB FC EC APB
CSP onset latency (ms)
CSP duration (ms)
2 £ ST
8 £ ST
2 £ ST
8 £ ST b
– – 77.2 ^ 7.3 78.8 ^ 7.6 82.0 ^ 6.8
– 59.4 ^ 3.4 61.5 ^ 2.9 62.7 ^ 3.1 65.0 ^ 7.0
– – 18.1 ^ 6.1 19.2 ^ 6.4 19.0 ^ 5.9
– 20.4 ^ 15.6 41.7 ^ 12.3 42.1 ^ 13.2 56.0 ^ 8.6
a BB, biceps brachii; TB, triceps brachii; FC, flexor carpi; EC, extensor carpi; APB, abductor pollicis brevis. b The CSP significantly increased from proximal to distant muscles.
Fig. 5. Low-threshold (2 £ ST) CSP modifications induced by the ischaemic test in a representative patient. After 10 min (upper trace), the CSP duration is slightly reduced. After 20 min (middle trace) the CSP duration is drastically reduced and no consistent inhibition is recorded 30 min after the beginning of the test (lower trace).
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
and muscular contraction levels. Second, the low-threshold CSP easily habituates; both the frequency of stimulation and the number of trials become critical for a correct evaluation of the CSP. Therefore, the relationship between stimulus intensity and the CSP is far from being elucidated, even if we consider that some inhibition of tonic EMG activity can be elicited simply by fingertip tapping (Caccia et al., 1973). In our opinion, since the publication of findings of Uncini et al. in 1991, research on CSP has focused excessively on the nocifensive nature of this reflex, disregarding the effects of low-intensity stimuli. 4.2. Central neural pathways mediating the CSP It is generally accepted that the CSP mediated by A-delta fibres is an oligosynaptic inhibitory spinal reflex (Uncini et al., 1991; Leis et al., 1995; Inghilleri et al, 1997; Logigian et al., 1999). In fact, the long latency of this reflex (60–70 ms) is due to a long afferent conduction time rather than a long central delay (Manconi et al., 1998). As the central processing time (CPT) for this reflex has been estimated to be about 15 ms (Shefner and Logigian, 1993), it is unlikely to involve the activation of brainstem or cortical structures. Moreover, Logigian et al. (1999) showed that an inhibitory effect corresponding to the CSP occurs in the lower limbs of patients with complete spinal cord injury. By contrast, since the cutaneous afferents that mediate the low-threshold CSP belong to group 2 fibres conducting at about 50 ms 21, the longer onset latency of this reflex can only be due to a longer central delay. Considering that the CPT may be estimated by the following formula CPT ¼ CSP latency 2 efferent time ([F 1 M]/2) 2 afferent time (distance between index finger to C6 root/conduction velocity of afferent fibres) (Shefner and Logigian, 1993), the central delay for the low-threshold CSP is about 55 ms. This is a feasible period of time for a polysynaptic circuit, such as brainstem spinobulbospinal or transcortical circuits. The central neural substrates producing alpha-motor neuron inhibition may be investigated by studying the habituation and recovery cycle using the double stimulus technique (Kimura, 1973; Cruccu et al., 1984; Inghilleri et al., 1997). With regard to the CSP evoked by high-intensity stimuli, we obtained results similar to those reported by other authors (Uncini et al., 1991; Inghilleri et al., 1997) showing a lack of habituation and fast recovery cycle. These findings suggest that the underlying neuronal network of the CSP is oligosynaptic. The CSP evoked by non-painful stimuli, on the contrary, habituated and exhibited a slow recovery cycle. This pattern is typical of polysynaptic exteroceptive reflexes (Cruccu et al., 1984; Cruccu et al., 1991; Priori et al., 1998). Thus, our findings provide evidence that the CSP evoked by low-intensity electrical finger stimulation is mediated by group 2 fibres, possibly by activating transcortical or brainstem polysynaptic circuits inhibiting the alpha-motor neurons.
2013
4.3. Relationship between the CSP and the CMRs The low-intensity CSP, as described in this paper, shows a behaviour similar to the so-called CMRs evoked by electrical non-painful stimulation of the fingers. Several researchers have reported that, in the upper limbs, lowintensity stimulation of the digital nerves of the index finger produces reflex responses in the hand muscle and in the forearm (Caccia et al., 1973; Jenner and Stephens, 1982, Evans et al., 1989). In the adult, this reflex response is characterised by several identifiable components comprising an initial short latency increase in EMG (the E1 component), followed by a decrease (the I1 component), followed by a second inhibition (the I2 component) and a second increase (the E2 component). The first two components are believed to be mediated by the activation of oligosynaptic spinal pathways, whereas the I2 and E2 components are thought to involve longer pathways, probably long loops extending to the cortex (Caccia et al., 1973; Jenner and Stephens, 1982; Issler and Stephens, 1983) Our data indicate that the CSP evoked by low-threshold fibres corresponds to the I2 component of the cutaneous muscular reflexes (Caccia et al., 1973; Brezinova, 1988). In fact, the I2 response share the same neurophysiological characteristics in terms of onset latency, effect of stimulus strength on reflex activity, habituation and distribution of reflex effects (Caccia et al., 1973). It is noteworthy that by increasing levels of digital stimulation EMG inhibition increases, while excitatory components decrease; the final result is that the I1 and I2 phases merge (Caccia et al., 1973; Clouston et al., 1995). On the basis of our data, we suggest that the CSP evoked by noxious stimuli may reflect a complex summation of the inhibitory components mediated by group 3 and group 2 fibres.
4.4. Functional significance of the CSP By studying the effects of finger stimulation on several muscles, we observed after low- and high-intensity stimulation, a common pattern of responses in distal and proximal muscles consisting of stronger reflex effects in the hand and forearm muscles, as reported by other authors (Caccia et al., 1973; Uncini et al., 1991; Inghilleri et al., 1997). This pattern of ‘distal’ inhibition may be interpreted from a functional viewpoint. As suggested by Leis (1998), the CSP mediated by Adelta fibres may be considered as part of the circuitry that mediates withdrawal flexor reflexes. The physiological role of the CSP evoked by low-threshold stimuli must be placed within the general pattern of excitability changes induced by the non-nociceptive cutaneous input in upper limb motor neurons in humans. The similarity between the low-threshold CSP and the I2 component of the CMRs suggests that the inhibitory response may
2014
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
help modulate the motorneuron activity which contributes to explorative movements of the hands. At the same time, sensory information from cutaneous large-diameter afferents may contribute to maintaining the excitability of the nocifensive reflex responses and thus improve the reflex gain in the different circumstances. In fact, a protective motor reaction to a nociceptive input would be accelerated and enhanced by a concomitant inhibition of the hands by other non-nociceptive afferent fibres. 5. Conclusions Our findings support the notion that low-threshold afferents participate in the production of the CSP. Further studies are needed to establish the relative contribution of the different classes of afferents generating the CSP. Nevertheless, by observing strict experimental paradigms, the study of the CSP at low- and high-intensity stimulation may provide additional information on different neural pathways, including the afferent sensory pathways and the central mechanisms modulating motor recruitment. References Brezinova V. Cutaneomuscular reflex in a peripheral nerve lesion. Electromyogr Clin Neurophysiol 1988;28:183–189. Caccia MR, McComas AJ, Upton RM, Blogg T. Cutaneous reflexes in small muscles of the hands. J Neurol Neurosurg Psychiatry 1973;36:960–977. Clouston PD, Kiers L, Menkes D, Sander H, Chiappa K, Cros D. Modulation of motor activity by cutaneous input: inhibition of the magnetic motor evoked potential by digital electrical stimulation. Electroenceph clin Neurophysiol 1995;97:114–125. Cruccu G, Agostino R, Fornarelli M, Inghilleri M, Manfredi M. Recovery cycle of the masseter inhibitory reflex in man. Neurosci Lett 1984;49:63–68. Cruccu G, Pauletti G, Agostino R, Berardelli A, Manfredi M. Masseter inhibitory reflex in movement disorders. Huntington’s chorea, Parkin-
son’s disease, dystonia, and unilateral masticatory spasm. Electroenceph clin Neurophysiol 1991;81:24–30. Evans AL, Harrison LM, Stephens JA. Task-dependent changes in cutaneous reflexes recorded from various muscles controlling finger movement in man. J Physiol (Lond) 1989;418:1–12. Garnett R, Stephens JA. The reflex response of single motor units in human first dorsal interosseous muscle following cutaneous afferent stimulation. J Physiol (Lond) 1980;303:351–364. Inghilleri M, Cruccu G, Argenta M, Polidori L, Manfredi M. Silent period in upper limb muscles after noxious cutaneous stimulation in man. Electroenceph clin Neurophysiol 1997;105:109–115. Issler H, Stephens JA. The maturation of cutaneous reflexes studied in the upper limb in man. J Physiol (Lond) 1983;335:643–654. Jenner JR, Stephens JA. Cutaneous reflex responses and their central nervous pathways studied in man. J Physiol (Lond) 1982;333:415–419. Kimura J. Disorders of interneurons in parkinsonism. The orbicularis oculi reflex to paired stimuli. Brain 1973;96:87–96. Leis AA. Conduction abnormalities detected by silent period testing. Electroenceph clin Neurophysiol 1994;93:444–449. Leis AA. Cutaneous silent period. Muscle Nerve 1998;21:1243–1245. Leis AA, Kofler M, Ross MA. The silent period in pure sensory neuronopathy. Muscle Nerve 1992;15:1345–1348. Leis AA, Stetkarova I, Beric A, Stokic DS. Spinal motor neuron excitability during the cutaneous silent period. Muscle Nerve 1995;18:1464–1470. Leis AA, Stokic DS, Fuhr P, Kofler M, Kronenberg MF, Wissel J, Glocker FX, Seifert C, Stetkarova I. Nociceptive fingertip stimulation inhibits synergistic motoneuron pools in the human upper limb. Neurology 2000;55:1305–1309. Logigian EL, Plotkin GM, Shefner JM. The cutaneous silent period is mediated by spinal inhibitory reflex. Muscle Nerve 1999;22:467–472. Manconi FM, Syed NA, Floeter MK. Mechanisms underlying spinal motor neuron excitability during the cutaneous silent period in humans. Muscle Nerve 1998;21:1256–1264. Priori A, Berardelli A, Inghilleri M, Pedace F, Giovannelli M, Manfredi M. Electrical stimulation over muscle tendons in humans. Evidence favouring presynaptic inhibition of Ia fibres due to the activation of group III tendon afferents. Brain 1998;121:373–380. Shefner JM, Logigian EL. Relationship between stimulus strength and cutaneous silent period. Muscle Nerve 1993;16:278–282. Syed NA, Sandbrink F, Luciano CA, Altarescu G, Weibel T, Schiffmann R, Floetr MK. Cutaneous silent period in patients with Fabry disease. Muscle Nerve 2000;23:1179–1186. Uncini A, Kujirai T, Gluck B, Pullman S. Silent period induced by cutaneous stimulation. Electroenceph clin Neurophysiol 1991;81:344–352.
Cutaneous afferents mediating the cutaneous silent period in the upper limbs: evidences for a role of low-threshold sensory fibres Mariano Serrao a,*, Leoluca Parisi a, Francesco Pierelli b, Paolo Rossi a a
Istituto di Clinica delle Malattie Nervose e Mentali, II Clinica Neurologica, Universita` degli Studi di Roma ‘La Sapienza’, Viale dell’Universita` 3000185, Rome, Italy b IRCCS Neuromed, Pozzilli (IS), Italy Accepted 10 September 2001
Abstract Objectives: To evaluate the contribution of the low-threshold afferents to the production of the cutaneous silent period (CSP) in the upper limbs. Methods: The CSP was studied in 10 healthy adults and 4 patients with Friedreich’s ataxia. The following neurophysiological aspects were studied: (a) relationship between sensory threshold (ST), sensory action potential (SAP) amplitude and CSP parameters; (b) habituation and recovery cycle of the CSP at different stimulus intensities (2 £ ST and 8 £ ST); (c) pattern of responses in distal and proximal muscles at different stimulus intensities (2 £ ST and 8 £ ST). Results: (a) The CSP occurred at low intensities (1 £ ST and 2 £ ST) and increased abruptly between 3.5 £ ST and 4 £ ST (corresponding to the pain threshold). The SAP amplitude was saturated before CSP saturation. In the patients with Friedreich’s ataxia, the CSP appeared only at higher stimulus intensities (6 £ ST–8 £ ST). (b) The CSP evoked at 2 £ ST showed a fast habituation and slow recovery cycle whereas the opposite behaviour was found at 8 £ ST. (c) Low-threshold stimuli induced an inhibitory response restricted to the distal muscles. High-intensity stimulation produced an electromyographic suppression, significantly increasing from proximal to distal muscles. Conclusions: Our findings support the notion that low-threshold afferents participate in the production of the CSP in the upper limbs. The different afferents may activate different central neural networks with separate functional significance. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cutaneous silent period; Low-threshold afferents; Central neural networks
1. Introduction During sustained muscle contraction, activation of cutaneous fibres following electrical nerve stimulation produces a transient suppression of electromyographic (EMG) activity known as the cutaneous silent period (CSP) (Caccia et al., 1973; Uncini et al., 1991; Leis et al., 1994). It has been reported that, in order to evoke a CSP, a single cutaneous stimulus must be perceived as painful (Shefner and Logigian, 1993); moreover, several reports have shown that the afferents of the CSP are mainly mediated by slow conducting A-delta-type nociceptive fibres (Uncini et al., 1991; Leis et al., 1992; Shefner and Logigian, 1993). Thus, it has been suggested that the CSP evoked by high electrical stimulation of the digital nerves may be considered as part of the circuitry that mediates withdrawal flexor reflexes (Leis, 1998, Leis et al., 2000). * Corresponding author. Tel.: 139-649914815; fax: 139-64454294. E-mail address: [email protected] (M. Serrao).
It is also known that low intensity electrical stimulation of the digital nerves of the index finger produces complex inhibitory and excitatory responses on EMG activity (cutaneomuscular reflexes, CMRs) in the hand and forearm muscles (Caccia et al., 1973; Garnett and Stephens, 1980, Jenner and Stephens, 1982, Evans et al., 1989). The CMRs are thought to play a modulatory role in muscle activity during finger movements. Caccia et al. (1973), after studying the cutaneous reflexes of the hands, suggested that the I2 inhibitory response, whose latency and duration overlap the CSP, was mediated by low-threshold group 2 fibres. Since the afferents mediating the CSP have yet to be definitively established, further investigation in this direction should be undertaken. As the high stimulation intensity required to evoke the CSP activates both nociceptive high threshold fibres and a wide group of non-nociceptive low-threshold cutaneous fibres, the CSP may be produced by both high- and lowthreshold cutaneous sensory afferents. In this regard, the CSP elicited by the A-delta fibres (Group 3) may dissemble
1388-2457/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(01)00675-7
CLINPH 2001542
2008
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
a component of the CSP due to the low-threshold cutaneous sensory afferents (Group 2). In this study, we address the question of the contribution of the low-threshold afferents to the production of the CSP.
between 22 and 248C. Skin temperature was kept at 34– 368C by an automatically controlled infrared heating element. The same operator performed all the CSP measurements.
2. Methods
2.3. Relationship between sensory threshold (ST), SAP and SP
2.1. Subjects and patients Recordings were made from 10 healthy adults (7 M, 3 F), aged from 25 to 38 years, with written informed consent and local ethical committee approval. In order to understand better the role of large myelinated sensory fibres in the origin of the CSP, we studied 4 patients aged 24–35 years with Friedreich’s ataxia. All patients had a genetically confirmed diagnosis, absent deep tendon reflexes, loss of proprioceptive, vibratory, but preserved pain sensations and absent surface sensory action potentials (SAPs) in the upper limbs. 2.2. CSP measurements and data analysis The CSP was recorded during an isometric contraction of the thumb on a horizontal plane against a fixed bar, while cutaneous electrical stimuli were delivered to the index finger. Ring electrodes over the D2 interphalangeal joints (digital nerve) were used to apply 0.1, 0.5 and 1 ms constant current square wave electrical stimuli (stimulus intensity: 2– 60 mA). Voluntary EMG activity was recorded through standard surface Ag/AgCl electrodes from the abductor pollicis brevis (APB), extensor carpi (EC), flexor carpi (FC), triceps brachii (TB) and biceps brachii (BB). Sensitivity was set at 500– 1000 mV/div., with a 30–3000 Hz bandpass. Subjects had to maintain approximately 50% of the maximum voluntary isometric contraction of the target muscles with the aid of EMG acoustic and visual feedback from an oscilloscope screen displaying the force level. EMG activity from the target muscles was full-wave rectified and averaged over 10 trials for each condition recorded. The CSP was identified by a decrease in the mean rectified EMG activity lasting at least 10 ms in duration compared with a baseline level obtained during a 40 ms epoch preceding the stimulus. The CSP onset latency was determined by inspection of the rectified EMG at the point in the traces at which the average EMG amplitude dropped below 50% of pre-stimulus levels. The duration of the CSP was calculated from the stimulus artefact to the return to the EMG baseline value after the SP (S–X interval) and from the onset of the CSP latency to the point where the EMG activity amplitude returned above 50% of the pre-stimulus level. The area of the CSP (mV ms) was automatically computed between these two points. In two healthy subjects, we tested the effects of low-intensity stimuli (2 £ ST) during different levels of tonic muscular contraction (20, 40, 60, 80, 100% of the maximal force). During the study, room temperature was maintained at
The ST was measured by stimulation through the ring electrodes at one stimulus per second, while gradually changing the stimulus strength. The threshold was taken to be the stimulus voltage when the subject began to feel each stimulus distinctly, reported as a regular tapping sensation 5 times over 5 trials. The stimulus intensity was expressed in multiples of the ST perception intensity ( £ ST) up to a maximum of 12 times the threshold. Each subject’s sensory perceptual threshold (ST), determined at the beginning of each session, was approximately 3–4 mA at 0.1 ms, 2–3 mA at 0.5 ms and 2–2.5 mA at 1 ms stimulus duration. Subjects were asked to subjectively evaluate the nature of each stimulus intensity as not painful, mildly painful, moderately painful and strongly painful. By stimulating the index finger at a fixed ST multiple, we recorded the SAP from the median nerve at the wrist using bipolar electrodes. SAP amplitudes, SP latency and duration were normalised as a percentage of the corresponding maximum value. Stimulus intensity was expressed as a multiple of the ST and curves were calculated for the two functions i.e. stimulus intensity/SP duration and stimulus intensity/SAP amplitude (Inghilleri et al., 1997). 2.4. Habituation and recovery cycle of the CSP The excitability state of the neuronal substrate of CSP was studied by measuring the habituation and the recovery cycle of the EMG suppression in the APB. This experiment was conducted at low- and high-intensity levels (2 £ ST and 8 £ ST). Habituation was investigated by using a train of 20 stimuli at 3 Hz and recording the changes in the mean EMG area of the rectified and averaged EMG signal (habituation index ¼ last CSP/baseline CSP percentage). The recovery cycle was investigated by delivering paired stimuli of the same intensity at different interstimulus intervals: 500, 300, 250, 200, 150 and 100 ms. For each time interval, a series of 10 trials was repeated at 0.2 Hz frequency. To prevent fatigue and habituation, the subjects rested for 2 min at the end of each series. The recovery cycle was expressed by plotting the interstimulus interval on the X axis and the area of responses to the second shock (test) as a percentage of the area to the first shock (conditioning) on the Y axis (Cruccu et al., 1984). 2.5. Ischaemic test In order to identify the afferent fibres responsible for the observed effects, ischaemia was induced by a tourniquet
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
2009
Table 1 Relationship between CSP onset, SAP onset and pain threshold (stimulus duration 0.2 ms) Subjects
1 2 3 4 5 6 7 8 9 10
CSP onset (threshold units)
1£ 1£ 2£ 1£ 1£ 1£ 2£ 2£ 1£ 2£
SAP onset (threshold units)
2£ 1.4 £ 2£ 1.2 £ 1.5 £ 1.5 £ 1.5 £ 1.5 £ 2£ 2£
inflated above systolic blood pressure (220–250 mmHg) across the elbow proximally to the target APB muscle in 3 patients in which we were able to record an H-reflex from the FC. The CSP evoked at low (2 £ ) and high (8 £ ) stimulus intensity were evaluated at 10, 20 and 30 min. The Hreflex, SAP and cMAP from the median nerve were monitored continuously during the experiment, for approximately 30 min. Immediately after these measurements, the tourniquet was gradually released over a period of 2–3 min. The excitability parameters were remeasured 5–10 min after the starting of tourniquet deflation. 2.6. Statistical analysis Data were evaluated by analysis of variance (ANOVA) for repeated measurements. P , 0:05 were considered statistically significant. All values are reported as mean ^ SD. 3. Results 3.1. Relationship between ST, SAP and SP The relative stimulus intensities inducing the CSP by stimulation of the digital nerves are shown in Table 1. A stable reproducible CSP was identified at the perceptual threshold in 6 subjects for every stimulus duration (Fig. 1). At 2 £ ST, the CSP was detected in all subjects. In most subjects the SP was seen just before or at the point in which the SAP appeared. The pain threshold (mildly painful) ranged between 3 £ ST and 5 £ ST. Strongly painful threshold varied between 8 £ ST and 12 £ ST; at these stimulus intensities the CSP had plateaued. The effect of increasing stimulus intensity on the voluntary EMG activity of the APB and the effect on the SAP amplitude are shown in (Fig. 2). Both functions were sigmoidal but separated and differently shaped. The CSP required lower intensities to appear and increased abruptly between 3.5 £ ST and 4 £ ST. The SAP amplitude was saturated before CSP saturation (Fig. 2).
Pain threshold (threshold units) Mildly
Moderately
Strongly
4£ 3.5 £ 4£ 3£ 4.5 £ 3£ 5£ 3£ 3.5 £ 3.5 £
6£ 6£ 7£ 5£ 6.5 £ 5£ 7.5 £ 5£ 6£ 6£
10 £ 10 £ 9£ 9£ 10 £ 8£ 12 £ 8£ 9£ 8£
While the total duration of EMG inhibition increased, with increasing stimulus intensities that of the onset latency of the SP was reduced (Table 2). The S–X interval was not significantly modified by increasing stimulus intensities (Table 2). In both subjects in which the CSP at 2 £ ST was evaluated during different levels of contraction (20–100%), the duration of the SP was dependent on the intensity of the contraction. The SP was inversely related to the intensity of the contraction. The SP was present in both subjects up to 80% of the maximal force. At 100% of the maximal force, the CSP disappeared in the rectified and average EMG tracing. However, at this strength level the CSP was present in few traces of the raw EMG signal. The ST in the patients with Friedreich’s ataxia was 3– 5 mA. At 2 £ ST, no CSP was evoked in ABP muscle. The CSP appeared only at higher stimulus intensities perceived as painful (6 £ ST–8 £ ST). 3.2. Habituation and recovery cycle Repetitive stimulation at 2 £ ST disclosed, in every
Fig. 1. CSP elicited in APB muscle with stimulation of the index finger at 2 £ (upper trace) and 8 £ (lower trace) ST. Rectified and averaged EMG signal during isometric contraction at 50% of maximal strength. The effect of stimulus intensity is represented by a decrease in onset latency and an increase in the duration.
2010
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 2. Function stimulus 2 intensity/CSP duration vs. function stimulus 2 intensity/SAP amplitude in 10 healthy subjects. Stimulus intensity is expressed as a multiple of ST. Both SAP amplitude and CSP duration are expressed as a percentage of maximum values. The mean pain threshold is indicated by a vertical dashed line. Both functions were sigmoidal but separated and shaped differently. The CSP (triangles) required lower intensities to appear and increased abruptly between 3.5 £ ST and 4 £ ST. The SAP amplitude (squares) was saturated before CSP saturation. Bars, standard deviation. A schematic model of cutaneous afferents involved in CSP production, based on neurophysiological findings is given above the graph.
subject, a fast habituation of the inhibitory response in the APB (Fig. 3a). The CSP obtained in response to high intensity (8 £ ST) repetitive stimulation did not habituate (Habituation Index ¼ 1.09 ^ 0.3) (Fig. 3b). The results of the recovery cycle varied according to the threshold stimuli (Fig. 4). High intensity paired stimuli did not modify the test CSP. Low intensity paired stimuli induced a reduction of EMG inhibition at 100–150 interstimulus interval, recovering the baseline level of EMG suppression, from 200 up to 500 ms. 3.3. Pattern of responses in distal and proximal muscles after digital nerve stimulation The stimulation of the index finger evoked a different pattern of inhibitory effects in the proximal and distal muscles depending on the stimulus intensity. Low-threshold stimuli (2 £ ST) induced an inhibitory response in the ABP, EC and FC. No change in EMG activity of the BB and TB Table 2 Mean latency, duration and S–X interval of the CSP at different stimulus intensities (stimulus duration 0.2 ms) Stimulus intensity (threshold units)
Onset latency (ms) a
Duration (ms) a
S–X interval (ms)
1£ 2£ 3£ 4£ 6£ 8£
87.8 ^ 6.4 82.0 ^ 6.8 78.8 ^ 7.3 67.0 ^ 7.2 65.4 ^ 7.3 65.0 ^ 7.0
14.0 ^ 5.7 19.0 ^ 5.9 23.4 ^ 7.1 45.0 ^ 8.3 52.8 ^ 8.2 56.0 ^ 8.6
106.3 ^ 10.2 106.2 ^ 10.8 107.0 ^ 11.2 110.0 ^ 13.9 111.0 ^ 13.8 115.0 ^ 14.2
a The CSP onset latency significantly decreased and the duration significantly increased at higher stimulus intensities (ANOVA for repeated measures, P , 0:05).
was found (Table 3). High-intensity stimulation produced EMG suppression in the small hand muscles, a shorter CSP in the forearm muscles and an even shorter one in the TB (ANOVA, P , 0:01). No inhibitory response was found in the BB. 3.4. Ischaemic test The subjects in whom ischaemia was induced by the inflated tourniquet, experienced paraesthesiae in the hand during the first minutes of the ischaemic period. Touch and deep sensitivity were reduced earlier than pain– temperature sensation. Change in low-threshold CSP, SAP and H-reflex, paralleled touch and deep sensation. A 50% reduction of the low-threshold CSP duration occurred after 10 min. At 20 min of the ischaemic period the low-threshold CSP, the H-reflex and SAP were abolished (Fig. 5), whereas the cMAP and the high-threshold CSP decreased (highthreshold CSP reduction ranged between 25 and 30%). At the end of the ischaemic period (30 min) the block was virtually complete. At 5 min after release of the cuff lowthreshold CSP was fully restored in both subjects.
4. Discussion 4.1. Peripheral afferents mediating the cutaneous SP There is a significant amount of evidence which suggest that afferent impulses generating the CSP are carried by Adelta fibres (Uncini et al., 1991; Leis, 1992; Shefner and Logigian, 1993; Leis et al., 1994; Inghilleri et al., 1997). These fibres, which range in diameter from 1 to 6 mM, have a higher threshold for activation than large fibres and
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 3. (A) Raw EMG activity recorded from APB muscle after trains of stimuli delivered at 3 Hz and low stimulus intensity (2 £ ST). The CSP promptly habituates. (B) Raw EMG activity recorded from APB muscle after trains of stimuli delivered at 3 Hz and high-stimulus intensity (8 £ ST). The CSP does not show habituation.
mediate impulses resulting from nociceptive stimuli. (Uncini et al., 1991; Leis et al., 1994). However, the possibility that large-diameter afferents may help to induce the CSP cannot be dismissed. Caccia et al. (1973), after studying the cutaneous reflexes of the hands, suggested that the inhibitory response (I2), whose latency and duration overlap the CSP, was already present when the stimulus was at the perceptual threshold level and was mediated by low-threshold group 2 fibres. Recently, Syed et al. (2000) assessed the CSP in patients with Fabry’s disease to verify whether smalldiameter afferents are responsible for producing the CSP. Their findings showed minor abnormalities in the CSP parameters, thus suggesting the possibility that large-diameter fibres may further contribute in producing the CSP. In our study, the CSP evoked at low stimulus intensity, already appeared at or just above the sensory perception
2011
threshold (1 £ ST–2 £ ST). The onset latency of the lowthreshold CSP ranged between 80 and 90 ms and the duration between 25 and 12 ms. Thus, a suppression of the EMG signal appeared at lower current levels than those required to evoke the SAP, as observed by Caccia et al. (1973). When the stimulus intensity was increased (from 3 £ ST to 12 £ ST), the onset latency of the CSP decreased, whereas the duration increased and the S–X interval remained stable. The stimulus intensity/function of the CSP (Fig. 2) became steeper between 3.5 £ ST and 5 £ ST, at the very point in which the stimulus was perceived as painful. These findings suggest that the CSP evoked by lowthreshold stimuli (up to 2 £ ST–3.5 £ ST), is mediated by large-diameter sensory fibres that do not convey painful impulses. This is confirmed by the fact that low-threshold CSP was absent in patients affected by Friedreich’s ataxia with no recordable SAPs. Additionally, the CSP evoked at low stimulus intensity early decreased after the ischaemic test indicating that type 2 afferents are involved in the origin of the CSP at these stimulus intensities. The suppressive effect on EMG of cutaneous stimulation increased markedly when the stimulus was perceived as painful, which indicates that in this condition the CSP is mainly mediated by A-delta fibres, as reported by several authors (Uncini et al., 1991; Shefner and Logigian, 1993, Inghilleri et al., 1997). The study of the stimulus/intensity function for SAP amplitude and CSP duration showed an almost parallel increase in both parameters, suggesting that the contribution of group 2 fibres may not be limited to the lower stimulus intensities and is probably masked by the recruitment of the group 3 fibres at higher intensities. The observed reduction of high-threshold CSP duration during the ischaemic block at the time in which the low-threshold CSP disappeared, further supports the contribution of the type 2 fibres in the generation of the high-threshold CSP. As the S–X interval was not significantly modified by the current levels, the effect of the nociceptive fibres was to shorten the latency of the CSP. In this regard, group 2 fibres may play a role in inducing the later part of the CSP, while group 3 afferents may be responsible for the earlier part. As mentioned earlier, many authors, though not all, have failed to find marked, stable EMG suppression after low-intensity cutaneous stimulation (Uncini et al., 1991; Inghilleri et al., 1997; Leis, 1994; Leis et al., 2000). These discrepancies are difficult to explain. It may be that, as the CSP is not a wellstandardised technique, the variability in the results obtained in different laboratories is due to different stimulation, recording and measurement paradigms. For instance, Uncini et al. (1991) computed the CSP in 3 different time windows, inappropriate for identifying a short duration CSP as that evoked by low-threshold stimuli. In fact, to obtain the SP from the averaged trials, Uncini used an off-line analysis program calculating the integrated EMG activity during 4 time periods in the EMG tracing. The first period
2012
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
Fig. 4. Mean recovery cycles of the CSP evoked by paired shocks at high (8 £ ST, black triangles) and low (2 £ , squares) stimulus intensity in 10 subjects. High intensity paired stimuli did not modify the CSP test. Low intensity paired stimuli lowered of EMG inhibition at a 100–150 interstimulus interval, recovering the baseline level of suppression from 200 up to 500 ms.
consisted of 100 ms of baseline activity (PRE) from the prestimulus epoch. The other 3 periods, taken from the poststimulus epoch, consisted of the SP, 50 ms before (PI) the SP and 50 ms after (PIII) the SP itself. As a confirmation for the determination of the SP by visual inspection of the rectified averaged EMG tracings, Uncini considered an SP definitely present when the SP/PI ratio was less than 0.6 (mean value of SP/PI 1 3 SD). It is obvious that the SP period measured by Uncini totally corresponds to the EMG suppression activity evoked at high intensity of stimulation, starting at 50–60 ms after the stimuli and lasting 50 ms. In this view, he found a very significant value from the SP/PI ratio (0.29 ^ 0.16). In our study, the CSP evoked at low-threshold stimuli beginning at 80–90 ms corresponds to the end of the SP period reported by Uncini. If we consider that SP evoked at low-threshold lasts 14–19 ms, there will be a drawback in the SP/PI ratio for this reflex. In fact, in this case the low-threshold CSP could correspond to both SP and PIII period reported by Uncini.
With regard to the specific aim of this paper, several technical factors have to be taken into account. First, an occlusion in the excitatory pathways between descending inputs, peripheral feedback activity (during tonic contraction) and the peripheral volley might explain the difficulty in recording inhibition of the muscle activity on the raw EMG signal, after low-intensity stimuli or during excessive muscle contraction. Thus, the visual inspection of the raw EMG signal as used by Shefner and Logigian (1993) may not be suitable for the measurement of the CSP. Moreover, when few trials are averaged and full-wave rectified during intense muscular contraction, as done by Inghilleri et al. (1997), low-level afferent inputs may collide with an intense excitatory output and consequently fail to show significant inhibition on the EMG activity. This is supported by our findings of an inverse relationship between CSP duration
Table 3 Mean latency, duration and S–X interval of the CSP elicited by index finger stimulation in different muscles of the upper limbs at different stimulus intensities (stimulus duration 0.2 ms) a Muscle
BB TB FC EC APB
CSP onset latency (ms)
CSP duration (ms)
2 £ ST
8 £ ST
2 £ ST
8 £ ST b
– – 77.2 ^ 7.3 78.8 ^ 7.6 82.0 ^ 6.8
– 59.4 ^ 3.4 61.5 ^ 2.9 62.7 ^ 3.1 65.0 ^ 7.0
– – 18.1 ^ 6.1 19.2 ^ 6.4 19.0 ^ 5.9
– 20.4 ^ 15.6 41.7 ^ 12.3 42.1 ^ 13.2 56.0 ^ 8.6
a BB, biceps brachii; TB, triceps brachii; FC, flexor carpi; EC, extensor carpi; APB, abductor pollicis brevis. b The CSP significantly increased from proximal to distant muscles.
Fig. 5. Low-threshold (2 £ ST) CSP modifications induced by the ischaemic test in a representative patient. After 10 min (upper trace), the CSP duration is slightly reduced. After 20 min (middle trace) the CSP duration is drastically reduced and no consistent inhibition is recorded 30 min after the beginning of the test (lower trace).
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
and muscular contraction levels. Second, the low-threshold CSP easily habituates; both the frequency of stimulation and the number of trials become critical for a correct evaluation of the CSP. Therefore, the relationship between stimulus intensity and the CSP is far from being elucidated, even if we consider that some inhibition of tonic EMG activity can be elicited simply by fingertip tapping (Caccia et al., 1973). In our opinion, since the publication of findings of Uncini et al. in 1991, research on CSP has focused excessively on the nocifensive nature of this reflex, disregarding the effects of low-intensity stimuli. 4.2. Central neural pathways mediating the CSP It is generally accepted that the CSP mediated by A-delta fibres is an oligosynaptic inhibitory spinal reflex (Uncini et al., 1991; Leis et al., 1995; Inghilleri et al, 1997; Logigian et al., 1999). In fact, the long latency of this reflex (60–70 ms) is due to a long afferent conduction time rather than a long central delay (Manconi et al., 1998). As the central processing time (CPT) for this reflex has been estimated to be about 15 ms (Shefner and Logigian, 1993), it is unlikely to involve the activation of brainstem or cortical structures. Moreover, Logigian et al. (1999) showed that an inhibitory effect corresponding to the CSP occurs in the lower limbs of patients with complete spinal cord injury. By contrast, since the cutaneous afferents that mediate the low-threshold CSP belong to group 2 fibres conducting at about 50 ms 21, the longer onset latency of this reflex can only be due to a longer central delay. Considering that the CPT may be estimated by the following formula CPT ¼ CSP latency 2 efferent time ([F 1 M]/2) 2 afferent time (distance between index finger to C6 root/conduction velocity of afferent fibres) (Shefner and Logigian, 1993), the central delay for the low-threshold CSP is about 55 ms. This is a feasible period of time for a polysynaptic circuit, such as brainstem spinobulbospinal or transcortical circuits. The central neural substrates producing alpha-motor neuron inhibition may be investigated by studying the habituation and recovery cycle using the double stimulus technique (Kimura, 1973; Cruccu et al., 1984; Inghilleri et al., 1997). With regard to the CSP evoked by high-intensity stimuli, we obtained results similar to those reported by other authors (Uncini et al., 1991; Inghilleri et al., 1997) showing a lack of habituation and fast recovery cycle. These findings suggest that the underlying neuronal network of the CSP is oligosynaptic. The CSP evoked by non-painful stimuli, on the contrary, habituated and exhibited a slow recovery cycle. This pattern is typical of polysynaptic exteroceptive reflexes (Cruccu et al., 1984; Cruccu et al., 1991; Priori et al., 1998). Thus, our findings provide evidence that the CSP evoked by low-intensity electrical finger stimulation is mediated by group 2 fibres, possibly by activating transcortical or brainstem polysynaptic circuits inhibiting the alpha-motor neurons.
2013
4.3. Relationship between the CSP and the CMRs The low-intensity CSP, as described in this paper, shows a behaviour similar to the so-called CMRs evoked by electrical non-painful stimulation of the fingers. Several researchers have reported that, in the upper limbs, lowintensity stimulation of the digital nerves of the index finger produces reflex responses in the hand muscle and in the forearm (Caccia et al., 1973; Jenner and Stephens, 1982, Evans et al., 1989). In the adult, this reflex response is characterised by several identifiable components comprising an initial short latency increase in EMG (the E1 component), followed by a decrease (the I1 component), followed by a second inhibition (the I2 component) and a second increase (the E2 component). The first two components are believed to be mediated by the activation of oligosynaptic spinal pathways, whereas the I2 and E2 components are thought to involve longer pathways, probably long loops extending to the cortex (Caccia et al., 1973; Jenner and Stephens, 1982; Issler and Stephens, 1983) Our data indicate that the CSP evoked by low-threshold fibres corresponds to the I2 component of the cutaneous muscular reflexes (Caccia et al., 1973; Brezinova, 1988). In fact, the I2 response share the same neurophysiological characteristics in terms of onset latency, effect of stimulus strength on reflex activity, habituation and distribution of reflex effects (Caccia et al., 1973). It is noteworthy that by increasing levels of digital stimulation EMG inhibition increases, while excitatory components decrease; the final result is that the I1 and I2 phases merge (Caccia et al., 1973; Clouston et al., 1995). On the basis of our data, we suggest that the CSP evoked by noxious stimuli may reflect a complex summation of the inhibitory components mediated by group 3 and group 2 fibres.
4.4. Functional significance of the CSP By studying the effects of finger stimulation on several muscles, we observed after low- and high-intensity stimulation, a common pattern of responses in distal and proximal muscles consisting of stronger reflex effects in the hand and forearm muscles, as reported by other authors (Caccia et al., 1973; Uncini et al., 1991; Inghilleri et al., 1997). This pattern of ‘distal’ inhibition may be interpreted from a functional viewpoint. As suggested by Leis (1998), the CSP mediated by Adelta fibres may be considered as part of the circuitry that mediates withdrawal flexor reflexes. The physiological role of the CSP evoked by low-threshold stimuli must be placed within the general pattern of excitability changes induced by the non-nociceptive cutaneous input in upper limb motor neurons in humans. The similarity between the low-threshold CSP and the I2 component of the CMRs suggests that the inhibitory response may
2014
M. Serrao et al. / Clinical Neurophysiology 112 (2001) 2007–2014
help modulate the motorneuron activity which contributes to explorative movements of the hands. At the same time, sensory information from cutaneous large-diameter afferents may contribute to maintaining the excitability of the nocifensive reflex responses and thus improve the reflex gain in the different circumstances. In fact, a protective motor reaction to a nociceptive input would be accelerated and enhanced by a concomitant inhibition of the hands by other non-nociceptive afferent fibres. 5. Conclusions Our findings support the notion that low-threshold afferents participate in the production of the CSP. Further studies are needed to establish the relative contribution of the different classes of afferents generating the CSP. Nevertheless, by observing strict experimental paradigms, the study of the CSP at low- and high-intensity stimulation may provide additional information on different neural pathways, including the afferent sensory pathways and the central mechanisms modulating motor recruitment. References Brezinova V. Cutaneomuscular reflex in a peripheral nerve lesion. Electromyogr Clin Neurophysiol 1988;28:183–189. Caccia MR, McComas AJ, Upton RM, Blogg T. Cutaneous reflexes in small muscles of the hands. J Neurol Neurosurg Psychiatry 1973;36:960–977. Clouston PD, Kiers L, Menkes D, Sander H, Chiappa K, Cros D. Modulation of motor activity by cutaneous input: inhibition of the magnetic motor evoked potential by digital electrical stimulation. Electroenceph clin Neurophysiol 1995;97:114–125. Cruccu G, Agostino R, Fornarelli M, Inghilleri M, Manfredi M. Recovery cycle of the masseter inhibitory reflex in man. Neurosci Lett 1984;49:63–68. Cruccu G, Pauletti G, Agostino R, Berardelli A, Manfredi M. Masseter inhibitory reflex in movement disorders. Huntington’s chorea, Parkin-
son’s disease, dystonia, and unilateral masticatory spasm. Electroenceph clin Neurophysiol 1991;81:24–30. Evans AL, Harrison LM, Stephens JA. Task-dependent changes in cutaneous reflexes recorded from various muscles controlling finger movement in man. J Physiol (Lond) 1989;418:1–12. Garnett R, Stephens JA. The reflex response of single motor units in human first dorsal interosseous muscle following cutaneous afferent stimulation. J Physiol (Lond) 1980;303:351–364. Inghilleri M, Cruccu G, Argenta M, Polidori L, Manfredi M. Silent period in upper limb muscles after noxious cutaneous stimulation in man. Electroenceph clin Neurophysiol 1997;105:109–115. Issler H, Stephens JA. The maturation of cutaneous reflexes studied in the upper limb in man. J Physiol (Lond) 1983;335:643–654. Jenner JR, Stephens JA. Cutaneous reflex responses and their central nervous pathways studied in man. J Physiol (Lond) 1982;333:415–419. Kimura J. Disorders of interneurons in parkinsonism. The orbicularis oculi reflex to paired stimuli. Brain 1973;96:87–96. Leis AA. Conduction abnormalities detected by silent period testing. Electroenceph clin Neurophysiol 1994;93:444–449. Leis AA. Cutaneous silent period. Muscle Nerve 1998;21:1243–1245. Leis AA, Kofler M, Ross MA. The silent period in pure sensory neuronopathy. Muscle Nerve 1992;15:1345–1348. Leis AA, Stetkarova I, Beric A, Stokic DS. Spinal motor neuron excitability during the cutaneous silent period. Muscle Nerve 1995;18:1464–1470. Leis AA, Stokic DS, Fuhr P, Kofler M, Kronenberg MF, Wissel J, Glocker FX, Seifert C, Stetkarova I. Nociceptive fingertip stimulation inhibits synergistic motoneuron pools in the human upper limb. Neurology 2000;55:1305–1309. Logigian EL, Plotkin GM, Shefner JM. The cutaneous silent period is mediated by spinal inhibitory reflex. Muscle Nerve 1999;22:467–472. Manconi FM, Syed NA, Floeter MK. Mechanisms underlying spinal motor neuron excitability during the cutaneous silent period in humans. Muscle Nerve 1998;21:1256–1264. Priori A, Berardelli A, Inghilleri M, Pedace F, Giovannelli M, Manfredi M. Electrical stimulation over muscle tendons in humans. Evidence favouring presynaptic inhibition of Ia fibres due to the activation of group III tendon afferents. Brain 1998;121:373–380. Shefner JM, Logigian EL. Relationship between stimulus strength and cutaneous silent period. Muscle Nerve 1993;16:278–282. Syed NA, Sandbrink F, Luciano CA, Altarescu G, Weibel T, Schiffmann R, Floetr MK. Cutaneous silent period in patients with Fabry disease. Muscle Nerve 2000;23:1179–1186. Uncini A, Kujirai T, Gluck B, Pullman S. Silent period induced by cutaneous stimulation. Electroenceph clin Neurophysiol 1991;81:344–352.