Clinical Neurophysiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Cutaneous silent period recordings in demyelinating and axonal polyneuropathies Diego Lopergolo a, Baris Isak b,c, Maria Gabriele a, Emanuela Onesti a, Marco Ceccanti a, Gelsomina Capua a, Laura Fionda a, Antonella Biasiotta a, Giulia Di Stefano a, Silvia La Cesa a, Vittorio Frasca a, Maurizio Inghilleri a,⇑ a b c
Department of Neurology and Psychiatry, University ‘‘Sapienza’’, Viale dell’Università 30, 00185 Rome, Italy Marmara University Hospital School of Medicine, Department of Neurology, Fevzi Cakmak Mah. Mimar Sinan Cad. No: 41, 34899 Ust Kaynarca/Pendik, Istanbul, Turkey Department of Clinical Neurophysiology, Aarhus Universitets hospital, Nørrebrogade 44, 8000 Aarhus C, Denmark
a r t i c l e
i n f o
Article history: Accepted 15 November 2014 Available online xxxx Keywords: Cutaneous silent period A-delta fibres Axonal polyneuropathy Demyelinating polyneuropathy Neuropathic pain
h i g h l i g h t s The evaluation of cutaneous silent period (CSP) is a useful method to detect dysfunction of A-delta
fibres in patients with axonal polyneuropathy (PNP) as well as demyelinating PNP. The number of axons and conduction properties of A-delta fibres play respectively crucial roles in CSP
duration and CSP latencies. CSP parameters do not differ between patients with and without neuropathic pain.
a b s t r a c t Objective: To investigate the cutaneous silent period (CSP), a spinal inhibitory reflex mainly mediated by A-delta fibres, in demyelinating and axonal polyneuropathy (PNP) and evaluate whether CSP parameters differ between patients with and without neuropathic pain. Methods: Eighty-four patients with demyelinating PNP, 178 patients with axonal PNP and 265 controls underwent clinical examination, DN4 questionnaire, standard nerve conduction study, motor-root stimulation and CSP recordings from abductor digiti minimi. We calculated the afferent conduction time of CSP (a-CSP time) with the formula: CSP latency root motor evoked potential latency. Results: In the demyelinating PNP group the a-CSP time was significantly longer; in the axonal PNP group, CSP duration was shorter than the demyelinating group (p = 0.010) and controls (p = 0.001). CSP parameters were not different between patients with and without neuropathic pain. Conclusions: The abnormality of a-CSP time in the demyelinating PNP group suggests the crucial role of A-delta fibres in the mechanism of CSP; the shorter CSP duration in the axonal PNP group supports the strong influence of the number of axons on this parameter. Conclusions: Our study suggests that neuropathic pain could be related to pathophysiological mechanisms differing from mere A-delta fibre loss. Significance: CSP evaluation is effective in detecting A-delta fibre dysfunction in axonal as well as demyelinating PNP. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Polyneuropathy (PNP) is a neurological disorder caused by various conditions, with a prevalence of approximately 2.400 per 100.000 population (England and Asbury, 2004; England et al., ⇑ Corresponding author. Tel.: +39 0649914485; fax: +39 0649914576. E-mail address: [email protected] (M. Inghilleri).
2005; Martyn and Hughes, 1997). Peripheral neuropathy usually manifests with symmetric motor and sensory involvement. Sensory disturbances including hypesthesia, paraesthesia and neuropathic pain are associated with nociceptive pathway damage (Koltzenburg and Scadding, 2001). The diagnosis of PNP is based on a combination of clinical symptoms, signs and electrodiagnostic findings (Dyck et al., 1992, 1997; Feldman et al., 1994; Teunissen et al., 1997; Peripheral Nerve Society, 1995; Proceedings of a
http://dx.doi.org/10.1016/j.clinph.2014.11.013 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
2
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
consensus development conference on standardized measures in diabetic neuropathy, 1992). Standard neurophysiological techniques such as nerve conduction studies (NCSs) and somatosensory evoked potentials are useful in demonstrating damage along peripheral and central sensory pathways, but they do not assess the function of small fibres (Cruccu et al., 2004). Without techniques specific for the study of small fibres (i.e., experimental blocks, laser evoked potentials and contact-heat evoked potentials) or special organ stimulation (i.e., cornea, tooth pulp and glans), electrical stimuli unavoidably also excite large fibres, because large sized non-nociceptive afferents are characterised by a lower electrical threshold than small sized afferents (i.e., A-delta and C-nociceptive fibres) (Cruccu et al., 2004). Unfortunately, these techniques are expensive and require the use of specialised equipment not widely available in most electrophysiological laboratories. The evaluation of cutaneous silent period (CSP) represents a useful electrophysiological method for investigating the function of small fibres (Caccia et al., 1973; Inghilleri et al., 1997) by using standard electromyography equipment (Floeter, 2003). The CSP is defined as a transient suppression of voluntary muscle contraction following painful electrical stimulation of cutaneous sensory nerves. It represents a nociceptive reflex which is mediated mostly by A-delta afferent fibres (Shahani and Young, 1973; Uncini et al., 1991; Inghilleri et al., 1997, 2002; Floeter, 2003). These fibres activate an oligosynaptic circuit that in turn inhibits the motoneurons (Inghilleri et al., 1997). Although pre-synaptic inhibition has also been proposed as an explanation for electromyographic (EMG) suppression by A-delta afferents at spinal level (Leis et al., 1995, 1996), there is widespread agreement that motor neurons receive post-synaptic inhibition during the CSP and that the inhibition is transmitted through spinal inhibitory interneurons (Logigian et al., 1999; Inghilleri et al., 1997; Floeter, 2003). However, although the spinal circuitry of CSP is still not well known, the last-order interneuron is inhibitory and capable of producing a secure post-synaptic inhibition (Floeter, 2003). Although electrical stimuli used to evoke the CSP activate both large and small-diameter fibres, and a contribution to CSP generation of large-diameter afferents has also been suggested (Serrao et al., 2001), the long latency of the CSP is more compatible with a spinal reflex produced by slowly conducting afferents (Floeter, 2003). Moreover, normal CSPs were recorded in patients with profound large-diameter fibre sensory neuropathies with absent SNAPs or somatosensory evoked potentials (Uncini et al., 1991; Leis et al., 1992; Inghilleri et al., 1995). The CSP latency reflects the conduction function of A-delta afferents, efferent motor axons and synaptic delay; the CSP duration depends on the amount of activated fibres and on changes in nociceptive input (Leis et al., 1992, 1996; Inghilleri et al., 1997; Gilio et al., 2008). As a contrast from laser evoked potentials, that investigate nociceptive pathways up to cortex by activating A-delta and C fibres and by evoking scalp potential related to A-delta fibres and, through dedicated techniques, also C fibres (Treede, 2003), the evaluation of CSP could explore the sensory pathway from sensory endings to spinal level and the functional integrity of the spinal circuitry, in particular centromedullary connections, being activated by A-delta fibres (Stetkarova et al., 2001; Kofler et al., 2003a; Stetkarova and Kofler, 2009). CSPs have been demonstrated to be altered in different types of neuropathies (Leis, 1994; Syed et al., 2000; Corsi et al., 2002; Osio et al., 2004; Svilpauskaite et al., 2006a; Koo et al., 2010; Kim et al., 2010; Onal et al., 2010). Nevertheless, to our knowledge, no studies have evaluated the utility of CSP in a large number of patients with PNP. Additionally, the modification of CSP parameters in different
pathogenetic processes (i.e., axonal vs. demyelinating) of PNPs has not yet been analysed in comparative studies. In this study we aimed to determine A-delta fibre dysfunction in patients with demyelinating PNP vs. patients with axonal PNP by evaluating the CSP and other standard electrophysiological parameters of conventional NCSs. Furthermore, we evaluated whether CSP parameters varied between patients with and without neuropathic pain in either patient group. 2. Methods 2.1. Subjects Over a period of four years (from January 2008 to January 2012) 84 patients with demyelinating PNP (37 females, 47 males) and 178 patients with axonal PNP (83 females, 95 males) were consecutively recruited from the inpatient and outpatient clinics of the Department of Neurology and Psychiatry of ‘‘Sapienza’’ University in Rome. A group of 265 healthy subjects (123 females, 142 males) was also enrolled. The study protocol was in accordance with the Helsinki Declaration of Human Rights and was approved by the local Ethics Committee. Written informed consent was provided by all the patients and controls to participating in the study. 2.1.1. Control group Normative values for the electrophysiological parameters were obtained from 265 healthy controls. 2.1.2. Patients The diagnosis of axonal and demyelinating PNP was based on established criteria and recommendations of the American Academy of Neurology and the American Association of Neuromuscular and Electrodiagnostic Medicine (England and Asbury, 2004; England et al., 2005). Patients with demyelinating PNP met the clinical diagnostic criteria for Chronic Inflammatory Demyelinating Polyneuropathy (CIDP), with a symptomatic sensorimotor neuropathy involving upper and lower limbs for at least two months (Joint Task Force of the EFNS and the PNS, 2010); patients with monoclonal gammopathy, paraneoplastic, metabolic disorders such as diabetes, disimmune diseases such as sarcoidosis, systemic lupus erythematosus and angiitis were excluded from the study. In patients with axonal PNP, diabetes was the etiological cause for 77% of the group (137 patients), chronic idiopathic axonal polyneuropathy being the etiological cause for the remaining 23% (41 patients). Exclusion criteria for both groups were central nervous system diseases and cognitive impairment. All subjects enrolled were not taking medication for neuropathic pain and were able to achieve a state of interferential muscular contraction. 2.2. Clinical evaluations All patients were investigated with detailed tests for muscle strength and sensory examination. Muscle strength was assessed with the Medical Research Council (MRC) Score for Muscle Strength (Medical Research Council, 1976) with a cumulative score ranging between 0 and 80 for upper limbs (shoulder abduction, elbow flexion, elbow extension, wrist flexion, wrist extension, thumb opposition, finger flexion, finger extension) and between 0 and 70 for lower limbs (hip flexion, knee extension, ankle dorsiflexion, ankle plantar flexion, toe flexion, toe extension).
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
All patients were divided into two groups according to the presence or absence of neuropathic pain. Screening for neuropathic pain was performed using the DN4 questionnaire, a clinicianadministered qualitative screening tool that indicates neuropathic pain when the score is P4 (Bouhassira et al., 2005). All clinical evaluations and NCSs were performed by the same examiners in order to minimise investigator variations. 2.3. Electrophysiological evaluations All electrophysiological data were recorded using a Micromed Myoquick 1400 EMG machine (Micromed S.p.A., Treviso, Italy). The experimental procedures were completed with the subjects seated in a comfortable reclining chair with a mounted head rest. During the electrophysiological evaluation, skin temperature was controlled and maintained between 31 and 34 °C. Conventional surface electrode techniques were used. In all patients and healthy subjects electrical motor root stimulation, motor NCS and CSP were performed recording from right abductor digiti minimi (ADM) muscle. Orthodromic sensory NCS for median and ulnar nerves was respectively achieved by stimulation of second and fifth digits. Furthermore, in the context of CSP recordings, average pre-stimulus EMG amplitude was also calculated in order to evaluate and obtain a level of muscle contraction that was similar between patients and healthy subjects. The electrophysiological investigations also included the exploration of peroneal, medial plantar and sural nerves, but for the purpose of this study only upper limb-related neurophysiological responses were taken into account. The methods used were coherent to the recommendations of experts of the International Federation of Clinical Neurophysiology (Kimura, 2006). 2.3.1. Electrical motor-root stimulation Electrical motor-root stimulation was performed by high voltage transcutaneous electrical stimulation over the cervical vertebral column, at level of cervical enlargement of the spinal cord (Lange et al., 1992; Mills and Murray, 1986). Stimulation of the cervical region was accomplished by using a Digitimer Stimulator D180 (Digitimer Ltd., Welwyn Garden City, UK) which was able to deliver maximal electrical shocks of 750 V (50–100 ls decay time). Cervical electrodes were placed on the neck, the cathode slightly caudal to the C-7 spinous process and the anode midline between T1 and occiput (approximately at the level of C4 process). The subjects were instructed to relax during cervical stimulation. The initial intensity of stimulation corresponded to 150 V and increments of 20 V were applied. In this way the intensity was gradually increased until maximal root motor evoked potential (root-MEP) amplitude was achieved. The actual intensities used were about 350–400 V. The stimulation intensity was adjusted by selecting the stimulus voltage which produced the shortest response latency. Root-MEPs were recorded from ADM muscle by Ag/AgCl surface electrodes attached to the target muscle by the belly-tendon method. Filter settings were 5 Hz–5 kHz, sweep duration was 50 or 100 ms, and the sensitivity was 0.5–5 mV/division. Latency was measured from the stimulus artifact to the onset of the root-MEPs. The root-MEPs with the largest possible amplitudes and the shortest latencies were evaluated. 2.3.2. Motor NCS (M-NCS) The M-NCS was performed by stimulating the ulnar nerve at the wrist and the elbow and recording from ADM muscle. The level of the stimulus intensity was slowly increased until maximal compound muscle action potentials (CMAPs) were obtained. CMAPs were recorded by Ag/AgCl surface electrodes attached to the target muscle by the belly-tendon method. The filter settings
3
were 5 Hz–5 kHz, sweep duration was 30 ms, and the sensitivity was 1 mV/division. Distal motor latency (DML) to the onset of the negative peak, baseline to peak amplitude of distal CMAP (following wrist stimulation) and the motor nerve conduction velocity (M-NCV) (elbow to wrist) were determined. 2.3.3. Sensory NCS (S-NCS) The S-NCS was performed orthodromically from median and ulnar nerves. Percutaneous stimuli were delivered until adequate maximal sensory action potentials (SNAPs) were obtained; recordings were performed with surface electrodes. Filters were set at 5 Hz–5 kHz, sweep duration was 30 ms and sensitivity was 5 lV/ division. Distal sensory latency (DSL) at the valley of the positive deflection, positive to negative peak amplitude of SNAPs and the distal sensory nerve conduction velocity (S-NCV) (digit II or V to wrist) were determined. 2.3.4. CSP Single consecutive electrical stimuli (duration 0.5 ms) were delivered using ring electrodes placed around the second and third digits of the right hand with the cathode proximally positioned. In a preliminary study, the sensory threshold (STh) was explored in 20 patients with axonal PNP, 20 patients with demyelinating PNP and 20 healthy controls. The STh was defined as the minimum stimulus intensity perceived by the subject five times over five trials. With respect to our results, in all participants of our study the stimulus intensity was set at 80 mA, which corresponds to a stimulus intensity more than 20 times the mean STh of the three samples in our preliminary study. EMG signals were recorded from right ADM muscle through a pair of surface Ag/AgCl electrodes by the belly-tendon method. Patients and healthy subjects underwent painful electrical stimulation in the second and third digits. In order to match the level of background EMG activation, all patients maintained maximal voluntary isometric contraction of ADM muscle for about 1 s before and after finger stimulation, whereas the healthy subjects had to maintain a submaximal ADM contraction ranging between 30% and 50% of the maximal effort (which corresponded to a range of contraction in which there are no more fluctuations in CSP parameters) (Kofler et al., 2007). Filters were set at 50 Hz–5 kHz, sweep duration was 500 ms and sensitivity was 100 lV/division. In order to monitor the steadiness of contraction, EMG activity was displayed on a free-run trace and was also played over a loudspeaker to provide visual and auditory feedback to the subject and examiner. Signals were amplified, full-wave rectified and off-line averaged (10 trials). The average pre-stimulus EMG amplitude was calculated as the average value of the EMG signal during a 100 ms time window prior to the stimulus. The latency of the CSP was determined by inspection of the rectified EMG signal at the point in which the average EMG amplitude dropped below 50% of the pre-stimulus levels. In accordance with the recommendations of the International Federation of Clinical Neurophysiology (IFCN) (Kimura et al., 1994), we measured the CSP duration rather than the amplitude of suppression, because CSP duration is less dependent on the EGM background level (Romaniello et al., 2004).The duration of the CSP was calculated from the onset of the CSP up to the point where the EMG activity amplitude returned above 50% of the pre-stimulus level (Gilio et al., 2008). In the case of absent CSPs, CSP duration was considered equivalent to 0 ms and a null value was attributed to CSP latency.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
4
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
2.3.5. Afferent conduction time of CSP In all participants, we calculated a parameter which we referred to as ‘‘afferent CSP time’’ (a-CSP time). It demonstrates the difference between the CSP latency and the root-MEP latency. Thus, we can exclude the efferent time from CSP latency. This parameter facilitated the investigators in analysing the conduction of the afferent fibre pool (mostly A-delta), this proved particularly useful for the patients with unavailable SNAPs and available CSP responses.
2.3.6. Estimated afferent and efferent conduction velocity of CSP In order to estimate the afferent conduction velocity, we measured in each subject the distance between the proximal phalanx of the second and third digits and C-7 spinous process. We then divided this distance by the a-CSP time. In the case of the efferent conduction velocity of CSP, we measured in each subject the distance between the ulnar nerve at the wrist and the C-7 spinous process; we then divided this distance by the difference between root-MEP latency and DML.
2.4. Statistical analysis The Mann–Whitney U test was used to analyse the differences in non-normally distributed values. Differences in the frequency of neuropathic pain, demographic data and the absence of evocable CSPs or SNAPs across the different aetiologies were analysed with Fisher’s exact test. Spearman’s rank correlation coefficient was used for the correlation analysis between CSP parameters (CSP duration and a-CSP time) and NCS parameters. Normal distribution of demographic, clinical and electrophysiological parameters were tested with the Shapiro–Wilk normality test. All values are reported as means ± standard deviation (SD). The statistical significance limit was accepted as p < 0.05. A Bonferroni correction was applied for Mann–Whitney U test and Fisher’s exact test for multiple comparisons: a p value of less than 0.017 was considered in order to indicate significant differences. Data were analysed using the Statistical Package for Social Sciences (SPSS 19.0).
3.3. Electrophysiological evaluation 3.3.1. Control group Normative values of standard NCSs and CSP study obtained from our controls are shown respectively in Tables 2 and 3. The STh detected in the preliminary study in a sample of 20 subjects resulted 2.3 ± 0.3 mA (range 1.8–3.1 mA). The mean distance used to estimate the afferent conduction velocity of CSP (distance between the proximal phalanx of the second and third digits and C-7 spinous process) was 826 ± 42 mm. As for the estimation of efferent conduction velocity, the mean distance between the wrist and C-7 spinous process was 718 ± 45 mm. 3.3.2. Patients with axonal PNP The STh detected in the preliminary study was 3.1 ± 0.3 mA (range 2.5–3.5 mA). The axonal PNP group showed longer ulnar DML and DSL, rootMEP latency, and a-CSP time (Fig. 1), reduced SNAP and CMAP amplitudes, and slower M-NCV and S-NCV when compared to the healthy subjects (Tables 2 and 4). The CSP duration was shorter compared to the demyelinating PNP group and the healthy subjects (Table 3, Figs. 2 and 3). The absence of CSP was detected only in the group with axonal PNP [26 patients, 14.6%, vs. 0 patients with demyelinating PNP, vs. 0 healthy subjects (Fisher’s exact test, p < 0.001)]. We found absent SNAPs from both ulnar and median nerves in 44 patients (24.7%). This frequency was different from that of patients with demyelinating PNP [39 patients, 46.4% (Fisher’s exact test, p < 0.001)] and controls [0 subjects, 0% (Fisher’s exact test, p < 0.001)]. Among patients with absent SNAPs, only 14 (8%) also indicated the absence of CSP responses. The mean distances used to estimate the afferent and efferent conduction velocity of CSP were respectively 807 ± 54 mm and 704 ± 55 mm. Indirect estimation of the afferent and efferent conduction velocity of the fibres mediating the CSP respectively yielded values of 12.5 ± 2.1 m/s and 55.0 ± 7.3 m/s. The average pre-stimulus EMG amplitude analysis revealed no statistically significant differences compared to the other groups (Table 3). A-CSP time and CSP duration did not show any significant correlation with the other considered electrophysiological parameters (Table 4).
3. Results 3.1. Demographic values The demographic data of patients and healthy subjects are described in Table 1. No statistical differences were evident among the patients with axonal PNP and the patients with demyelinating PNP (p > 0.05 for all comparisons).
3.2. Clinical evaluation Clinical data of patients and healthy subjects are described in Table 1. The demyelinating PNP group suffered more loss of strength in upper and lower limbs as evidenced with MRC score compared to the axonal PNP group (p < 0.001). Patients with axonal PNP showed a more compromised pin-prick sensation (p < 0.001) and neuropathic pain was more present in comparison with patients with demyelinating PNP (52 patients, 29.2%, in axonal PNP group vs. 9 patients, 10.7%, in demyelinating PNP group, p < 0.001). The other clinical data did not show further differences among the patient groups (Table 1).
3.3.3. Patients with demyelinating PNP The STh detected in the preliminary study was 3.0 ± 0.4 mA (range 2.1–3.6 mA). In the demyelinating PNP group, the DMLs, DSLs, root-MEP latency and a-CSP time (Fig. 1) were significantly longer, the amplitudes of CMAP and SNAP were significantly lower, and the S-NCV and M-NCV were significantly slower compared to the axonal PNP group and the healthy subjects (Tables 2 and 4). The CSP duration was not different to the healthy subjects (Table 3, Figs. 2 and 3). CSP was always present, but 39 (46.4%) patients presented the absence of SNAPs from both ulnar and median nerves. This frequency was different from that of patients with axonal PNP [44 patients, 24.7% (Fisher’s exact test, p < 0.001)] and controls [0 subjects, 0% (Fisher’s exact test, p < 0.001)]. The mean distances used to estimate the afferent and efferent conduction velocity of CSP were respectively 803 ± 41 mm and 700 ± 41 mm. Indirect estimation of the afferent and efferent conduction velocity of the fibres mediating the CSP, respectively yielded values of 11.5 ± 3.5 m/s and 35.8 ± 9.2 m/s.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
5
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx Table 1 Demographic and clinical characteristics of patients and healthy subjects (mean ± SD). Characteristics
Demyelinating group (n = 84)
Axonal group (n = 178)
A vs. D p
Sex (F;M) Age (years) Upper limbs MRC Trophism Tactile sensation Pin-prick sensation Vibration sensation Lower limbs MRC Trophism Tactile sensation Pin-prick sensation Vibration sensation Neuropathic pain
37 F; 47 M 61.2 ± 10.1
83 F; 95 M 61.7 ± 10.6
74.4 ± 7.1 0.07 ± 0.3 1.3 ± 1.4 1.0 ± 1.3 2.9 ± 1.1
78.3 ± 5.5 0.1 ± 0.3 1.2 ± 1.5 1.8 ± 1.5 3.0 ± 0.7
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Cutaneous silent period recordings in demyelinating and axonal polyneuropathies Diego Lopergolo a, Baris Isak b,c, Maria Gabriele a, Emanuela Onesti a, Marco Ceccanti a, Gelsomina Capua a, Laura Fionda a, Antonella Biasiotta a, Giulia Di Stefano a, Silvia La Cesa a, Vittorio Frasca a, Maurizio Inghilleri a,⇑ a b c
Department of Neurology and Psychiatry, University ‘‘Sapienza’’, Viale dell’Università 30, 00185 Rome, Italy Marmara University Hospital School of Medicine, Department of Neurology, Fevzi Cakmak Mah. Mimar Sinan Cad. No: 41, 34899 Ust Kaynarca/Pendik, Istanbul, Turkey Department of Clinical Neurophysiology, Aarhus Universitets hospital, Nørrebrogade 44, 8000 Aarhus C, Denmark
a r t i c l e
i n f o
Article history: Accepted 15 November 2014 Available online xxxx Keywords: Cutaneous silent period A-delta fibres Axonal polyneuropathy Demyelinating polyneuropathy Neuropathic pain
h i g h l i g h t s The evaluation of cutaneous silent period (CSP) is a useful method to detect dysfunction of A-delta
fibres in patients with axonal polyneuropathy (PNP) as well as demyelinating PNP. The number of axons and conduction properties of A-delta fibres play respectively crucial roles in CSP
duration and CSP latencies. CSP parameters do not differ between patients with and without neuropathic pain.
a b s t r a c t Objective: To investigate the cutaneous silent period (CSP), a spinal inhibitory reflex mainly mediated by A-delta fibres, in demyelinating and axonal polyneuropathy (PNP) and evaluate whether CSP parameters differ between patients with and without neuropathic pain. Methods: Eighty-four patients with demyelinating PNP, 178 patients with axonal PNP and 265 controls underwent clinical examination, DN4 questionnaire, standard nerve conduction study, motor-root stimulation and CSP recordings from abductor digiti minimi. We calculated the afferent conduction time of CSP (a-CSP time) with the formula: CSP latency root motor evoked potential latency. Results: In the demyelinating PNP group the a-CSP time was significantly longer; in the axonal PNP group, CSP duration was shorter than the demyelinating group (p = 0.010) and controls (p = 0.001). CSP parameters were not different between patients with and without neuropathic pain. Conclusions: The abnormality of a-CSP time in the demyelinating PNP group suggests the crucial role of A-delta fibres in the mechanism of CSP; the shorter CSP duration in the axonal PNP group supports the strong influence of the number of axons on this parameter. Conclusions: Our study suggests that neuropathic pain could be related to pathophysiological mechanisms differing from mere A-delta fibre loss. Significance: CSP evaluation is effective in detecting A-delta fibre dysfunction in axonal as well as demyelinating PNP. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Polyneuropathy (PNP) is a neurological disorder caused by various conditions, with a prevalence of approximately 2.400 per 100.000 population (England and Asbury, 2004; England et al., ⇑ Corresponding author. Tel.: +39 0649914485; fax: +39 0649914576. E-mail address: [email protected] (M. Inghilleri).
2005; Martyn and Hughes, 1997). Peripheral neuropathy usually manifests with symmetric motor and sensory involvement. Sensory disturbances including hypesthesia, paraesthesia and neuropathic pain are associated with nociceptive pathway damage (Koltzenburg and Scadding, 2001). The diagnosis of PNP is based on a combination of clinical symptoms, signs and electrodiagnostic findings (Dyck et al., 1992, 1997; Feldman et al., 1994; Teunissen et al., 1997; Peripheral Nerve Society, 1995; Proceedings of a
http://dx.doi.org/10.1016/j.clinph.2014.11.013 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
2
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
consensus development conference on standardized measures in diabetic neuropathy, 1992). Standard neurophysiological techniques such as nerve conduction studies (NCSs) and somatosensory evoked potentials are useful in demonstrating damage along peripheral and central sensory pathways, but they do not assess the function of small fibres (Cruccu et al., 2004). Without techniques specific for the study of small fibres (i.e., experimental blocks, laser evoked potentials and contact-heat evoked potentials) or special organ stimulation (i.e., cornea, tooth pulp and glans), electrical stimuli unavoidably also excite large fibres, because large sized non-nociceptive afferents are characterised by a lower electrical threshold than small sized afferents (i.e., A-delta and C-nociceptive fibres) (Cruccu et al., 2004). Unfortunately, these techniques are expensive and require the use of specialised equipment not widely available in most electrophysiological laboratories. The evaluation of cutaneous silent period (CSP) represents a useful electrophysiological method for investigating the function of small fibres (Caccia et al., 1973; Inghilleri et al., 1997) by using standard electromyography equipment (Floeter, 2003). The CSP is defined as a transient suppression of voluntary muscle contraction following painful electrical stimulation of cutaneous sensory nerves. It represents a nociceptive reflex which is mediated mostly by A-delta afferent fibres (Shahani and Young, 1973; Uncini et al., 1991; Inghilleri et al., 1997, 2002; Floeter, 2003). These fibres activate an oligosynaptic circuit that in turn inhibits the motoneurons (Inghilleri et al., 1997). Although pre-synaptic inhibition has also been proposed as an explanation for electromyographic (EMG) suppression by A-delta afferents at spinal level (Leis et al., 1995, 1996), there is widespread agreement that motor neurons receive post-synaptic inhibition during the CSP and that the inhibition is transmitted through spinal inhibitory interneurons (Logigian et al., 1999; Inghilleri et al., 1997; Floeter, 2003). However, although the spinal circuitry of CSP is still not well known, the last-order interneuron is inhibitory and capable of producing a secure post-synaptic inhibition (Floeter, 2003). Although electrical stimuli used to evoke the CSP activate both large and small-diameter fibres, and a contribution to CSP generation of large-diameter afferents has also been suggested (Serrao et al., 2001), the long latency of the CSP is more compatible with a spinal reflex produced by slowly conducting afferents (Floeter, 2003). Moreover, normal CSPs were recorded in patients with profound large-diameter fibre sensory neuropathies with absent SNAPs or somatosensory evoked potentials (Uncini et al., 1991; Leis et al., 1992; Inghilleri et al., 1995). The CSP latency reflects the conduction function of A-delta afferents, efferent motor axons and synaptic delay; the CSP duration depends on the amount of activated fibres and on changes in nociceptive input (Leis et al., 1992, 1996; Inghilleri et al., 1997; Gilio et al., 2008). As a contrast from laser evoked potentials, that investigate nociceptive pathways up to cortex by activating A-delta and C fibres and by evoking scalp potential related to A-delta fibres and, through dedicated techniques, also C fibres (Treede, 2003), the evaluation of CSP could explore the sensory pathway from sensory endings to spinal level and the functional integrity of the spinal circuitry, in particular centromedullary connections, being activated by A-delta fibres (Stetkarova et al., 2001; Kofler et al., 2003a; Stetkarova and Kofler, 2009). CSPs have been demonstrated to be altered in different types of neuropathies (Leis, 1994; Syed et al., 2000; Corsi et al., 2002; Osio et al., 2004; Svilpauskaite et al., 2006a; Koo et al., 2010; Kim et al., 2010; Onal et al., 2010). Nevertheless, to our knowledge, no studies have evaluated the utility of CSP in a large number of patients with PNP. Additionally, the modification of CSP parameters in different
pathogenetic processes (i.e., axonal vs. demyelinating) of PNPs has not yet been analysed in comparative studies. In this study we aimed to determine A-delta fibre dysfunction in patients with demyelinating PNP vs. patients with axonal PNP by evaluating the CSP and other standard electrophysiological parameters of conventional NCSs. Furthermore, we evaluated whether CSP parameters varied between patients with and without neuropathic pain in either patient group. 2. Methods 2.1. Subjects Over a period of four years (from January 2008 to January 2012) 84 patients with demyelinating PNP (37 females, 47 males) and 178 patients with axonal PNP (83 females, 95 males) were consecutively recruited from the inpatient and outpatient clinics of the Department of Neurology and Psychiatry of ‘‘Sapienza’’ University in Rome. A group of 265 healthy subjects (123 females, 142 males) was also enrolled. The study protocol was in accordance with the Helsinki Declaration of Human Rights and was approved by the local Ethics Committee. Written informed consent was provided by all the patients and controls to participating in the study. 2.1.1. Control group Normative values for the electrophysiological parameters were obtained from 265 healthy controls. 2.1.2. Patients The diagnosis of axonal and demyelinating PNP was based on established criteria and recommendations of the American Academy of Neurology and the American Association of Neuromuscular and Electrodiagnostic Medicine (England and Asbury, 2004; England et al., 2005). Patients with demyelinating PNP met the clinical diagnostic criteria for Chronic Inflammatory Demyelinating Polyneuropathy (CIDP), with a symptomatic sensorimotor neuropathy involving upper and lower limbs for at least two months (Joint Task Force of the EFNS and the PNS, 2010); patients with monoclonal gammopathy, paraneoplastic, metabolic disorders such as diabetes, disimmune diseases such as sarcoidosis, systemic lupus erythematosus and angiitis were excluded from the study. In patients with axonal PNP, diabetes was the etiological cause for 77% of the group (137 patients), chronic idiopathic axonal polyneuropathy being the etiological cause for the remaining 23% (41 patients). Exclusion criteria for both groups were central nervous system diseases and cognitive impairment. All subjects enrolled were not taking medication for neuropathic pain and were able to achieve a state of interferential muscular contraction. 2.2. Clinical evaluations All patients were investigated with detailed tests for muscle strength and sensory examination. Muscle strength was assessed with the Medical Research Council (MRC) Score for Muscle Strength (Medical Research Council, 1976) with a cumulative score ranging between 0 and 80 for upper limbs (shoulder abduction, elbow flexion, elbow extension, wrist flexion, wrist extension, thumb opposition, finger flexion, finger extension) and between 0 and 70 for lower limbs (hip flexion, knee extension, ankle dorsiflexion, ankle plantar flexion, toe flexion, toe extension).
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
All patients were divided into two groups according to the presence or absence of neuropathic pain. Screening for neuropathic pain was performed using the DN4 questionnaire, a clinicianadministered qualitative screening tool that indicates neuropathic pain when the score is P4 (Bouhassira et al., 2005). All clinical evaluations and NCSs were performed by the same examiners in order to minimise investigator variations. 2.3. Electrophysiological evaluations All electrophysiological data were recorded using a Micromed Myoquick 1400 EMG machine (Micromed S.p.A., Treviso, Italy). The experimental procedures were completed with the subjects seated in a comfortable reclining chair with a mounted head rest. During the electrophysiological evaluation, skin temperature was controlled and maintained between 31 and 34 °C. Conventional surface electrode techniques were used. In all patients and healthy subjects electrical motor root stimulation, motor NCS and CSP were performed recording from right abductor digiti minimi (ADM) muscle. Orthodromic sensory NCS for median and ulnar nerves was respectively achieved by stimulation of second and fifth digits. Furthermore, in the context of CSP recordings, average pre-stimulus EMG amplitude was also calculated in order to evaluate and obtain a level of muscle contraction that was similar between patients and healthy subjects. The electrophysiological investigations also included the exploration of peroneal, medial plantar and sural nerves, but for the purpose of this study only upper limb-related neurophysiological responses were taken into account. The methods used were coherent to the recommendations of experts of the International Federation of Clinical Neurophysiology (Kimura, 2006). 2.3.1. Electrical motor-root stimulation Electrical motor-root stimulation was performed by high voltage transcutaneous electrical stimulation over the cervical vertebral column, at level of cervical enlargement of the spinal cord (Lange et al., 1992; Mills and Murray, 1986). Stimulation of the cervical region was accomplished by using a Digitimer Stimulator D180 (Digitimer Ltd., Welwyn Garden City, UK) which was able to deliver maximal electrical shocks of 750 V (50–100 ls decay time). Cervical electrodes were placed on the neck, the cathode slightly caudal to the C-7 spinous process and the anode midline between T1 and occiput (approximately at the level of C4 process). The subjects were instructed to relax during cervical stimulation. The initial intensity of stimulation corresponded to 150 V and increments of 20 V were applied. In this way the intensity was gradually increased until maximal root motor evoked potential (root-MEP) amplitude was achieved. The actual intensities used were about 350–400 V. The stimulation intensity was adjusted by selecting the stimulus voltage which produced the shortest response latency. Root-MEPs were recorded from ADM muscle by Ag/AgCl surface electrodes attached to the target muscle by the belly-tendon method. Filter settings were 5 Hz–5 kHz, sweep duration was 50 or 100 ms, and the sensitivity was 0.5–5 mV/division. Latency was measured from the stimulus artifact to the onset of the root-MEPs. The root-MEPs with the largest possible amplitudes and the shortest latencies were evaluated. 2.3.2. Motor NCS (M-NCS) The M-NCS was performed by stimulating the ulnar nerve at the wrist and the elbow and recording from ADM muscle. The level of the stimulus intensity was slowly increased until maximal compound muscle action potentials (CMAPs) were obtained. CMAPs were recorded by Ag/AgCl surface electrodes attached to the target muscle by the belly-tendon method. The filter settings
3
were 5 Hz–5 kHz, sweep duration was 30 ms, and the sensitivity was 1 mV/division. Distal motor latency (DML) to the onset of the negative peak, baseline to peak amplitude of distal CMAP (following wrist stimulation) and the motor nerve conduction velocity (M-NCV) (elbow to wrist) were determined. 2.3.3. Sensory NCS (S-NCS) The S-NCS was performed orthodromically from median and ulnar nerves. Percutaneous stimuli were delivered until adequate maximal sensory action potentials (SNAPs) were obtained; recordings were performed with surface electrodes. Filters were set at 5 Hz–5 kHz, sweep duration was 30 ms and sensitivity was 5 lV/ division. Distal sensory latency (DSL) at the valley of the positive deflection, positive to negative peak amplitude of SNAPs and the distal sensory nerve conduction velocity (S-NCV) (digit II or V to wrist) were determined. 2.3.4. CSP Single consecutive electrical stimuli (duration 0.5 ms) were delivered using ring electrodes placed around the second and third digits of the right hand with the cathode proximally positioned. In a preliminary study, the sensory threshold (STh) was explored in 20 patients with axonal PNP, 20 patients with demyelinating PNP and 20 healthy controls. The STh was defined as the minimum stimulus intensity perceived by the subject five times over five trials. With respect to our results, in all participants of our study the stimulus intensity was set at 80 mA, which corresponds to a stimulus intensity more than 20 times the mean STh of the three samples in our preliminary study. EMG signals were recorded from right ADM muscle through a pair of surface Ag/AgCl electrodes by the belly-tendon method. Patients and healthy subjects underwent painful electrical stimulation in the second and third digits. In order to match the level of background EMG activation, all patients maintained maximal voluntary isometric contraction of ADM muscle for about 1 s before and after finger stimulation, whereas the healthy subjects had to maintain a submaximal ADM contraction ranging between 30% and 50% of the maximal effort (which corresponded to a range of contraction in which there are no more fluctuations in CSP parameters) (Kofler et al., 2007). Filters were set at 50 Hz–5 kHz, sweep duration was 500 ms and sensitivity was 100 lV/division. In order to monitor the steadiness of contraction, EMG activity was displayed on a free-run trace and was also played over a loudspeaker to provide visual and auditory feedback to the subject and examiner. Signals were amplified, full-wave rectified and off-line averaged (10 trials). The average pre-stimulus EMG amplitude was calculated as the average value of the EMG signal during a 100 ms time window prior to the stimulus. The latency of the CSP was determined by inspection of the rectified EMG signal at the point in which the average EMG amplitude dropped below 50% of the pre-stimulus levels. In accordance with the recommendations of the International Federation of Clinical Neurophysiology (IFCN) (Kimura et al., 1994), we measured the CSP duration rather than the amplitude of suppression, because CSP duration is less dependent on the EGM background level (Romaniello et al., 2004).The duration of the CSP was calculated from the onset of the CSP up to the point where the EMG activity amplitude returned above 50% of the pre-stimulus level (Gilio et al., 2008). In the case of absent CSPs, CSP duration was considered equivalent to 0 ms and a null value was attributed to CSP latency.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
4
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
2.3.5. Afferent conduction time of CSP In all participants, we calculated a parameter which we referred to as ‘‘afferent CSP time’’ (a-CSP time). It demonstrates the difference between the CSP latency and the root-MEP latency. Thus, we can exclude the efferent time from CSP latency. This parameter facilitated the investigators in analysing the conduction of the afferent fibre pool (mostly A-delta), this proved particularly useful for the patients with unavailable SNAPs and available CSP responses.
2.3.6. Estimated afferent and efferent conduction velocity of CSP In order to estimate the afferent conduction velocity, we measured in each subject the distance between the proximal phalanx of the second and third digits and C-7 spinous process. We then divided this distance by the a-CSP time. In the case of the efferent conduction velocity of CSP, we measured in each subject the distance between the ulnar nerve at the wrist and the C-7 spinous process; we then divided this distance by the difference between root-MEP latency and DML.
2.4. Statistical analysis The Mann–Whitney U test was used to analyse the differences in non-normally distributed values. Differences in the frequency of neuropathic pain, demographic data and the absence of evocable CSPs or SNAPs across the different aetiologies were analysed with Fisher’s exact test. Spearman’s rank correlation coefficient was used for the correlation analysis between CSP parameters (CSP duration and a-CSP time) and NCS parameters. Normal distribution of demographic, clinical and electrophysiological parameters were tested with the Shapiro–Wilk normality test. All values are reported as means ± standard deviation (SD). The statistical significance limit was accepted as p < 0.05. A Bonferroni correction was applied for Mann–Whitney U test and Fisher’s exact test for multiple comparisons: a p value of less than 0.017 was considered in order to indicate significant differences. Data were analysed using the Statistical Package for Social Sciences (SPSS 19.0).
3.3. Electrophysiological evaluation 3.3.1. Control group Normative values of standard NCSs and CSP study obtained from our controls are shown respectively in Tables 2 and 3. The STh detected in the preliminary study in a sample of 20 subjects resulted 2.3 ± 0.3 mA (range 1.8–3.1 mA). The mean distance used to estimate the afferent conduction velocity of CSP (distance between the proximal phalanx of the second and third digits and C-7 spinous process) was 826 ± 42 mm. As for the estimation of efferent conduction velocity, the mean distance between the wrist and C-7 spinous process was 718 ± 45 mm. 3.3.2. Patients with axonal PNP The STh detected in the preliminary study was 3.1 ± 0.3 mA (range 2.5–3.5 mA). The axonal PNP group showed longer ulnar DML and DSL, rootMEP latency, and a-CSP time (Fig. 1), reduced SNAP and CMAP amplitudes, and slower M-NCV and S-NCV when compared to the healthy subjects (Tables 2 and 4). The CSP duration was shorter compared to the demyelinating PNP group and the healthy subjects (Table 3, Figs. 2 and 3). The absence of CSP was detected only in the group with axonal PNP [26 patients, 14.6%, vs. 0 patients with demyelinating PNP, vs. 0 healthy subjects (Fisher’s exact test, p < 0.001)]. We found absent SNAPs from both ulnar and median nerves in 44 patients (24.7%). This frequency was different from that of patients with demyelinating PNP [39 patients, 46.4% (Fisher’s exact test, p < 0.001)] and controls [0 subjects, 0% (Fisher’s exact test, p < 0.001)]. Among patients with absent SNAPs, only 14 (8%) also indicated the absence of CSP responses. The mean distances used to estimate the afferent and efferent conduction velocity of CSP were respectively 807 ± 54 mm and 704 ± 55 mm. Indirect estimation of the afferent and efferent conduction velocity of the fibres mediating the CSP respectively yielded values of 12.5 ± 2.1 m/s and 55.0 ± 7.3 m/s. The average pre-stimulus EMG amplitude analysis revealed no statistically significant differences compared to the other groups (Table 3). A-CSP time and CSP duration did not show any significant correlation with the other considered electrophysiological parameters (Table 4).
3. Results 3.1. Demographic values The demographic data of patients and healthy subjects are described in Table 1. No statistical differences were evident among the patients with axonal PNP and the patients with demyelinating PNP (p > 0.05 for all comparisons).
3.2. Clinical evaluation Clinical data of patients and healthy subjects are described in Table 1. The demyelinating PNP group suffered more loss of strength in upper and lower limbs as evidenced with MRC score compared to the axonal PNP group (p < 0.001). Patients with axonal PNP showed a more compromised pin-prick sensation (p < 0.001) and neuropathic pain was more present in comparison with patients with demyelinating PNP (52 patients, 29.2%, in axonal PNP group vs. 9 patients, 10.7%, in demyelinating PNP group, p < 0.001). The other clinical data did not show further differences among the patient groups (Table 1).
3.3.3. Patients with demyelinating PNP The STh detected in the preliminary study was 3.0 ± 0.4 mA (range 2.1–3.6 mA). In the demyelinating PNP group, the DMLs, DSLs, root-MEP latency and a-CSP time (Fig. 1) were significantly longer, the amplitudes of CMAP and SNAP were significantly lower, and the S-NCV and M-NCV were significantly slower compared to the axonal PNP group and the healthy subjects (Tables 2 and 4). The CSP duration was not different to the healthy subjects (Table 3, Figs. 2 and 3). CSP was always present, but 39 (46.4%) patients presented the absence of SNAPs from both ulnar and median nerves. This frequency was different from that of patients with axonal PNP [44 patients, 24.7% (Fisher’s exact test, p < 0.001)] and controls [0 subjects, 0% (Fisher’s exact test, p < 0.001)]. The mean distances used to estimate the afferent and efferent conduction velocity of CSP were respectively 803 ± 41 mm and 700 ± 41 mm. Indirect estimation of the afferent and efferent conduction velocity of the fibres mediating the CSP, respectively yielded values of 11.5 ± 3.5 m/s and 35.8 ± 9.2 m/s.
Please cite this article in press as: Lopergolo D et al. Cutaneous silent period recordings in demyelinating and axonal polyneuropathies. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.013
5
D. Lopergolo et al. / Clinical Neurophysiology xxx (2014) xxx–xxx Table 1 Demographic and clinical characteristics of patients and healthy subjects (mean ± SD). Characteristics
Demyelinating group (n = 84)
Axonal group (n = 178)
A vs. D p
Sex (F;M) Age (years) Upper limbs MRC Trophism Tactile sensation Pin-prick sensation Vibration sensation Lower limbs MRC Trophism Tactile sensation Pin-prick sensation Vibration sensation Neuropathic pain
37 F; 47 M 61.2 ± 10.1
83 F; 95 M 61.7 ± 10.6
74.4 ± 7.1 0.07 ± 0.3 1.3 ± 1.4 1.0 ± 1.3 2.9 ± 1.1
78.3 ± 5.5 0.1 ± 0.3 1.2 ± 1.5 1.8 ± 1.5 3.0 ± 0.7
