Med Biol Eng Comput DOI 10.1007/s11517-015-1292-9

ORIGINAL ARTICLE

Recording and assessment of evoked potentials with electrode arrays N. Miljkovic´1,2 · N. Maleševic´1,2 · V. Kojic´1,2 · G. Bijelic´2 · T. Keller3 · D. B. Popovic´1 

Received: 7 October 2013 / Accepted: 26 March 2015 © International Federation for Medical and Biological Engineering 2015

Abstract  In order to optimize procedure for the assessment of evoked potentials and to provide visualization of the flow of action potentials along the motor systems, we introduced array electrodes for stimulation and recording and developed software for the analysis of the recordings. The system uses a stimulator connected to an electrode array for the generation of evoked potentials, an electrode array connected to the amplifier, A/D converter and computer for the recording of evoked potentials, and a dedicated software application. The method has been tested for the assessment of the H-reflex on the triceps surae muscle in six healthy humans. The electrode array with 16 pads was positioned over the posterior aspect of the thigh, while the recording electrode array with 16 pads was positioned over the triceps surae muscle. The stimulator activated all the pads of the stimulation electrode array asynchronously, while the signals were recorded continuously at all the recording sites. The results are topography maps (spatial distribution of evoked potentials) and matrices (spatial visualization of nerve excitability). The software allows the automatic selection of the lowest stimulation intensity to achieve maximal H-reflex amplitude and selection of the recording/stimulation pads Electronic supplementary material  The online version of this article (doi:10.1007/s11517-015-1292-9) contains supplementary material, which is available to authorized users. * N. Miljković [email protected] 1

University of Belgrade - Faculty of Electrical Engineering, Bulevar kralja Aleksandra 73, 11000 Belgrade, Serbia

2

Tecnalia Serbia Ltd., Vladetina 13/6, Belgrade, Serbia

3

Tecnalia, Health Unit, Paseo Mikeletegi 1, 20011 San Sebastian/Donostia, Spain



according to predefined criteria. The analysis of results shows that the method provides rich information compared with the conventional recording of the H-reflex with regard the spatial distribution. Keywords  Asynchronous electrical stimulation · Evoked potential · H-reflex · Surface electrode array · Topography

1 Introduction When a mixed peripheral nerve is stimulated, causatively generated action potential propagates in both descending and ascending directions. Activation of the descending pathways at stimulation intensity above the motor threshold results with a direct motor response (M-wave), while the activation of ascending neurons generates a monosynaptic response of the same motor system (H-reflex) at supramaximal stimulation levels [23]. Tuning of the stimulation intensity changes the amplitudes of the evoked responses: Lowlevel stimulation current activates large neurons (afferent fibers) and generates the H-reflex and a small M-wave, whereas high-level stimulation activates all the fibers and H-reflex is eliminated by the collision with the antidromic motor valley. In order to obtain desirable reflex response, stimulation current tuning is a must. A standard clinical procedure for recording of the M-wave and H-reflex is as follows: placement of the recording electrodes over the muscle, placement of electrical stimulation (ES) electrodes over the appropriate nerve, and the adjustment of the stimulation parameters that results in the M-wave and H-reflex. In some cases and under certain measurement conditions, the H-reflex is very difficult or impossible to record [6]. H-reflex size can be affected by both technical and physiological factors that must be considered for valid

13



H-reflex measurement [4, 29]. The H-reflex is influenced by multiple neural pathways, and the H-reflex variation is a useful measure for analyzing neural mechanisms responsible for the control of movement [1, 6]. Conditioning of the H-reflex has been used for studying the effects of the central nervous system on the peripheral nervous system by transcranial electrical stimulation (TES), transcranial magnetic stimulation (TMS), and galvanic stimulation (GS) of the vestibular system [8, 18–20]. The placement of the stimulation electrodes for H-reflex generation differs from subject to subject, and it requires expertise related to nerve location since the anatomy differs between individuals. One of the first attempts to use a transcutaneous electrode array was reported in [10], in which a manual switchboard was used to select 8 out of 61 channels for stimulation. A manual 24-button transcutaneous electrode array box with circular electrode pads was introduced in [32]. The use of the electrode array for ES has proven to be more effective than conventional ES, because it can be adjusted to a specific user [25]. An electromyography (EMG) electrode array introduces an insight into the action potential distribution over much larger muscle areas than in the standard bipolar recording, and thus, it provides enhanced information on the spatial amplitude distribution [5, 22]. EMG amplitude topography is a method that produces a two-dimensional image with information on the spatial/amplitude distribution over the recorded muscle areas [22, 34]. In [22], the recording was performed by reference electrode distanced to recording site with bipolar signals produced by offline subtraction, whereas maps were calculated and plotted for both signals. Bipolar recordings performed from adjacent electrodes can be also used for calculation of EMG topography maps [34]. It has been proven that EMG topography maps created from root mean square (RMS) values contain valuable information regarding the anatomical and physiological characteristics of muscle, usually described relevantly to force alteration [22, 34]. We present an EMG topography visualization of signals recorded with reference electrode distanced to the recording site for studying the physiological changes related to the spatial/amplitude distribution of both the M-wave and the H-reflex and for estimation of nerve localization by the use of recording/stimulation electrode arrays. Regardless of the existing methodology recommendations, the optimal intensity level and positioning of the stimulation electrodes are still debatable [4, 23, 29]. It is of interest to have an instrument that allows the appropriate positioning and the appropriate adjustment of the stimulation intensity in an objective and unsupervised application. The goals of this paper are (1) to demonstrate the use of array electrodes for recording/stimulation of evoked potentials from muscles for obtaining optimal location and (2) to

13

Med Biol Eng Comput

present the use of array electrodes for spatial muscle activation and for estimation of nerve localization. We assumed that the electrode array would enable practitioners to record evoked motor- potentials easily, effectively, and unsupervisedly and that the novel method would provide a tool for estimation of nerve localization and for investigation of spatial distribution of evoked potentials. The designed method enables automatic detection of the optimal site for stimulation, and it provides rich information about the generated response. The optimal site selection is based on the criterion that the recorded signal has the highest amplitude compared with other stimulation sites. It has been shown that H-reflexes are depressed progressively, relative to the initial response, at stimulation frequencies above 0.1 Hz. This phenomenon is called postactivation depression (PAD) [7]. In this study, we show that PAD can be studied by means of dynamic EMG topography maps of the PTP values of H-reflexes recorded by electrode arrays. We presented the functioning of the proposed method for the detection of optimal stimulation/recording sites, by the means of electrode arrays for both stimulation and recording of the motor-evoked potentials, and for estimation of nerve localization, which is presented in a case study. The measurements were performed in three static postures: standing, prone, and sitting.

2 Methods and materials 2.1 Hardware and software setup The new instrument used in this study (Fig. 1) consists of an electrical stimulator, an amplifier of electrophysiological signals, an electrode array for stimulation, an electrode array for recording, a dedicated unit for synchronization with other hardware (e.g., TMS, TES, and GS), a motion sensor, and the actuators for external mechanical perturbations. The instrument is controlled by a computer, which includes an analog to digital (A/D) converter. The algorithm switching of the stimulation, optimal detection of the recording and stimulation sites, and data acquisition and processing were developed by using the LabVIEW software (National Instruments Inc., Austin, USA). In this study, we used a 6 × 10 cm rectangular recording electrode array with 16 silver elliptical pads (1 × 0.5 cm) in a 4 × 4 arrangement for the H-reflex measurements and a geometrically identical electrode array for the stimulation of the tibial nerve (Fig. 2). The recording electrode array pads are individually covered with AG2550 gel (Axelgaard, Manufacturing Co., Ltd., Denmark) and isolated

Med Biol Eng Comput Fig. 1  The instrument that uses electrode arrays for the stimulation and recording. A PC with an A/D converter controls the instrument. The optional hardware of the instrument includes sensors for the assessment of movement, an actuator for perturbation, and a unit for synchronization with other stimulation devices (e.g., TMS, TES, and GS). The sketch shows the placement for monopolar recordings of the H-reflex (H) and M-wave (M) on the triceps surae muscle when the tibial nerve is stimulated in subject ID = 3

Fig. 2  Stimulation and recording procedures for optimal site selection. The INTFES stimulator, designed and produced by Tecnalia Serbia Ltd., Belgrade, Serbia, has a Bluetooth link to the PC and a pad selection multiplexer for asynchronous stimulation of the pads of the stimulation electrode array. An A/D card is connected to the PC via a PCMCIA interface

from each other, whereas the stimulation pads are covered with AG702 gel (Axelgaard, Manufacturing Co., Ltd., Denmark). Both electrodes were designed and produced for research purposes by Tecnalia Serbia Ltd., Belgrade, Serbia (available on request). The stimulation anode was a rectangular PALS Platinum (Axelgaard, Manufacturing Co., Ltd., Denmark) electrode (7.5 × 10 cm), which was placed over the patella. We used 2 × 3 cm pre-gelled disposable,

surface Ag/AgCl Ambu Neuroline 720 electrodes (Ambu, Neuroline, Ballerup, Denmark) as the reference electrode. The position of the reference electrode (monopolar recordings) was in close proximity to the recording electrode array preferably over muscle tendon, as shown in Fig. 1. Placement of reference electrode is a result of our initial measurements performed in order to decrease artifacts. We did not aim at supervised and skilled use of multi-array

13



electrode placed according to anatomical landmarks, but rather to arbitrary positioning of the multilayer electrode (in the midst in medial/lateral direction and in the midst in proximal/distal direction between the ankle and knee). A ground electrode (GCB-geliMED KG, Bad Segeberg, Germany) was placed between the stimulation and recording electrode arrays in order to reduce stimulation artifact (Fig.  1). The signals that are equivalent to bipolar recordings can be obtained by subtracting the signals from two pads on the recording electrode (e.g., two neighboring electrodes in the caudal direction) [22]. EMG signals were amplified with the 16-channel Grass RPS 107 amplifier (Grass Technologies, An Astro-Med, Inc, West Warwick, USA). Gain was set to 500, and the band-pass filter frequency band was set from 2 to 3000 Hz. Signals were digitized at 5000 samples per second with a resolution of 16 bits (PCMCIA A/D card DAQCard-6062E, National Instruments, Inc, Austin, USA). We used an INTFES stimulator developed for research purposes by Tecnalia Serbia Ltd., Belgrade, Serbia [25]. Frequency of ES was set to three pulses per second (pps), pulse duration was set to 300 µs, and eight pulses were delivered to each site, since our initial measurements confirmed the parameter selection for reliable and repeatable measurement of H-reflex. To determine the optimal recording site, we calculated EMG amplitude difference between the second maximal and first minimal peak of the H-reflex (PTP). For each stimulation current, and for each stimulation site, we also calculated the PTP from the most appropriate recording site and compared them. The current range was chosen for each subject separately. We changed the current amplitude from an initial value of 5 mA (in 1-mA steps) to the point where the H-reflex started to decrease and/or the M-wave reached saturation level. All stimulation sites were activated asynchronously by the implementation of the multiplexer, as shown in Fig. 2. The stimulation site that in combination with the current amplitude produced the highest PTP was considered as the optimal one [1, 4]. If at any time the subject experienced painful sensations, we selected a lower intensity paradigm (70 % of maximal PTP of the H-reflex). For dynamic topography representation, we calculated the average matrix from three successive topography maps, referred to later as Bin 1, Bin 2, etc. The program also estimates the PTP values and topography maps for the M-waves in parallel with the maps for the H-reflexes. 2.4 Outcome measures The monopolar H-reflex consists of two local maxima and one global minimum in a specific order [14]. For estimation of PTP, ΔT, and ΔT HM, we measured the amplitude of maximal peak of the recorded H-reflex with longer latency (Fig. 1) and considered it a maximal peak of the H-reflex. We calculated ΔT as the difference between the H-reflex H H maximum tmax_ampl and the H-reflex minimum tmin_ampl from the stimulation artifact (Eq. 1). H H T = tmax_ampl − tmin_ampl

(1)

M The latency between the M-wave maximum tmax_ampl H and the H-reflex maximal peak tmax_ampl was determined (Eq. 2).

Med Biol Eng Comput Fig. 3  Monopolar H-reflex recorded with recording electrode array for stimulation with current amplitude of 19 mA, stimulation frequency of 3 Hz, and impulse duration of 300 µs for subject with ID = 5. Optimal stimulation site and current amplitude were chosen automatically. H-reflexes are visible on all sites. The sketch of the recording electrode array is presented in the right-hand panel. The signals are shown for the interval lasting 43 ms (from 22 to 65 ms after the stimulation pulse)

T

HM

=

H tmax_ampl

M − tmax_ampl

(2)

PTP values were estimated according to Eq. 3. H H PTP = Vmax_ampl − Vmin_ampl

(3)

H where Vmax_ampl (mV) is the amplitude of the maximal peak H of the H-reflex and Vmin_ampl (mV) is the value of the minimal peak of the H-reflex. In order to estimate recording pad location effect on H-reflex PTP, for optimally chosen stimulation pad and current amplitude, we calculated the relative difference of optimal recording pad with other recording pads as in Eq. 4 and then averaged the results for all subjects.   H − VH Vopt i rel_dif = abs × 100 (4) H Vopt

In Eq. 4, rel_dif (%) stands for variation of EMG electrode H stands for optimal PTP, and V H stands for placement, Vopt i PTP of non-optimal neighboring pads. 2.5 Procedure We presented new method for the assessment of evoked motor potentials from six healthy volunteers with ID = 1–6 (two females and four males with 28 ± 3.5 years of age, 184  ± 6 cm height, and 74 ± 10 kg mass). H-reflex was recorded on the dominant right side. All volunteers signed an informed consent approved by the local ethics committee. During the measurements, the subjects lay on a bed in a comfortable prone position with their head oriented to the

floor in anterior direction (Fig. 1). We aimed at avoiding the head rotation since it has been reported that the head position can modulate the motoneuron pool and thus affect the H-reflex amplitude [20]. The activation of antagonist was controlled by the palpation of anterior tendons. The insets in Fig. 1 show the signals that are used for the H-reflex assessment. We recorded data during two additional sessions, in which the subjects were standing and sitting with knee flexed at 90°. For one subject (ID = 6), we completed the measurements in prone, standing, and sitting positions during additional session with 2-day intersession interval to assess the day to day variability. This subject also underwent the standard clinical recording procedure [4] during both sessions as described in Sect. 2.2.

3 Results In Fig. 3, the recorded monopolar H-reflexes measured at the triceps surae muscle are presented in the left-hand panel. The sketch of the recording electrode array and its direction with respect to the muscle fibers is presented in the right-hand panel. Figure 4 shows the evoked potentials versus time for two stimulation sites (top panels) and normalized EMG topography maps calculated from the PTP values for two sets of H-reflexes (bottom panels). Figure 4a shows the colorcoded spatial distribution of the PTP values; red color corresponds to higher PTP values (at least 70 % of the maximum PTP) and blue color to lower PTP values. Figure 4b shows the topography map of non-optimally chosen stimulation site; hence, the H-reflex is not elicited. Sketches of the recording and stimulation electrode arrays are shown

13



Med Biol Eng Comput

Fig. 4  Recorded evoked potentials for two stimulation sites for subject with ID = 5 (top panels). The signals are shown for 43 ms (from 22 to 65 ms after the stimulation pulse). Normalized EMG topography maps calculated from PTP values of H-reflex over triceps surae

muscle: a Spatial distribution of the amplitude is color coded; red corresponds to high PTP values and blue to low PTP values. b Topography map when the stimulation site was not optimal

in the right-hand panel in Fig. 5. Topography maps for all stimulation sites are shown in the left-hand upper panel for the optimal stimulation intensity of 19 mA. The lower lefthand panel shows the topography map for the case when stimulus was delivered to the (d, d) pad. The maps mostly colored in red present higher excitability for stimulation of the corresponding pads. FigSUP. 1 in supplementary material shows the stimulation matrices generated from

stimulation at five current amplitudes from 15 to 19 mA (steps of 1 mA) for subject ID = 5. The optimization presented here is achieved by sequential activation of the elements of the stimulation electrode array (Fig. 5). In Fig. 6, the topography matrices for the M-wave and H-reflex are presented for the stimulation current intensity of 17 mA. In Fig. 7, a dynamic topography map measured in one subject is shown. First and second topography maps

13

Med Biol Eng Comput Fig. 5  EMG topography maps for all 16 stimulation sites for subject with ID = 5 (left-hand upper panel): normalized topography map for the one stimulation pad [marked (d, d)] and framed in pink rectangular is presented in the left-hand lower panel. The sketch of the placement of the stimulation and recording electrode arrays on the lower leg is in the right-hand panel with framed stimulation pad (d, d) that corresponds to magnified topography map

of the recorded H-reflexes from the electrode array are presented. Other maps are plotted as three successive averaged topography maps (Bin 1, Bin 2, etc.). Overall number of displayed H-reflexes is 68 in Fig. 7. In Fig. 8, topography maps from three postures are shown. Each map is normalized to 70 % of the maximal value in the corresponding position. Averaged PTP values for optimal recording pads over all pulses delivered of H-reflexes with standard deviations in Fig. 8 are 2.60 ± 0.98, 0.61 ± 0.23, and 5.25 ± 0.67 mV. PTP values obtained from additional session in this subject are presented in supplementary material. The procedure presented in Fig. 5 by topography maps and matrices can be

understood as the physical relocation of the stimulation electrode. In parallel, the electrode array allows the shift of the stimulation site during the measurements in cases where it is of interest to record the H-reflex while the subject moves (e.g., a change in the posture from prone to a sitting or standing position presented in Fig. 8). For each subject and current amplitude during the stimulation of the optimal stimulation site and recording from the optimal recording site, we calculated the maximal PTP values, ΔT, and ΔT HM parameters. The optimal recording pad did not vary significantly among subjects, while the optimal stimulation pad varied in relation to current intensity. The results showed that two or three adjacent stimulation pads produced optimal

13



Med Biol Eng Comput

Fig. 6  Stimulation matrices for M-wave and H-reflex for current amplitude of 17 mA in subject ID = 4. A color bar with normalized PTP values is presented in the right-hand panel

Fig. 7  Dynamic topography of normalized topography maps of PTP values calculated from H-reflexes recorded with recording electrode array in subject ID = 4 with stimulation intensity of 17 mA, and

active stimulation pad was (a, c). Bin presents three successive averaged maps of recorded signals. A color bar of normalized PTP topography maps is given in the right-hand panel

H-reflex. ΔT and ΔT HM have average SD values of 11.77 and 2.91 %, respectively. PTP values had SD of 41.86 % on average. The placement of EMG electrodes produces PTP variation of on average 35.37 ± 19.16 % (maximum value is 60 %, and minimum is 0.54 %) with respect to neighboring pads (parameter calculation is described in Sect. 2). TableSUP 1 from supplementary material also contains the results recorded during standard clinical procedure for subject ID = 6.

4 Discussion

13

Recorded H-reflexes presented along the array sketch in Fig. 3 have different amplitudes, which correspond to variations of the EMG amplitude in relation to the electrode placement [16]. It is reported that on average, the changes in conditioned H-reflex by GS of vestibular system can be 5 % of the H-reflex amplitude [20]. We showed much

Med Biol Eng Comput

Fig. 8  Normalized topography maps of PTP values calculated from PTP of H-reflexes recorded with recording electrode array in subject ID = 6 in three positions: prone, sitting, and standing, for an electri-

cal stimulation with current intensity of 22 mA. A color bar of normalized PTP topography maps is given in the right-hand panel

higher variation comparing neighboring sites on the recording electrode array, which indicates that special care should be taken when placing recording electrodes. Averaged PTP values, recorded from the optimal recording site, showed relatively large standard deviations in all subjects, which is probably a consequence of the PAD phenomenon [27], since we delivered 3 pps. The difference in PTP values among optimally stimulated pad and neighboring pads is affected by both pad localization and dimension. Namely, the recording array electrodes cover large area of triceps surae muscle, which results both in cross talk from gasfchanges across baseline sessionstrocnemius and soleus muscles and in attenuation from electrically inactive tissue. The recording pads used in this study were larger than in previously reported work using recording electrode arrays [11–13, 28], since other researchers were investigating properties of motor units wherewith high spatial resolution is needed. It has been stated that the reflexes in gastrocnemius muscle are significantly smaller than in soleus muscle [2]. This finding can be used for future advanced application design, where the automatic algorithm would differentiate various muscles and potentially contribute to advanced cross talk reduction. Nevertheless, one should have in mind that PTP differentiation cannot be performed in this study and compared with previous results, since the authors in [2] used 0.2 pps (pulses per second) in comparison with 3 pps used here. The visualization of topography maps allows the investigation of the evoked potential distribution over the recorded area of muscle [22, 34] as presented in Figs. 4, 6, 7, 8 in this study. The nonlinear distribution is probably a consequence of the muscle fiber orientation. The

recommendation for bipolar electrode placement is that the electrodes should be placed over muscle belly, parallel to the direction of the muscle fibers [16]. In this study, we assessed the optimal placement of the recording electrodes for monopolar recording and only the muscle belly localization was of interest. If spatial distribution of latencies can estimate muscle fiber direction as shown in [24], and spatial distribution of amplitude can provide muscle belly localization, then the recording pads could be chosen/ placed optimally by both criteria. Optimal stimulation site selection besides maximization of the H-reflex amplitude included minimization of painful sensation. The topography matrix shows how each topography map corresponds to relocation of appropriate stimulation pad, as shown in the upper panel in Fig. 5. It can be concluded that the amplitude of H-reflex varies with the location of the stimulation site and the current amplitude [4, 23, 29]. The algorithm tunes the stimulation current intensity automatically in a way that the output recorded at the electrode array results in high-amplitude H-reflex with the minimum stimulation intensity. We assumed that the unpleasant painful sensation could be decreased by automatic current tuning and by providing to the subject a current level control (“panic” button which resulted in current decrease as described in Sect. 2). No substantial difference was observed when H-reflex was recorded during standard clinical procedure and by the proposed method, except for the H-reflex PTP (4.24. ± 1.92 mV by standard and 2.60 ± 0.98 mV by proposed novel method). Maximal values of H-reflex PTP differed correspondingly. This PTP disagreement is probably a consequence of applied pressure to the stimulation site by

13



clinician and stimulation cathode dimension. The manual placement with applied pressure probably reaches different nerve areas (whereas afferent fibers are bundled within the nerve), and correspondingly, the stimulation is delivered to higher number of afferent fibers. Though skilled clinicians can easily place the electrodes, the presented method provides unsupervised use of an objective tool for automatic recording/stimulation electrode placement. In addition to optimal electrode location, the topography maps and matrices allowed studying various aspects of H-reflex measurement. The stimulation matrix representation (Fig. 6) can be used for studying M-wave and H-reflex PTP values in parallel to recordings performed while for one active stimulation pad at the time and current tuning: As M-wave has high PTP, the H-reflex has smaller PTP, and vice versa. This is likely the consequence of the recruitment characteristics responsible for the M-wave and H-reflex [30] and of spatial distribution/dependence of M-wave and H-reflex PTPs. Further assessment might allow the study of the spatial recruitment of the nerve fibers both ascending and descending, since the method allows the simultaneous visualization of M-wave and H-reflex. Typically, as a result of PAD, relative to the initial response, the second reflex is depressed by 80 % at frequencies of 4–10 Hz, and it can be reduced further by frequencies above 10 Hz. PAD of the H-reflex caused by changes of stimulation frequency has been reported by means of PTP values of the H-reflex recorded with a bipolar EMG setup and corresponds to our results [7, 17, 26]. We showed that, by dynamic topography presented in this study, PDA as one of the commonly known H-reflex phenomenon can be visualized concerning the spatial distribution of evoked potentials (Fig. 7). It is known that the amplitude of the H-reflex changes due to posture [31], and researchers are likely to change the stimulation site when analyzing the evoked potentials in various postures. In Fig. 8, we showed the topography maps in one subject during three different postures: prone, sitting, and standing. Although the optimal stimulation site for the generation of the H-reflex changes with the variation of current intensity, the optimal recording site does not change as it was shown in [4, 23, 29]. ΔT and ΔT HM remained relatively constant with small SD values as it was expected. See Supplementary materials for more information. The application of the proposed method in various body postures has potentials for studying reflexes in non-anatomical positions in patients. The method presented in this study enables visualization of spatial distribution of evoked potentials and an estimation of peripheral nerve location. Electrical nerve stimulation has been a common practice for nerve localization when initiating peripheral nerve block as local anesthesia [15]. It has been stated that the use of this

13

Med Biol Eng Comput

technique can be challenging due to definition of anatomical landmarks in growing children and that superficial nerve mapping with the means of ES can assist sound anatomical knowledge [3]. We assume that the use of the presented method for estimation of nerve localization can be used for nerve mapping as a preceding step during peripheral nerve block application. The proposed method can also be used for the assessment of peripheral nerve localization as a complementary objective measure to ultrasonography [21]. Limitations and main considerations of the study: 1. Additional stratification with ultrasound imaging or other objective assessment procedures used for estimation of nerve localization were not performed, and they should be included in future work. 2. We reported H-reflex PTP for M-wave set to 20 % of its maximal value, but H-reflex amplitude can be also normalized (e.g., Hmax/Mmax ratio [23] or conditioned H-reflex changes across baseline sessions [33]) or presented as a result of different protocol (e.g., Hmax PTP [27]). 3. The method presented in this study for visualization of evoked potentials by topography maps and for estimation of nerve localization should be further explored, especially in patients. 4. We performed the monopolar recording of the H-reflex, whereas future studies should explore bipolar configuration or precise replacement of reference electrode in relation to inter-electrode spacing [9].

5 Supplementary material Supplementary material contains a FigSUP. 1 with the five stimulation matrices for subject with ID = 5 showing optimal stimulation pad displacement due to current intensity tuning. Normalized stimulation matrices of PTP values of H-reflexes recorded in subject with ID = 5 when the stimulation current ranged from 15 to 19 mA (1 mA steps) are presented in FigSUP. 1. The sketch of the placement of the stimulation electrode array with respect to the popliteal fossa is shown in the bottom right-hand panel in FigSUP. 1. In FigSUP. 2, topography maps from three postures during 2 sessions in subject ID = 6 are shown. TableSUP 1 in supplementary material presents optimal stimulation pad location; optimal recording pad location is presented, averaged maximal PTP, averaged ΔT, and averaged ΔT HM parameters with their standard deviations (SD) for each subject. Additional parameters for subject ID = 6 from repeated measurement during session 2 and from standard clinical procedure are presented in TableSUP 1. Stimulation current amplitude and the corresponding maximal PTP, ΔT, RF, and SF are presented too.

Med Biol Eng Comput Acknowledgments  The work on this project was supported partly by the Ministry of Education, Science and Technological Development, Republic of Serbia, Grant OI175016. The authors would like to thank all the volunteers involved in this study and PhD MD Laszlo Schwirtlih, MD Radoje Čobeljić, PhD MD Ksenija Ribarić, and MT Mihajlo Tancˇić for their valuable help in the design of the method and measurements. We would like to acknowledge both Tecnalia Serbia Ltd. for designing the electrode arrays for stimulation and EMG measurements and for providing us with the stimulator. The authors would like to thank Prof. Roberto Merletti for his valuable advice on the preparation of the manuscript and Prof. Milan R. Dimitrijević for his valuable advice during all phases of the study.

References 1. Aagaard P, Simonsen EB, Andersen JL, Magnusson P, DyhrePoulsen P (2002) Neural adaptation to resistance training: changes in evoked V-wave and H-reflex responses. J Appl Physiol 92:2309–2318 2. Alrowayeh HN, Sabbahi MA, Etnyre B (2011) Similarities and differences of the soleus and gastrocnemius H-reflexes during varied body postures, foot positions, and muscle function: multifactor designs for repeated measures. BMC Neurol 11:65 3. Bosenberg AT, Raw R, Boezaart AP (2002) Surface mapping of peripheral nerves in children with a nerve stimulator. Pediatr Anesth 12(5):398–403 4. Brinkworth RSA, Tuncer M, Tucker KJ, Jaberzadeh S, Turker KS (2007) Standardization of H-reflex analysis. J Neurosci Methods 162(1–2):1–7 5. Campanini I, Merlo A, Degola P, Merletti R, Vezzosi G, Farina D (2007) Effect of electrode location on EMG signal envelope in leg muscles during gait. J Electromyogr Kinesiol 17(4):515–526 6. Chen YS, Zhou S (2011) Soleus H-reflex and its relation to static postural control. Gait Posture 33(2):169–178 7. Clair JM, Anderson-Reid JM, Graham CM, Collins DF (2011) Postactivation depression and recovery of reflex transmission during repetitive electrical stimulation of the human tibial nerve. J Neurophysiol 106(1):184–192 8. Costa J, Guzman J, Vallderiola F, Rumia J, Tolosa E, Casanova-Molla J, Valls-Sole J (2011) Modulation of the soleus H reflex by electrical subcortical stimuli in humans. Exp Brain Res 212(3):439–448 9. De Luca CJ, Kuznetsov M, Gilmore LD, Roy SH (2011) Interelectrode spacing of surface EMG sensors: reduction of crosstalk contamination during voluntary contractions. J Biomech 45(3):555–561 10. Elsaify A (2005) A self-optimizing portable FES system using an electrode array and movement sensors. PhD Thesis, Department of Engineering, University of Leicester, UK 11. Farina D, Enoka RM (2011) Surface EMG decomposi tion requires an appropriate validation. J Neurophysiol 105(2):981–982 12. Farina D, Falla D (2008) Estimation of muscle fiber conduction velocity from two-dimensional surface EMG recordings in dynamic tasks. Biomed Signal Process 3(2):138–144 13. Farina D, Negro F, Gazzoni M, Enoka RM (2008) Detecting the unique representation of motor-unit action potentials in the surface electromyogram. J Neurophysiol 100(3):1223–1233 14. Gerilovsky L, Tsvetinov P, Trenkova G (1989) Peripheral effects on the amplitude of monopolar and bipolar H-reflex potentials from the soleus muscle. Exp Brain Res 76(1):173–181 15. Hadzic A, Vloka JD, Claudio RE, Hadzic N, Thys DM, Santos AC (2005) Electrical nerve localization: effects of cutaneous electrode placement and duration of the stimulus on motor response. Anesthesiology 100(4):976–981

16. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G (2000) Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 10(5):361–374 17. Imura S, Kishikawa M, Wada T, Iwai A, Fuziwara M (1997) Changes in the H-reflex amplitude causes by fluctuation of stimulus frequency. J Phys Ther Sci 9(2):111–119 18. Inghilleri M, Lorenzano C, Conte A, Frasca V, Manfredi M, Berardelli A (2003) Effects of transcranial magnetic stimulation on the H reflex and F wave in hand muscles. Clin Neurophysiol 114(6):1096–1101 19. Kennedy PM, Cresswell AG, Chua R, Inglis JT (2004) Vestibulospinal influences on lower limb motoneurons. Can J Physiol Pharmacol 82(8–9):675–681 20. Kennedy PM, Inglis JT (2002) Interaction effects of galvanic vestibular stimulation and head position on the soleus H reflex in humans. Clin Neurophysiol 113(11):1709–1714 21. Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS (2013) Correlation of nerve ultrasound, electrophysiological, and clinical findings in post Guillain–Barre syndrome. J Peripher Nerv Syst 18(3):232–240 22. Kleine BU, Schumann NP, Stegeman DF, Scholle HC (2000) Surface EMG mapping of human trapezius muscle: the topography of monopolar and bipolar surface EMG amplitude and spectrum parameters at varied forces and in fatigue. Clin Neurophysiol 111(4):686–693 23. Knikou M (2008) The H-reflex as a probe: pathways and pitfalls. J Neurosci Methods 171(1):1–12 24. Kojić V, Miljković N, Malešević N, Popović DB (2012) H-reflex recorded by multi-pad EMG electrodes. 11th Symposium on Neural Network Applications in Electrical Engineering, NEUREL 2012. IEEE Press, Belgrade, pp 119–122 25. Malešević NM, Popović Maneski LZ, Ilić V, Jorgovanović N, Bijelić G, Keller T, Popović DB (2012) A multi-pad electrode based functional electrical stimulation system for restoration of grasp. J Neuroeng Rehabil 9:66 26. McNulty PA, Jankelowitz SK, Wiendels TM, Burke D (2008) Postactivation depression of the soleus H reflex measured using threshold tracking. J Neurophysiol 100(6):3275–3284 27. McNulty PA, Shiner CT, Thayaparan GK, Burke D (2012) The stability of Mmax and Hmax amplitude over time. Exp Brain Res 218(4):601–607 28. Merletti R, Holobar A, Farina D (2008) Analysis of motor units with high-density surface electromyogram. J Electromyogr Kinesiol 18(16):879–890 29. Misiaszek JE (2003) The H-reflex as a tool in neurophysiology: its limitations and uses in understanding nervous system function. Muscle Nerve 28(2):144–160 30. Palmieri RM, Ingersoll CD, Hoffman MA (2004) The Hoffmann reflex: methodologic considerations and applications for use in sports medicine and athletic training research. J Athl Train 39(3):268–277 31. Pinniger GJ, Nordlund MM, Steele JR, Cresswell AG (2001) H-reflex modulation during passive lengthening and shortening of the human triceps surae. J Physiol 534(3):913–923 32. Popović-Bijelić A, Bijelić G, Jorgovanović N, Bojanić D, Popović MB, Popović DB (2005) Multi-field surface electrode for selective electrical stimulation. Artif Organs 29(6):448–452 33. Thompson AK, Chen XY, Wolpow JR (2009) Acquisi tion of a simple motor skill: task-dependent adaptation plus long-term change in the human soleus H-reflex. J Neurosci 29(18):5784–5792 34. Tucker K, Falla D, Graven-Nielsen T, Farina D (2009) Electromyographic mapping of the erector spinae muscle with varying load and during sustained contraction. J Electromyogr Kines 19(3):373–379

13

Recording and assessment of evoked potentials with electrode arrays.

In order to optimize procedure for the assessment of evoked potentials and to provide visualization of the flow of action potentials along the motor s...
1MB Sizes 0 Downloads 8 Views