- Email: [email protected]
Influence of reference electrode position on the compound muscle action potential
Journal Pre-proofs Influence of reference electrode position on the compound muscle action potential Sanjeev D. Nandedkar, Paul E. Barkhaus PII: DOI: Reference:
S1388-2457(19)31295-7 https://doi.org/10.1016/j.clinph.2019.11.008 CLINPH 2009039
To appear in:
Clinical Neurophysiology
Received Date: Revised Date: Accepted Date:
29 December 2018 22 September 2019 10 November 2019
Please cite this article as: Nandedkar, S.D., Barkhaus, P.E., Influence of reference electrode position on the compound muscle action potential, Clinical Neurophysiology (2019), doi: https://doi.org/10.1016/j.clinph. 2019.11.008
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 International Federation of Clinical Neurophysiology. Published by Elsevier B.V. All rights reserved.
Influence of reference electrode position on the compound muscle action potential Sanjeev D Nandedkar, Ph.D.1, Paul E Barkhaus, M.D.2
1
Natus Neurology, Hopewell Junction, New York, USA
2
Departments of Neurology, Medical College of Wisconsin, Milwaukee, Wisconsin USA
Corresponding author: Sanjeev D Nandedkar, 15 Dartantra Drive, Hopewell Junction, NY 12533 Email: [email protected] Phone: + 1 8452168107
Keywords: Compound muscle action potential; CMAP; Reference electrode; Volume conduction; Nerve conduction studies; Muscle origin.
Highlights A new montage, called ‘Proximal E2’, is proposed for nerve conduction studies. It reduces contribution of the reference (E2) electrode to the compound muscle action potential (CMAP). The effect of new montage on CMAP waveform and measurements is discussed.
Abstract Objective: When the compound muscle action potential (CMAP) is recorded in motor nerve conduction studies, the reference (E2) electrode can make a significant contribution to the CMAP. This study investigates the E2 recorded signal and its effect on CMAP measurements when E2 electrode is placed at different sites. Methods: The CMAP was recorded using the active electrode on the muscle belly and 4 different E2 electrodes placed at distal and proximal sites. The signal recorded by each electrode was also measured using a reference electrode on the contralateral limb. Signals were recorded from the abductor pollicis brevis, abductor digiti minimi, tibialis anterior and biceps muscles. Results: The E2 recorded a smaller signal when it was placed near or off the proximal tendon or muscle origin. This affected CMAP latency, duration for tibialis anterior. Contrary to expectation, initial upward deflection was noted for E2 signal.
1
Conclusion: A proximal E2 position records a lower volume conducted signal and yields a CMAP that is more representative of the muscle over which the E1 (active) electrode is placed. Significance: The proposed ‘Proximal E2’ montage may be better suited to assess pathology.
2
Introduction In motor nerve conduction studies the compound muscle action potential (CMAP) is characterized by its amplitude, latency, duration, etc. The CMAP is recorded using a differential amplifier which requires three recording electrodes; E1, E2, and E0 (Robinson et al, 2016). All voltage measurements in the amplifier are made relative to the E0 (or “ground”) electrode. This electrode is also necessary to reduce background noise and interference. The E1 (or “active”) electrode is placed over the muscle belly. The E2 (or “reference”) electrode is conventionally placed over a supposedly electrically quiet or “inactive” location. In most instances this is at the distal muscle tendon or even further distally. Hence the CMAP recorded is thought to primarily represent the electrical activity of the muscle over which the E1 electrode is placed, and the E2 considered to be minimally contributory to the signal. In actual practice, the E2 electrode records significant voltage from other muscles innervated by the stimulated nerve (Kincaid et al, 1993; Brashear and Kincaid, 1996; Nandedkar and Barkhaus, 2007, Barkhaus et al, 2011; Soono et al, 2011). This is most easily appreciated in tibial motor nerve conduction studies recorded from the foot muscle (abductor hallucis). The objective of this study is to investigate the effect of the E2 position on the CMAP waveform and its measurements. By reducing E2 contribution, the CMAP will better represent the electrical activity from the tested muscle, both in normal muscle and muscles affected by neuromuscular disorders (Higashihara et al, 2010, Kawamore et al, 2013, Nandedkar et al, 2015).
Methods Studies were performed at the second author’s institution where it was approved by the local institutional review board. All subjects gave written consent. Ten presumed healthy subjects (based on their medical history) participated in the study (Age 25 – 69; 8 females and 2 males). Recordings were made using a Nicolet Synergy EDX system (Natus, Middleton, WI, USA) with 8 channel amplifier, and disposable surface electrodes (Natus Part # 019-415200). The single or multi-channel motor nerve conduction program was used to record the CMAP at supramaximal stimulation intensity. The band-pass filter settings were 3 Hz and 10,000 Hz for the low and high frequencies, respectively. The notch filter was not used. Stimulus duration was 0.2 milliseconds for all recordings. The following nerve and muscle combinations were used for CMAP recordings (figures 1 and 2): fibular (peroneal) – tibialis anterior (TA), ulnar – abductor digiti minimi (ADM), median – abductor pollicis brevis (APB), and musculocutaneous – biceps brachii (BB). The stimulation site was wrist for median and ulnar nerve, above fibular head for fibular nerve, and axilla for musculocutaneous nerve. These recordings were used for analysis of CMAP waveform. Additionally, the median nerve was stimulated at elbow, ulnar nerve at above elbow, and fibular nerve at below fibular head to calculate conduction velocity. Proximal stimulation of musculocutaneous nerve at the Erb’s point was not tested. We do not calculate conduction velocity for this nerve using two points of stimulation; and find stimulation in axilla sufficiently selective to study the amplitude and latency of the evoked response. All subjects underwent bilateral studies.
3
We also performed similar studies in a few patients with amyotrophic lateral sclerosis (ALS).
The E1 electrode was placed over the muscle belly that was identified using anatomic landmarks and muscle palpation. We tested 4 different positions of the E2 electrode (see table 1 for summary). The first position was a few centimeters distal to the distal tendon. This is the standard position of E2 electrode in our laboratory, and we will denote it as ‘distal (D)’. Many laboratories place the E2 electrode over the distal tendon. This is commonly referred to as ‘belly-tendon’ montage. We will refer to this E2 position as ‘distal tendon (DT)’. The third position was at the proximal tendon and it is labelled as ‘PT’. The fourth position is called the ‘proximal (P)’ which was a few centimeters proximal and/or sideways from the PT site. These positions are shown in figures 1 and 2. The placements for E2 for the fibular nerve are distal tendon at ankle (‘DT’), more distal to DT on the dorsum of foot (‘D’), proximal origin of the tibialis anterior muscle at the knee (‘PT’), and at the medial aspect of the knee over the patella (‘P’). In the hand muscles, the ‘P’ position will be on the wrist, close to the stimulation site. Therefore, we used a site at the center of the dorsal wrist that was just distal to ulnar styloid process for ‘P’. The distance between stimulation site and P or PT electrodes was more than 3 cm. For the musculocutaneous nerve, the ‘DT’ site is at the middle anterior elbow, and the ‘D’ is on the olecranon. The ‘PT’ is at the anterior shoulder and the ‘P’ site is on the posterior upper shoulder. The distance between electrodes varied slightly depending upon the length of subject’s hand and foot. The distance between D and DT was roughly 4 cm for APB and ADM; 5 cm (arm thickness at elbow) for biceps; 5 cm for TA (Figures 1, 2). The distance between P and PT was roughly 4 cm for APB and ADM; 8 cm (along skin surface) for biceps and TA (figures 1 and 2) We began testing by performing a single channel CMAP recording with the E2 electrode placed at our laboratory standard positions: ‘DT’ for median and ulnar nerve, ‘D’ for musculocutaneous nerve and ‘P’ for fibular nerve. Appropriate muscle contraction and limb movement were observed to ensure adequate and proper nerve stimulation. With the musculocutaneous nerve stimulation, the muscle contraction was further verified by percutaneous palpation of the biceps brachii contraction. The stimulus intensity was supramaximal, i.e. roughly 15% higher than the level that gave maximal amplitude response. Using this intensity level, the E1 position was adjusted by a few millimeters to ascertain the location that gave the largest CMAP amplitude. Three to five different positions, along and perpendicular to muscle fiber direction, were tested while the electrode remained on the muscle belly. Attention was also paid to make sure that the waveform had an initial sharp upward (negative) deflection, and of the general quality of the potential (i.e. minimal irregularity suggesting co-activation of other muscle). This E1 position was used for all subsequent recordings. A 4 channel montage was used to study the effect of E2 position on the CMAP. The E1 and E0 electrodes were common to all channels. Electrodes were placed at sites D, DT, PT and P serving as the E2 electrode for the four channels. The CMAP was recorded simultaneously from all channels. As example, the 4 CMAPs recorded from the TA are shown in figure 3B. The electrode combination used for recording is also indicated. The onset latency (marker 1), amplitude of negative peak (markers 1 - 2) and duration of negative peak (markers 1 – 3) were measured. The markers were placed automatically by the system and we did not change their position. 4
In the last recording, we used a 5 channel configuration with E1, D, DT, PT and P sites as the ‘active’ electrode. An electrode placed on the contralateral limb, labelled ‘C’ (figure 1), served as the ‘reference’ (E2) input for all channels. This site is remote from the tested muscles and does not record any significant volume conducted potential. Hence it can be used to assess the signals recorded by these individual electrodes on the tested side (Kincaid et al, 1993; Brashear and Kincaid, 1996; Nandedkar and Barkhaus, 2007; Barkhaus et al, 2011). As example, the 5 recordings from the TA are shown in figure 3A. The electrode combination used for recording is also indicated. The top trace represents the signal recorded by the E1 electrode placed on the muscle belly. The bottom 4 traces are potentials recorded by the E2 electrode at different positions described above. Subtracting these from the top trace shows the CMAP for the corresponding combination of recording electrodes. As example, subtracting the 3rd trace (DT – C) from top trace (E1 – C) gives the CMAP for E1 – DT recording configuration (2nd trace on right). The signals in this recording (other than the E1-C derivation) had variable shapes (figure 3A). Some had initial upward deflection while others had an initial downward deflection. Therefore, the peak-to-peak amplitude of signals was measured for analysis. These measurements were performed manually. The amplitude, duration and latency results were tabulated and graphed for analysis using an Excel spreadsheet. Kruskal-Wallis statistical method was used to assess difference in various measurements. It is customary to use a statistical significance level of p<0.05. Due to multiple comparisons, we applied Bonferroni correction and used a level of p<0.017 for significance. For each muscle, the amplitude differences of E2 recorded signal was compared using all positions (4 groups, 3 degrees of freedom (DOF)) and also by comparing two positions at a time (DOF=1). The CMAP measurements, i.e. latency, amplitude and duration, were also compared in a similar manner.
Results Studies from control subjects were analyzed quantitatively. For each nerve type (i.e. median, ulnar, etc), the side-to-side comparison of amplitude, latency and duration showed no statistically significant difference. Analysis of data from left side only is described. Sample recordings from the fibular, ulnar and musculocutaneous nerves are shown in figures 3, 4 and 5, respectively. The position of the E2 electrode did not affect CMAP signal quality and artifact free CMAPs could be recorded with distal stimulation. Figure 6 shows the distribution of peak-to-peak amplitude of the E2 recorded signal for the 4 muscles. In all muscles the amplitude is high, reaching several millivolts. The APB recording had the lowest amplitude compared to other muscles. The table inset summarizes the mean and standard deviation of amplitude values. Comparison of all 4 positions together showed a significant amplitude difference among E2 sites (p<0.001). Comparing individual positions, the P and PT positions gave significantly less amplitude (p<0.001) compared to D and DT sites, with the exception of site PT in the median nerve (indicated by * in the table inset).
5
E2 recorded potentials had variable configurations with either initial upward or initial downward deflections (arrows in figures 3, 4, 5). The pattern of amplitude, latency and duration differences was not the same for all muscles, as described below. In fibular nerve studies, proximal placement of E2 (positions P and PT) gave a CMAP with shorter onset latency and shorter negative peak duration (p<0.01) compared to E2 placed at ‘D’ and ‘DP’ (Figure 3). In ulnar nerve conduction studies, negative peak of the CMAP coincided with the positive peak of the signal recorded by E2 electrode when the E2 was placed distally. This is indicated by the ● symbol in figure 4. This pattern was seen in all recordings (including the right side). Musculocutaneous nerve studies were challenging and sometime gave waveforms with complex morphology (figure 5 indicated by * and #). Median nerve studies showed no significant differences using different E2 positions. Pooled recordings from both sides were used to assess waveforms and artifacts with proximal stimulation. With proximal stimulation and E2 placed at ‘D’ or ‘DT’ locations, the CMAP waveform had initial upward deflection for all nerves. With E2 placed at ‘P’ or ‘PT’ sites, a low amplitude initial positive deflection was seen for median nerve conduction studies. It was large enough to cause incorrect marker placement for the CMAP onset, and a very high conduction velocity (> 90 m/sec) in 4 nerves (figure 7A). When the onset marker position was adjusted manually to the peak of the positive phase, i.e. onset of the negative spike, the velocity was similar to that using ‘D’ and ‘DT’ positions. This artifact was not seen for any fibular nerve conduction studies (Figure 7C). Only 1 ulnar nerve showed this artifact and it was seen unilaterally. There was no difference in velocity values for ulnar and fibular nerves using different E2 sites. In all nerves, regardless of the E2 position, the amplitude on proximal stimulation was similar or up to 20% smaller than the CMAP recorded with distal stimulation. The CMAPs recorded with ulnar nerve stimulation where E2 had been placed at ‘D’ and ‘P’ positions in 3 patients with ALS are compared in figure 8. In A, the waveforms show similar amplitude. However, in B and C, the amplitude using ‘P’ position is much smaller.
Discussion This study demonstrates the significant E2 electrode contribution to CMAP as described by other investigators (Kincaid et al, 1993; Brashaer and Kincaid, 1997; Nandedkar and Barkhaus, 2007). In addition, we also show the variable effect of E2 electrode on the CMAP waveform and its measurements depending upon its placement. Ideally, the CMAP should reflect only the electrical activity from the muscle recorded by the E1 electrode. This is not possible due to signals recorded from current E2 placements. One strategy to achieve this could be to place E2 on the contralateral limb (figures 3, 4 and 5). However, this gives noisy surface EMG recordings (SIP) in procedures such as motor unit number index (MUNIX) that analyzes 6
both CMAP and SIP (Nandedkar et al, 2010). On proximal stimulation, the E1 records a large amplitude signal from other co-stimulated muscles that is not cancelled by the contralateral E2 signal. Therefore, our strategy was to place the E2 close enough to the E1 electrode to minimize noise and artifact, and to record the least signal from other muscles. Historically the E2 position has always been distal to the E1 electrode. The reason for this choice is not clear. Perhaps by moving the E2 electrode away from the stimulation site the stimulus artifact was reduced, particularly when motor conductions were originally described and recording equipment was more prone to artifact (Hodes et al, 1948). In our recordings using modern amplifiers, the proximal E2 position gave high quality recordings with good separation between the CMAP and stimulus artifact (figures 3, 4, 5). The high E2 signal amplitude at distal sites may be due to its proximity to the tendons of multiple muscles innervated by the tested nerve. As example, the deep fibular nerve innervates the TA, extensor digitorum longus (EDL), and extensor hallucis longus (EHL) muscles in the anterior compartment of the leg. They are all stimulated when attempting to record the TA CMAP. Although the EDL and EHL are relatively lateral to the E1 electrode, their tendons are close to the E2 electrode when placed at or distal to the ankle (Leis and Trapani, 2000; Barkhaus et al, 2004). Their potentials not only add to the CMAP amplitude, but their temporal dispersion also increases the CMAP duration. Conversely, the origins of TA, EDL, and EHL are more separated. The origin of TA is on the lateral condyle and upper body of tibia, while the EDL and EHL fibers arise from the proximal and distal anterior shaft of fibula, respectively. This may explain the lower amplitude signal recorded using proximal sites for E2. There is an additional anatomical consideration regarding the TA muscle. Anatomic studies (Aquilonius et al, 1984) have shown that this muscle has several motor point zones. This is corroborated in our study, where moving the E1 from proximal to distal (mid-leg) shows relative lack of changes in CMAP amplitude. Based on these considerations we suggest position P (i.e., medial patella) of E2 and proximal positioning of E1 (proximal third) for recording TA. We have been doing this already for many years as a result of our experiences in MUNIX and have adopted this specific proximal recording site for TA MUNIX studies. (Neuwirth et al, 2011). Our findings in TA are consistent with those described by Escorcio-Bezerra and co-workers (2019). We were also surprised by the shape of the E2 signal. When an electrode is over the endplate zone, the signal has an initial upward deflection. Any downward initial deflection is interpreted as volume conducted signals from E1 electrode placement that is off the end-plate region. By this argument, all of our E2 recordings should have an initial positive deflection. But our recordings showed both initial positive and negative deflections (indicated by arrows in figures 3, 4, and 5). This positivity was also observed by Phongsamart and colleagues (2002) in ulnar nerve conduction studies. We have no explanation for this variable pattern. When the E2 signal is initially negative it will cancel the initial negative potential from the E1 electrode. Hence the CMAP onset will be seen later and it will have a longer distal latency. If the E2 signal is initially positive, it will add to the negativity of E1 signal by virtue of differential amplification. Hence the CMAP amplitude will be artificially higher. This is also seen in the fibular nerve studies (figures 3 and 6).
7
The ulnar nerve recordings show large amplitude peaks for the E1 and E2 signal. The peak occurs later in the E2 signal (indicated by ● in figure 4). The peak of E2 also coincided with the CMAP peak used for amplitude measurements. To properly study the individual muscle involvement of ulnar-innervated muscles in pathology (e.g., trauma, split hand distribution in motor neuron disease, etc.), the distal recording site of E2 may not be optimal (Higashihara et al, 2010, Sonoo et al, 2011). Greater sensitivity may be achieved by using proximal E2 sites. We suggest the ‘P’ site for recordings. The ‘PT’ site also gives low amplitude signals. However, the short distance between the resulting E1 and E2 position will make it inconvenient when the patient has a small hand such as in children. The median nerve recordings had low amplitude E2 signals. Hence, the position of E2 electrode does not have a large effect on the CMAP latency, amplitude and duration. Therefore, the waveforms of median conduction study are not presented. In the biceps recording the E2 signal waveform morphology varied at distal E2 positions. Under standard recording conditions, one cannot recognize the individual contributions of the E1 and E2 electrodes to the CMAP. When the signal does not have expected morphology, one tries different E1 positions. If the E2 contribution is significant, repositioning the E1 electrode may not help. This may explain some of the difficulties with CMAP recordings in the musculocutaneous nerve studies. Figure 5 showed a complex short duration waveform (indicated by *) that is much different from the expected waveform shown as dashed line in Figure 5B. When the multi-channel recording was made, the large amplitude E2 signal with distal positions (D and DT) is seen quite easily. At DT position the signal is so large that the CMAP looked inverted (indicated by #). Another technical point goes back to anatomical studies (Aquilonius et al, 1984). They demonstrated a single endplate zone for the biceps brachii. In this study, if E1 is wellplaced at mid-belly and the CMAP amplitude is good, only minimal movement of E1 is required to optimize it. In consideration of these issues, we suggest that a proximal position (“PT”) for E2 be used. The ‘P’ position gives the lowest amplitude E2 signal for the median nerve study. Hence it is optimal for the study of CMAP amplitude using distal stimulation. However, it is suboptimal when proximal stimulation is performed. It may give incorrect velocity values. With fibular and ulnar nerve, the ‘P’ position gave artifact free recordings with distal and proximal stimulation in practically all subjects. The shape of the CMAP on proximal stimulation was very similar to that from distal stimulation. The amplitude on proximal stimulation was similar or slightly smaller than distal stimulation using ‘P’ site. This indicates that ‘P’ position of E2 can be used for routine ulnar and fibular conduction studies also. We have been using this position for fibular conduction study for many years, and have not experienced any issues with proximal stimulation. Our E2 position is close to the proximal tendon or the origin of the muscle. Hence we propose the term “Proximal E2” for this montage to distinguish it from the traditional “Belly-Tendon” montage. We use the term “Proximal” as opposite of “Distal” placement of E2 electrode, and not necessarily the position at or off the proximal tendon. Using the suggested positions of E2 will require new definitions of normal limits generated from a prospective study. Nevertheless, we feel that the use of the proposed montage gives a cleaner signal that is minimally influenced by E2 and is more representative of the muscle being recorded by E1. We have made very limited comparisons of CMAPs using different E2 positions in patient groups. Figure 8 shows a comparison of CMAPs recorded using the ‘D’ (traditional position) and 8
‘P’ (the new proposed position) of E2. The amplitudes are similar for the two positions in figure 8A. Recordings in B and C were made from patients with very weak muscles (strength 3/5 using Medical Research Council scale). We were surprised to see a CMAP with amplitude of 1.2 – 1.4 mV using position ‘D’ of E2 electrode. However, the amplitude is much less using ‘P’ position (< 0.5 mV), and it better reflects the muscle weakness. We emphasize that these are very preliminary observations, and future studies will be needed to fully assess the benefits of the ‘P’ position. Proximal E2 position is away from the ‘moving’ part of the limb during stimulation. This may also result in less artifacts in repetitive nerve stimulation.
Declaration of interest The first author is an employee of Natus Neurology. This work represents no conflict. The second author has no conflict to report. Acknowledgments The first author thanks Natus Neuro for support in this project.
9
References Aquilonius SM, Askmark H, Gillberg PG, Nandedkar S, Olsson Y, Stalberg E: Topographical localization of motor endplates in cryosections of whole human muscles. Muscle Nerve, 1984; 7:287-93. Barkhaus PB, Nandedkar SD and Roberts M: “Electronic Myoanatomic Atlas for Clinical Electromyography (Second Edition)”. EMG on DVD series: Volume I. Nandedkar Productions LLC, Hopewell Junction, NY, 2004. Barkhaus P, Kincaid J, Nandedkar S: Amplitude drop on proximal stimulation in the tibial motor nerve conduction study. Muscle Nerve, 2011; 44:776-782. Brashear A, Kincaid JC: The influence of the reference electrode on CMAP configuration: leg observations and an alternate E2 site. Muscle Nerve, 1996; 19:63-67. Escorcio-Bezerra ML, Abrahao A, Nunes KF, Sparapni FV, Braga NI, Robinsom LR, Zinman LManzano G: Optimal E2 (Reference) electrode placement in fibular motor nerve conduction studies recording from the tibialis anterior muscle. Muscle Nerve, 2019; 59:249-253. Higashihara M, Sonoo M, Imafuku I, Ugawa Y, Tsuji S: Origin of ulnar compound muscle action potential investigated in patients with ulnar neuropathy at the wrist. Muscle Nerve, 2010; 41: 704-706. Hodes R, Larrabee MG, German W: The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons. Arch Neuropsychiat, 1948; 60:340-365. Kawamore Y, Sonoo M, Higashihara M, Chiba T, Hatanaka Y: Origin of surface motor unit potentials in hypothenar motor unit number estimation. Muscle Nerve, 2013; 48:185-190. Kincaid JC, Brashear A, Markand ON: The influence of the reference electrode on CMAP configuration. Muscle Nerve, 1993; 16:392-396. Leis AA and Trapani VC: “Atlas of Electromyography”. Oxford University Press, New York, 2000. Nandedkar SD, Barkhaus PE: Contribution of reference electrode to the compound muscle action potential. Muscle Nerve, 2007; 36:87-92. Nandedkar SD, Barkhaus PE, Stalberg EV: Motor unit number index (MUNIX): Principle, method and findings in healthy subjects and in patients with motor neuron disease. Muscle Nerve, 2010; 42: 796807. Nandedkar SD, Barkhaus PE, Stalberg EV: Cumulative motor index (CMI): an index to study progression of amyotrophic lateral sclerosis. J Clin Neurophysiol, 2015; 32:79-85 Neuwirth C, Nandedkar S, Stålberg E, Barkhaus P, de Carvalho M, Furtula J, Van Dijk J, Baldinger R, Castro J, Costa J, Otto M, Sandberg A, Weber M: Motor unit number index (MUNIX): Reference values of five different muscles in healthy subjects from a multi-center study. Clin Neurophysiol, 2011; 122:18951898. 10
Phongsamart G, Wertsch J, Ferdjallah M, King J, Foster D: Effect of reference electrode position on the compound muscle action potential (CMAP) onset latency. Muscle Nerve, 2002; 25:816-821. Robinson LR, Christie M, Nandedkar SD: A message from ground electrode. Muscle Nerve, 2016; 54:1010-1011. Soono M, Kurokawa K, Higashihara M, Kurono H, Hokkoku K, Hatanaka Y, Shimizu T: Origin of far field potentials in the Ulnar nerve compound muscle action potentials. Muscle Nerve, 2011; 43:671-678.
11
Table 1: Summary of E2 electrode placements. The underlined positions are suggested for CMAP recordings. (Also see Figures 1 and 2). * For median nerve ‘P’ position is optimal for distal stimulation only. The ‘D’ position is better suited when proximal stimulation is also performed.
Nerve 1: Distal ‘D’ Fibular Dorsum Foot Median Thumb* Ulnar Digit V Musculocutaneous Olecranon
2: Distal Tendon ‘DT’ Dorsum Ankle Base Thumb Base Digit V Anterior Elbow
E2 Positions 3: Proximal tendon ‘PT’ Lateral Knee Distal Wrist Distal Wrist Anterior Shoulder
4: Proximal ‘P’ Medial patella Center Dorsal Wrist* Center Dorsal Wrist Posterior upper Shoulder
12
Figure legends Figure 1: Electrode placement for musculocutaneous and fibular motor nerve conduction studies. The star symbol shows E2 position that gave the lowest amplitude and hence suggested E2 position for future recordings. Refer to text for details. Figure 2: Electrode placement for median and ulnar motor nerve conduction studies. The star symbol shows E2 position that gave the lowest amplitude and hence suggested E2 position for future recordings. For the median nerve, the ‘D’ position gives higher E2 amplitude but less artifact on proximal stimulation. Refer to text for details. Figure 3: Fibular nerve study. (A) The signals recorded by the E1, D, DT, PT and P as active electrodes are shown. The E2 reference electrode was placed on the opposite leg. Arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. (B) The CMAPs using the 4 different E2 positions are shown. Note the shorter latency and duration of CMAP using the proximal site (PT and P) for the E2 electrode. (C) The table shows mean value and standard deviation (SD) of onset latency and duration for the E2 positions. The latency and duration are shorter (p<0.01) using P and PT sites compared to D and DT sites. This is also seen from waveforms in B. Figure 4: Ulnar nerve study. (A) The signals recorded by the E1, D, DT, PT and P as active electrodes are shown. The E2 (reference) electrode was on the opposite hand. The arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. The ● symbol shows the peak of the E2 recorded signal. (B) The CMAPs recorded using the 4 different E2 positions are shown. Figure 5: Musculocutaneous nerve study. (A) Signals recorded by the E1, D, DT, PT and P as active electrodes are shown on the left. The E2 (reference electrode) is on the opposite arm. The arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. (B) The CMAP recorded using the 4 different E2 positions are shown on the right (solid lines). One CMAP shows a short duration and complex waveform (indicated by *) with two distinct negative peaks using a distal E2 position. The waveform marked # shows a mainly positive waveform with a very low amplitude initial negative peak. In A, we can see high amplitude signals recorded by E2 electrode at distal positions while the signal recorded by E1 is small. Their phase cancellation gives the complex CMAP with short duration and low amplitude for the initial negative peak. The atypical waveforms in this recording are compared with the expected CMAP waveform recorded in another subjects (top traces in dashed line). Figure 6: Peak to peak amplitude of E2 recorded signal at 4 different sites (D, DT, PT and P) is compared for 4 muscles: (A) Abductor pollicis brevis (B) Abductor digiti minimi, (C) Biceps brachii, and (D) Tibialis anterior. All plots show the same scale for amplitude. Amplitude is affected by E2 position (p<0.001, DOF=3). While the amplitude is high, it is lowest for the APB muscle. The table inset shows the mean value and standard deviation (SD) of amplitude. The amplitude is lower (p < 0.001) using P and PT sites compared to D and DT sites (except for PT site in median nerve indicated by *).
13
Figure 7: CMAP recordings with distal and proximal stimulation of (A) Median (B) Ulnar and (C) Fibular nerve are compared. The traces on top are recorded with E2 placed at ‘P’. Traces below are recorded with E2 placed at ‘D’. Marker positions are shown as identified by the machine. In the median nerve, a low amplitude positive phase (indicated by arrow) is an artifact from muscles in the forearm. This gave very high conduction velocity. Upon manually correcting the onset marker, the velocity values were similar for the two E2 positions. In ulnar and fibular nerves, no technical issues are seen.
Figure 8: CMAP recordings from the abductor digiti minimi muscle of 3 patients with amyotrophic lateral sclerosis. Upon superimposition, traces recorded with E2 placed at ‘D’ are shown as dashed line. The solid lines show CMAP with E2 placed at ‘P’. Note the relative amplitude differences.
14
15
16
17
18
S1388-2457(19)31295-7 https://doi.org/10.1016/j.clinph.2019.11.008 CLINPH 2009039
To appear in:
Clinical Neurophysiology
Received Date: Revised Date: Accepted Date:
29 December 2018 22 September 2019 10 November 2019
Please cite this article as: Nandedkar, S.D., Barkhaus, P.E., Influence of reference electrode position on the compound muscle action potential, Clinical Neurophysiology (2019), doi: https://doi.org/10.1016/j.clinph. 2019.11.008
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 International Federation of Clinical Neurophysiology. Published by Elsevier B.V. All rights reserved.
Influence of reference electrode position on the compound muscle action potential Sanjeev D Nandedkar, Ph.D.1, Paul E Barkhaus, M.D.2
1
Natus Neurology, Hopewell Junction, New York, USA
2
Departments of Neurology, Medical College of Wisconsin, Milwaukee, Wisconsin USA
Corresponding author: Sanjeev D Nandedkar, 15 Dartantra Drive, Hopewell Junction, NY 12533 Email: [email protected] Phone: + 1 8452168107
Keywords: Compound muscle action potential; CMAP; Reference electrode; Volume conduction; Nerve conduction studies; Muscle origin.
Highlights A new montage, called ‘Proximal E2’, is proposed for nerve conduction studies. It reduces contribution of the reference (E2) electrode to the compound muscle action potential (CMAP). The effect of new montage on CMAP waveform and measurements is discussed.
Abstract Objective: When the compound muscle action potential (CMAP) is recorded in motor nerve conduction studies, the reference (E2) electrode can make a significant contribution to the CMAP. This study investigates the E2 recorded signal and its effect on CMAP measurements when E2 electrode is placed at different sites. Methods: The CMAP was recorded using the active electrode on the muscle belly and 4 different E2 electrodes placed at distal and proximal sites. The signal recorded by each electrode was also measured using a reference electrode on the contralateral limb. Signals were recorded from the abductor pollicis brevis, abductor digiti minimi, tibialis anterior and biceps muscles. Results: The E2 recorded a smaller signal when it was placed near or off the proximal tendon or muscle origin. This affected CMAP latency, duration for tibialis anterior. Contrary to expectation, initial upward deflection was noted for E2 signal.
1
Conclusion: A proximal E2 position records a lower volume conducted signal and yields a CMAP that is more representative of the muscle over which the E1 (active) electrode is placed. Significance: The proposed ‘Proximal E2’ montage may be better suited to assess pathology.
2
Introduction In motor nerve conduction studies the compound muscle action potential (CMAP) is characterized by its amplitude, latency, duration, etc. The CMAP is recorded using a differential amplifier which requires three recording electrodes; E1, E2, and E0 (Robinson et al, 2016). All voltage measurements in the amplifier are made relative to the E0 (or “ground”) electrode. This electrode is also necessary to reduce background noise and interference. The E1 (or “active”) electrode is placed over the muscle belly. The E2 (or “reference”) electrode is conventionally placed over a supposedly electrically quiet or “inactive” location. In most instances this is at the distal muscle tendon or even further distally. Hence the CMAP recorded is thought to primarily represent the electrical activity of the muscle over which the E1 electrode is placed, and the E2 considered to be minimally contributory to the signal. In actual practice, the E2 electrode records significant voltage from other muscles innervated by the stimulated nerve (Kincaid et al, 1993; Brashear and Kincaid, 1996; Nandedkar and Barkhaus, 2007, Barkhaus et al, 2011; Soono et al, 2011). This is most easily appreciated in tibial motor nerve conduction studies recorded from the foot muscle (abductor hallucis). The objective of this study is to investigate the effect of the E2 position on the CMAP waveform and its measurements. By reducing E2 contribution, the CMAP will better represent the electrical activity from the tested muscle, both in normal muscle and muscles affected by neuromuscular disorders (Higashihara et al, 2010, Kawamore et al, 2013, Nandedkar et al, 2015).
Methods Studies were performed at the second author’s institution where it was approved by the local institutional review board. All subjects gave written consent. Ten presumed healthy subjects (based on their medical history) participated in the study (Age 25 – 69; 8 females and 2 males). Recordings were made using a Nicolet Synergy EDX system (Natus, Middleton, WI, USA) with 8 channel amplifier, and disposable surface electrodes (Natus Part # 019-415200). The single or multi-channel motor nerve conduction program was used to record the CMAP at supramaximal stimulation intensity. The band-pass filter settings were 3 Hz and 10,000 Hz for the low and high frequencies, respectively. The notch filter was not used. Stimulus duration was 0.2 milliseconds for all recordings. The following nerve and muscle combinations were used for CMAP recordings (figures 1 and 2): fibular (peroneal) – tibialis anterior (TA), ulnar – abductor digiti minimi (ADM), median – abductor pollicis brevis (APB), and musculocutaneous – biceps brachii (BB). The stimulation site was wrist for median and ulnar nerve, above fibular head for fibular nerve, and axilla for musculocutaneous nerve. These recordings were used for analysis of CMAP waveform. Additionally, the median nerve was stimulated at elbow, ulnar nerve at above elbow, and fibular nerve at below fibular head to calculate conduction velocity. Proximal stimulation of musculocutaneous nerve at the Erb’s point was not tested. We do not calculate conduction velocity for this nerve using two points of stimulation; and find stimulation in axilla sufficiently selective to study the amplitude and latency of the evoked response. All subjects underwent bilateral studies.
3
We also performed similar studies in a few patients with amyotrophic lateral sclerosis (ALS).
The E1 electrode was placed over the muscle belly that was identified using anatomic landmarks and muscle palpation. We tested 4 different positions of the E2 electrode (see table 1 for summary). The first position was a few centimeters distal to the distal tendon. This is the standard position of E2 electrode in our laboratory, and we will denote it as ‘distal (D)’. Many laboratories place the E2 electrode over the distal tendon. This is commonly referred to as ‘belly-tendon’ montage. We will refer to this E2 position as ‘distal tendon (DT)’. The third position was at the proximal tendon and it is labelled as ‘PT’. The fourth position is called the ‘proximal (P)’ which was a few centimeters proximal and/or sideways from the PT site. These positions are shown in figures 1 and 2. The placements for E2 for the fibular nerve are distal tendon at ankle (‘DT’), more distal to DT on the dorsum of foot (‘D’), proximal origin of the tibialis anterior muscle at the knee (‘PT’), and at the medial aspect of the knee over the patella (‘P’). In the hand muscles, the ‘P’ position will be on the wrist, close to the stimulation site. Therefore, we used a site at the center of the dorsal wrist that was just distal to ulnar styloid process for ‘P’. The distance between stimulation site and P or PT electrodes was more than 3 cm. For the musculocutaneous nerve, the ‘DT’ site is at the middle anterior elbow, and the ‘D’ is on the olecranon. The ‘PT’ is at the anterior shoulder and the ‘P’ site is on the posterior upper shoulder. The distance between electrodes varied slightly depending upon the length of subject’s hand and foot. The distance between D and DT was roughly 4 cm for APB and ADM; 5 cm (arm thickness at elbow) for biceps; 5 cm for TA (Figures 1, 2). The distance between P and PT was roughly 4 cm for APB and ADM; 8 cm (along skin surface) for biceps and TA (figures 1 and 2) We began testing by performing a single channel CMAP recording with the E2 electrode placed at our laboratory standard positions: ‘DT’ for median and ulnar nerve, ‘D’ for musculocutaneous nerve and ‘P’ for fibular nerve. Appropriate muscle contraction and limb movement were observed to ensure adequate and proper nerve stimulation. With the musculocutaneous nerve stimulation, the muscle contraction was further verified by percutaneous palpation of the biceps brachii contraction. The stimulus intensity was supramaximal, i.e. roughly 15% higher than the level that gave maximal amplitude response. Using this intensity level, the E1 position was adjusted by a few millimeters to ascertain the location that gave the largest CMAP amplitude. Three to five different positions, along and perpendicular to muscle fiber direction, were tested while the electrode remained on the muscle belly. Attention was also paid to make sure that the waveform had an initial sharp upward (negative) deflection, and of the general quality of the potential (i.e. minimal irregularity suggesting co-activation of other muscle). This E1 position was used for all subsequent recordings. A 4 channel montage was used to study the effect of E2 position on the CMAP. The E1 and E0 electrodes were common to all channels. Electrodes were placed at sites D, DT, PT and P serving as the E2 electrode for the four channels. The CMAP was recorded simultaneously from all channels. As example, the 4 CMAPs recorded from the TA are shown in figure 3B. The electrode combination used for recording is also indicated. The onset latency (marker 1), amplitude of negative peak (markers 1 - 2) and duration of negative peak (markers 1 – 3) were measured. The markers were placed automatically by the system and we did not change their position. 4
In the last recording, we used a 5 channel configuration with E1, D, DT, PT and P sites as the ‘active’ electrode. An electrode placed on the contralateral limb, labelled ‘C’ (figure 1), served as the ‘reference’ (E2) input for all channels. This site is remote from the tested muscles and does not record any significant volume conducted potential. Hence it can be used to assess the signals recorded by these individual electrodes on the tested side (Kincaid et al, 1993; Brashear and Kincaid, 1996; Nandedkar and Barkhaus, 2007; Barkhaus et al, 2011). As example, the 5 recordings from the TA are shown in figure 3A. The electrode combination used for recording is also indicated. The top trace represents the signal recorded by the E1 electrode placed on the muscle belly. The bottom 4 traces are potentials recorded by the E2 electrode at different positions described above. Subtracting these from the top trace shows the CMAP for the corresponding combination of recording electrodes. As example, subtracting the 3rd trace (DT – C) from top trace (E1 – C) gives the CMAP for E1 – DT recording configuration (2nd trace on right). The signals in this recording (other than the E1-C derivation) had variable shapes (figure 3A). Some had initial upward deflection while others had an initial downward deflection. Therefore, the peak-to-peak amplitude of signals was measured for analysis. These measurements were performed manually. The amplitude, duration and latency results were tabulated and graphed for analysis using an Excel spreadsheet. Kruskal-Wallis statistical method was used to assess difference in various measurements. It is customary to use a statistical significance level of p<0.05. Due to multiple comparisons, we applied Bonferroni correction and used a level of p<0.017 for significance. For each muscle, the amplitude differences of E2 recorded signal was compared using all positions (4 groups, 3 degrees of freedom (DOF)) and also by comparing two positions at a time (DOF=1). The CMAP measurements, i.e. latency, amplitude and duration, were also compared in a similar manner.
Results Studies from control subjects were analyzed quantitatively. For each nerve type (i.e. median, ulnar, etc), the side-to-side comparison of amplitude, latency and duration showed no statistically significant difference. Analysis of data from left side only is described. Sample recordings from the fibular, ulnar and musculocutaneous nerves are shown in figures 3, 4 and 5, respectively. The position of the E2 electrode did not affect CMAP signal quality and artifact free CMAPs could be recorded with distal stimulation. Figure 6 shows the distribution of peak-to-peak amplitude of the E2 recorded signal for the 4 muscles. In all muscles the amplitude is high, reaching several millivolts. The APB recording had the lowest amplitude compared to other muscles. The table inset summarizes the mean and standard deviation of amplitude values. Comparison of all 4 positions together showed a significant amplitude difference among E2 sites (p<0.001). Comparing individual positions, the P and PT positions gave significantly less amplitude (p<0.001) compared to D and DT sites, with the exception of site PT in the median nerve (indicated by * in the table inset).
5
E2 recorded potentials had variable configurations with either initial upward or initial downward deflections (arrows in figures 3, 4, 5). The pattern of amplitude, latency and duration differences was not the same for all muscles, as described below. In fibular nerve studies, proximal placement of E2 (positions P and PT) gave a CMAP with shorter onset latency and shorter negative peak duration (p<0.01) compared to E2 placed at ‘D’ and ‘DP’ (Figure 3). In ulnar nerve conduction studies, negative peak of the CMAP coincided with the positive peak of the signal recorded by E2 electrode when the E2 was placed distally. This is indicated by the ● symbol in figure 4. This pattern was seen in all recordings (including the right side). Musculocutaneous nerve studies were challenging and sometime gave waveforms with complex morphology (figure 5 indicated by * and #). Median nerve studies showed no significant differences using different E2 positions. Pooled recordings from both sides were used to assess waveforms and artifacts with proximal stimulation. With proximal stimulation and E2 placed at ‘D’ or ‘DT’ locations, the CMAP waveform had initial upward deflection for all nerves. With E2 placed at ‘P’ or ‘PT’ sites, a low amplitude initial positive deflection was seen for median nerve conduction studies. It was large enough to cause incorrect marker placement for the CMAP onset, and a very high conduction velocity (> 90 m/sec) in 4 nerves (figure 7A). When the onset marker position was adjusted manually to the peak of the positive phase, i.e. onset of the negative spike, the velocity was similar to that using ‘D’ and ‘DT’ positions. This artifact was not seen for any fibular nerve conduction studies (Figure 7C). Only 1 ulnar nerve showed this artifact and it was seen unilaterally. There was no difference in velocity values for ulnar and fibular nerves using different E2 sites. In all nerves, regardless of the E2 position, the amplitude on proximal stimulation was similar or up to 20% smaller than the CMAP recorded with distal stimulation. The CMAPs recorded with ulnar nerve stimulation where E2 had been placed at ‘D’ and ‘P’ positions in 3 patients with ALS are compared in figure 8. In A, the waveforms show similar amplitude. However, in B and C, the amplitude using ‘P’ position is much smaller.
Discussion This study demonstrates the significant E2 electrode contribution to CMAP as described by other investigators (Kincaid et al, 1993; Brashaer and Kincaid, 1997; Nandedkar and Barkhaus, 2007). In addition, we also show the variable effect of E2 electrode on the CMAP waveform and its measurements depending upon its placement. Ideally, the CMAP should reflect only the electrical activity from the muscle recorded by the E1 electrode. This is not possible due to signals recorded from current E2 placements. One strategy to achieve this could be to place E2 on the contralateral limb (figures 3, 4 and 5). However, this gives noisy surface EMG recordings (SIP) in procedures such as motor unit number index (MUNIX) that analyzes 6
both CMAP and SIP (Nandedkar et al, 2010). On proximal stimulation, the E1 records a large amplitude signal from other co-stimulated muscles that is not cancelled by the contralateral E2 signal. Therefore, our strategy was to place the E2 close enough to the E1 electrode to minimize noise and artifact, and to record the least signal from other muscles. Historically the E2 position has always been distal to the E1 electrode. The reason for this choice is not clear. Perhaps by moving the E2 electrode away from the stimulation site the stimulus artifact was reduced, particularly when motor conductions were originally described and recording equipment was more prone to artifact (Hodes et al, 1948). In our recordings using modern amplifiers, the proximal E2 position gave high quality recordings with good separation between the CMAP and stimulus artifact (figures 3, 4, 5). The high E2 signal amplitude at distal sites may be due to its proximity to the tendons of multiple muscles innervated by the tested nerve. As example, the deep fibular nerve innervates the TA, extensor digitorum longus (EDL), and extensor hallucis longus (EHL) muscles in the anterior compartment of the leg. They are all stimulated when attempting to record the TA CMAP. Although the EDL and EHL are relatively lateral to the E1 electrode, their tendons are close to the E2 electrode when placed at or distal to the ankle (Leis and Trapani, 2000; Barkhaus et al, 2004). Their potentials not only add to the CMAP amplitude, but their temporal dispersion also increases the CMAP duration. Conversely, the origins of TA, EDL, and EHL are more separated. The origin of TA is on the lateral condyle and upper body of tibia, while the EDL and EHL fibers arise from the proximal and distal anterior shaft of fibula, respectively. This may explain the lower amplitude signal recorded using proximal sites for E2. There is an additional anatomical consideration regarding the TA muscle. Anatomic studies (Aquilonius et al, 1984) have shown that this muscle has several motor point zones. This is corroborated in our study, where moving the E1 from proximal to distal (mid-leg) shows relative lack of changes in CMAP amplitude. Based on these considerations we suggest position P (i.e., medial patella) of E2 and proximal positioning of E1 (proximal third) for recording TA. We have been doing this already for many years as a result of our experiences in MUNIX and have adopted this specific proximal recording site for TA MUNIX studies. (Neuwirth et al, 2011). Our findings in TA are consistent with those described by Escorcio-Bezerra and co-workers (2019). We were also surprised by the shape of the E2 signal. When an electrode is over the endplate zone, the signal has an initial upward deflection. Any downward initial deflection is interpreted as volume conducted signals from E1 electrode placement that is off the end-plate region. By this argument, all of our E2 recordings should have an initial positive deflection. But our recordings showed both initial positive and negative deflections (indicated by arrows in figures 3, 4, and 5). This positivity was also observed by Phongsamart and colleagues (2002) in ulnar nerve conduction studies. We have no explanation for this variable pattern. When the E2 signal is initially negative it will cancel the initial negative potential from the E1 electrode. Hence the CMAP onset will be seen later and it will have a longer distal latency. If the E2 signal is initially positive, it will add to the negativity of E1 signal by virtue of differential amplification. Hence the CMAP amplitude will be artificially higher. This is also seen in the fibular nerve studies (figures 3 and 6).
7
The ulnar nerve recordings show large amplitude peaks for the E1 and E2 signal. The peak occurs later in the E2 signal (indicated by ● in figure 4). The peak of E2 also coincided with the CMAP peak used for amplitude measurements. To properly study the individual muscle involvement of ulnar-innervated muscles in pathology (e.g., trauma, split hand distribution in motor neuron disease, etc.), the distal recording site of E2 may not be optimal (Higashihara et al, 2010, Sonoo et al, 2011). Greater sensitivity may be achieved by using proximal E2 sites. We suggest the ‘P’ site for recordings. The ‘PT’ site also gives low amplitude signals. However, the short distance between the resulting E1 and E2 position will make it inconvenient when the patient has a small hand such as in children. The median nerve recordings had low amplitude E2 signals. Hence, the position of E2 electrode does not have a large effect on the CMAP latency, amplitude and duration. Therefore, the waveforms of median conduction study are not presented. In the biceps recording the E2 signal waveform morphology varied at distal E2 positions. Under standard recording conditions, one cannot recognize the individual contributions of the E1 and E2 electrodes to the CMAP. When the signal does not have expected morphology, one tries different E1 positions. If the E2 contribution is significant, repositioning the E1 electrode may not help. This may explain some of the difficulties with CMAP recordings in the musculocutaneous nerve studies. Figure 5 showed a complex short duration waveform (indicated by *) that is much different from the expected waveform shown as dashed line in Figure 5B. When the multi-channel recording was made, the large amplitude E2 signal with distal positions (D and DT) is seen quite easily. At DT position the signal is so large that the CMAP looked inverted (indicated by #). Another technical point goes back to anatomical studies (Aquilonius et al, 1984). They demonstrated a single endplate zone for the biceps brachii. In this study, if E1 is wellplaced at mid-belly and the CMAP amplitude is good, only minimal movement of E1 is required to optimize it. In consideration of these issues, we suggest that a proximal position (“PT”) for E2 be used. The ‘P’ position gives the lowest amplitude E2 signal for the median nerve study. Hence it is optimal for the study of CMAP amplitude using distal stimulation. However, it is suboptimal when proximal stimulation is performed. It may give incorrect velocity values. With fibular and ulnar nerve, the ‘P’ position gave artifact free recordings with distal and proximal stimulation in practically all subjects. The shape of the CMAP on proximal stimulation was very similar to that from distal stimulation. The amplitude on proximal stimulation was similar or slightly smaller than distal stimulation using ‘P’ site. This indicates that ‘P’ position of E2 can be used for routine ulnar and fibular conduction studies also. We have been using this position for fibular conduction study for many years, and have not experienced any issues with proximal stimulation. Our E2 position is close to the proximal tendon or the origin of the muscle. Hence we propose the term “Proximal E2” for this montage to distinguish it from the traditional “Belly-Tendon” montage. We use the term “Proximal” as opposite of “Distal” placement of E2 electrode, and not necessarily the position at or off the proximal tendon. Using the suggested positions of E2 will require new definitions of normal limits generated from a prospective study. Nevertheless, we feel that the use of the proposed montage gives a cleaner signal that is minimally influenced by E2 and is more representative of the muscle being recorded by E1. We have made very limited comparisons of CMAPs using different E2 positions in patient groups. Figure 8 shows a comparison of CMAPs recorded using the ‘D’ (traditional position) and 8
‘P’ (the new proposed position) of E2. The amplitudes are similar for the two positions in figure 8A. Recordings in B and C were made from patients with very weak muscles (strength 3/5 using Medical Research Council scale). We were surprised to see a CMAP with amplitude of 1.2 – 1.4 mV using position ‘D’ of E2 electrode. However, the amplitude is much less using ‘P’ position (< 0.5 mV), and it better reflects the muscle weakness. We emphasize that these are very preliminary observations, and future studies will be needed to fully assess the benefits of the ‘P’ position. Proximal E2 position is away from the ‘moving’ part of the limb during stimulation. This may also result in less artifacts in repetitive nerve stimulation.
Declaration of interest The first author is an employee of Natus Neurology. This work represents no conflict. The second author has no conflict to report. Acknowledgments The first author thanks Natus Neuro for support in this project.
9
References Aquilonius SM, Askmark H, Gillberg PG, Nandedkar S, Olsson Y, Stalberg E: Topographical localization of motor endplates in cryosections of whole human muscles. Muscle Nerve, 1984; 7:287-93. Barkhaus PB, Nandedkar SD and Roberts M: “Electronic Myoanatomic Atlas for Clinical Electromyography (Second Edition)”. EMG on DVD series: Volume I. Nandedkar Productions LLC, Hopewell Junction, NY, 2004. Barkhaus P, Kincaid J, Nandedkar S: Amplitude drop on proximal stimulation in the tibial motor nerve conduction study. Muscle Nerve, 2011; 44:776-782. Brashear A, Kincaid JC: The influence of the reference electrode on CMAP configuration: leg observations and an alternate E2 site. Muscle Nerve, 1996; 19:63-67. Escorcio-Bezerra ML, Abrahao A, Nunes KF, Sparapni FV, Braga NI, Robinsom LR, Zinman LManzano G: Optimal E2 (Reference) electrode placement in fibular motor nerve conduction studies recording from the tibialis anterior muscle. Muscle Nerve, 2019; 59:249-253. Higashihara M, Sonoo M, Imafuku I, Ugawa Y, Tsuji S: Origin of ulnar compound muscle action potential investigated in patients with ulnar neuropathy at the wrist. Muscle Nerve, 2010; 41: 704-706. Hodes R, Larrabee MG, German W: The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons. Arch Neuropsychiat, 1948; 60:340-365. Kawamore Y, Sonoo M, Higashihara M, Chiba T, Hatanaka Y: Origin of surface motor unit potentials in hypothenar motor unit number estimation. Muscle Nerve, 2013; 48:185-190. Kincaid JC, Brashear A, Markand ON: The influence of the reference electrode on CMAP configuration. Muscle Nerve, 1993; 16:392-396. Leis AA and Trapani VC: “Atlas of Electromyography”. Oxford University Press, New York, 2000. Nandedkar SD, Barkhaus PE: Contribution of reference electrode to the compound muscle action potential. Muscle Nerve, 2007; 36:87-92. Nandedkar SD, Barkhaus PE, Stalberg EV: Motor unit number index (MUNIX): Principle, method and findings in healthy subjects and in patients with motor neuron disease. Muscle Nerve, 2010; 42: 796807. Nandedkar SD, Barkhaus PE, Stalberg EV: Cumulative motor index (CMI): an index to study progression of amyotrophic lateral sclerosis. J Clin Neurophysiol, 2015; 32:79-85 Neuwirth C, Nandedkar S, Stålberg E, Barkhaus P, de Carvalho M, Furtula J, Van Dijk J, Baldinger R, Castro J, Costa J, Otto M, Sandberg A, Weber M: Motor unit number index (MUNIX): Reference values of five different muscles in healthy subjects from a multi-center study. Clin Neurophysiol, 2011; 122:18951898. 10
Phongsamart G, Wertsch J, Ferdjallah M, King J, Foster D: Effect of reference electrode position on the compound muscle action potential (CMAP) onset latency. Muscle Nerve, 2002; 25:816-821. Robinson LR, Christie M, Nandedkar SD: A message from ground electrode. Muscle Nerve, 2016; 54:1010-1011. Soono M, Kurokawa K, Higashihara M, Kurono H, Hokkoku K, Hatanaka Y, Shimizu T: Origin of far field potentials in the Ulnar nerve compound muscle action potentials. Muscle Nerve, 2011; 43:671-678.
11
Table 1: Summary of E2 electrode placements. The underlined positions are suggested for CMAP recordings. (Also see Figures 1 and 2). * For median nerve ‘P’ position is optimal for distal stimulation only. The ‘D’ position is better suited when proximal stimulation is also performed.
Nerve 1: Distal ‘D’ Fibular Dorsum Foot Median Thumb* Ulnar Digit V Musculocutaneous Olecranon
2: Distal Tendon ‘DT’ Dorsum Ankle Base Thumb Base Digit V Anterior Elbow
E2 Positions 3: Proximal tendon ‘PT’ Lateral Knee Distal Wrist Distal Wrist Anterior Shoulder
4: Proximal ‘P’ Medial patella Center Dorsal Wrist* Center Dorsal Wrist Posterior upper Shoulder
12
Figure legends Figure 1: Electrode placement for musculocutaneous and fibular motor nerve conduction studies. The star symbol shows E2 position that gave the lowest amplitude and hence suggested E2 position for future recordings. Refer to text for details. Figure 2: Electrode placement for median and ulnar motor nerve conduction studies. The star symbol shows E2 position that gave the lowest amplitude and hence suggested E2 position for future recordings. For the median nerve, the ‘D’ position gives higher E2 amplitude but less artifact on proximal stimulation. Refer to text for details. Figure 3: Fibular nerve study. (A) The signals recorded by the E1, D, DT, PT and P as active electrodes are shown. The E2 reference electrode was placed on the opposite leg. Arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. (B) The CMAPs using the 4 different E2 positions are shown. Note the shorter latency and duration of CMAP using the proximal site (PT and P) for the E2 electrode. (C) The table shows mean value and standard deviation (SD) of onset latency and duration for the E2 positions. The latency and duration are shorter (p<0.01) using P and PT sites compared to D and DT sites. This is also seen from waveforms in B. Figure 4: Ulnar nerve study. (A) The signals recorded by the E1, D, DT, PT and P as active electrodes are shown. The E2 (reference) electrode was on the opposite hand. The arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. The ● symbol shows the peak of the E2 recorded signal. (B) The CMAPs recorded using the 4 different E2 positions are shown. Figure 5: Musculocutaneous nerve study. (A) Signals recorded by the E1, D, DT, PT and P as active electrodes are shown on the left. The E2 (reference electrode) is on the opposite arm. The arrows indicate initial upward and downward deflection of the signal recorded at sites away from the muscle belly. (B) The CMAP recorded using the 4 different E2 positions are shown on the right (solid lines). One CMAP shows a short duration and complex waveform (indicated by *) with two distinct negative peaks using a distal E2 position. The waveform marked # shows a mainly positive waveform with a very low amplitude initial negative peak. In A, we can see high amplitude signals recorded by E2 electrode at distal positions while the signal recorded by E1 is small. Their phase cancellation gives the complex CMAP with short duration and low amplitude for the initial negative peak. The atypical waveforms in this recording are compared with the expected CMAP waveform recorded in another subjects (top traces in dashed line). Figure 6: Peak to peak amplitude of E2 recorded signal at 4 different sites (D, DT, PT and P) is compared for 4 muscles: (A) Abductor pollicis brevis (B) Abductor digiti minimi, (C) Biceps brachii, and (D) Tibialis anterior. All plots show the same scale for amplitude. Amplitude is affected by E2 position (p<0.001, DOF=3). While the amplitude is high, it is lowest for the APB muscle. The table inset shows the mean value and standard deviation (SD) of amplitude. The amplitude is lower (p < 0.001) using P and PT sites compared to D and DT sites (except for PT site in median nerve indicated by *).
13
Figure 7: CMAP recordings with distal and proximal stimulation of (A) Median (B) Ulnar and (C) Fibular nerve are compared. The traces on top are recorded with E2 placed at ‘P’. Traces below are recorded with E2 placed at ‘D’. Marker positions are shown as identified by the machine. In the median nerve, a low amplitude positive phase (indicated by arrow) is an artifact from muscles in the forearm. This gave very high conduction velocity. Upon manually correcting the onset marker, the velocity values were similar for the two E2 positions. In ulnar and fibular nerves, no technical issues are seen.
Figure 8: CMAP recordings from the abductor digiti minimi muscle of 3 patients with amyotrophic lateral sclerosis. Upon superimposition, traces recorded with E2 placed at ‘D’ are shown as dashed line. The solid lines show CMAP with E2 placed at ‘P’. Note the relative amplitude differences.
14
15
16
17
18