- Email: [email protected]
The standard concentric needle cannula cannot replace the Macro EMG electrode
Clinical Neurophysiology 125 (2014) 406–410
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
The standard concentric needle cannula cannot replace the Macro EMG electrode Arne Sandberg ⇑ Department of Neuroscience, Clinical Neurophysiology, Uppsala University, Uppsala 75185, Sweden
a r t i c l e
i n f o
Article history: Accepted 31 July 2013 Available online 13 September 2013 Keywords: EMG CNEMG Macro EMG
h i g h l i g h t s Concentric needle EMG needle cannula cannot be used as a substitute for the costly standard Macro
EMG needle because there were different motor unit amplitudes between superficially and more deeply recorded motor units with the concentric needle cannula. There was no deep dependency when using the standard Macro electrode regarding the motor unit amplitudes. This deep dependency when using the concentric needle EMG cannula for recording is probable due to different degree of so called shunting effect which depends on the length of the exposed concentric needle cannula.
a b s t r a c t Objective: To establish the usefulness of the single use and affordable standard concentric EMG electrode as a substitute for the expensive standard macro electrode. Methods: Macro EMG performed with macro electrode is compared with recordings from the uninsulated cannula of a standard EMG electrode at two different recording depths in the tibialis anterior muscle. This was performed both in muscles with signs of collateral reinnervation and without. Results: The amplitude of the motor units recorded with the uninsulated concentric needle cannula were lower for the deeply recorded motor units compared to motor unit potential (MUP) amplitudes recorded with the standard macro electrode. The deeply recorded concentric needle (CN) cannula recorded MUPs amplitudes were also lower than superficially recorded CN cannula MUPs. The standard Macro EMG signals show no difference between deeply and superficially recorded motor units. Conclusion: The uninsulated cannula of the concentric needle electrode cannot replace the standard Macro EMG electrode due to technical reasons, probably from different effects of shunting of the bare cannula in deep vs. superficially recorded motor units. Significance: The standard CN electrode could not be used as substitute for the standard Macro EMG needle. Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The standard Macro EMG method developed by Stålberg (1980) is a well known and accepted method especially in estimating the resulting electrical size of the motor unit in reinnervation (Stålberg, 1982). The Macro EMG is a single fiber EMG triggered averaging technique which uses in this context a large recording surface electrode approximately covering the activity from the whole motor unit (MU) under study. It had been shown for instance that the method shows a higher degree of pathology in a condition that ⇑ Tel.: +46 186113440; fax: +46 18504768. E-mail address: [email protected]
commonly shows a great degree of collateral sprouting (e.g. prior polio) than concentric needle EMG (CNEMG) (Sandberg et al., 1999) and the new motor unit estimation technique of motor unit number index (MUNIX) (Sandberg et al., 2011). These differences between the techniques were in the case of the Macro/CNEMG comparison due to technical reasons, different degree of so called phase cancellation in the MUP for these two electrodes. In the Macro/MUNIX comparison the greater degree of pathology shown with the Macro was due to possible difference in motor unit populations studied with the different techniques. The Macro EMG is performed with a special Macro EMG needle electrode, which in practice is a modified standard single fiber EMG (SFEMG) needle with the SFEMG surface exposed in the center of
1388-2457/$36.00 Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinph.2013.07.021
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
the uninsulated cannula. The proximal part of the cannula is insulated with Teflon except the most distal 15 mm, from which recording of the macro signal is performed. These Macro EMG needle electrodes are commercially available. However, the standard macro electrodes are only available at a significant cost, mostly resulting in re-using the electrodes. It has not been possible so far to develop single use macro electrodes at a reasonable price due to technical reasons. It is therefore desirable to explore the possibility to replace the standard macro needle electrode with a single use inexpensive needle. There have been efforts in this direction earlier by Jabre (Jabre 1991, 1992; Bauermeister and Jabre, 1992) and Nix (Nix and Scherer, 1992). However, these earlier attempts have used a not commercially available special needle electrode with proximal Teflon insulation which result in the exposure of a fixed distal uptake area. In Stålbergs original report (Stålberg 1980) regarding Macro EMG a bare cannula without insulation was used, but soon replaced with a fixed length of the exposed cannula. Stålberg found that a fixed length of the cannula gave a lesser shunting effect (Stålberg et al., 2010) resulting in lesser difference between superficially and deeply situated motor units (MUs) in addition to an overall greater macro amplitude than with the uninsulated cannula. The results from that report showed a 30% reduction of macro amplitude regarding deeper recorded MUs compared to superficially recorded ones. That material was recorded from the biceps brachii, 20 motor unit potentials (MUPs) were recorded. I am not aware of any other study that’s confirming that this is also the case in other muscle in a reasonable quantitative study. The aim of this study is to investigate if the standard concentric needle is versatile or not regarding the reported difference in macro amplitude between superficially and deeply located MUs in another muscle than biceps brachii, namely the tibialis anterior (TA) muscle in the sense to replace the standard Macro-EMG electrode. 2. Materials and methods Thirteen subjects (mean 59 years of age, 7 female and 6 male) with the suspicion of different conditions where some degree of collateral reinnervation is expected, e.g. 4 prior polio patients, 8 patients with clinically suspected lumbal radiculopathy and one patient suffering from polyneuropathy were included in the study. The patients were investigated unilaterally in the TA muscle, except two polio patients that were investigated bilaterally. These underwent standard Macro EMG in the TA muscle in addition to use the CNEMG needle non- insulated cannula as uptake electrode for the macro signal and the concentric surface as a trigger of the MUPs. Recording were performed from the same area in the same TA muscle, first with the standard macro electrode superficially located in the muscle, than more deeply located MUPs were recorded by relocating the macro needle more deeply. Superficial location was defined as a penetration distance of 15–25 mm perpendicular with the skin surface. Deep location of the recordings were made from a penetration depth of 27–37 mm. Similar penetration depth was chosen for the concentric needle (CN) cannula recordings. This setup was chosen mainly because it was in the practical point of view feasible in the clinical recording situation since it was possible to use a simple ruler for measurement. I tested initially a more complicated setup in which I moved the cannula an additional 7.5 mm deeper to secure that the macro cannula and the CN cannula was at least partly located in the MU area under study, this setup was too complicated in the clinical situation and should make an odd modification of the macro technique in the longer perspective. This set up gives with a skin layer around 5 mm of thickness a superficial recording span of 10 mm and a deep recording span of 10 mm, thus separated by a depth of
407
2 mm. Any deeper penetration was not possible because the limited length of 37 mm regarding the penetrable part of the needle. In the macro recordings the TECA 26 G, 37 mm (part No. 17915) Macro EMG needle was used. In the CN cannula recordings the Alpine Biomed DCN 37 (26 G, 37 mm) was used. The needles were of course also moved in different perpendicular positions and at least 2 different insertions to record different motor units. With this recording setup, at least minimum of 10 mm length of the concentric needle CN cannula and macro surface is located in the muscle. The MUs recorded were selected as recommended for the standard Macro EMG. The triggering from the ‘‘active’’ CN surface located at the tip of the needle was in accordance with that reported in earlier studies regarding jitter analysis with CN (Ertas et al., 2000). The reference electrode for both the CN cannula recordings and the standard Macro EMG was located over the quadriceps tendon just proximal to the patella. The MUPs were sampled and analyzed in the Macro EMG program in the Keypoint EMG equipment. Identical or distorted MUPs were removed. At least 10 different MUPs were collected for each for the different recording locations and needles. This setup did not allow sampling of identical MUs for the two techniques. The filter settings for the standard Macro EMG single fiberchannel were 500 Hz–10 kHz, for the Standard Macro EMG channel 5 Hz–5 kHz. The filter settings for the CN triggering channel were 1000 Hz–10 kHz. The setting for the high-pass filter was chosen in accordance with the common setting in SFEMG recordings using the small concentric needle. The higher setting (1 kHz) for the CNEMG surface compared to the lower setting (500 Hz) for the SFEMG surface in the standard macro needle was chosen to get as selective recordings as possible from the CNEMG surface to avoid triggering from distant muscle fibers. The filter settings for the CN cannula recording channel were 5 Hz–5 kHz. The peak–peak amplitude were measured from the Macro and cannula channel recordings. These signals were averaged until an acceptable baseline and a constant MUP was obtained. The FD parameter in the standard Macro EMG investigation was not used in this study. The TA muscle was in addition to the above investigation also investigated with ordinary concentric EMG. The patients were separated into two groups depending of the CNEMG result in the investigated TA muscle, one group with normal CNEMG and the second group with signs of collateral reinnervation measured with CNEMG. Reference values from the laboratory were used. This setup was performed to investigate if the electrical motor unit size affects the EMG results from different recording depth in the muscle. The study was approved by the local ethics committee. All subjects gave their informed consent. 2.1. Statistical analysis ANOVA for repeated measures (rmANOVA) including least significant difference was used to compare the multiple recordings. When the so called test of sphericity indicated P < 0.05 (for the pooled material in the results) the Friedman repeated measurement ANOVA on Ranks were used. However, to get a decent indicator of differences between the different macro recording setups, so called pairwise comparisons in the rmANOVA (in addition to tests of between-subjects effect) were performed for all the different setups. Outliers were indentified separately for the groups with normal CNEMG and pathological CNEMG respectively, and were excluded from the study. All statistics were performed with the commercially available software SPSS. Differences were considered significant at P < 0.05.
408
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
Table 1 Material, CNEMG results and mean amplitude from the recordings; ‘‘SM mean amp’’ is superficially recorded mean MUPs with the Macro EMG needle, ‘‘DM mean amp’’ is the mean MUP amplitudes recorded from more deeply located MUs. ‘‘SCN mean amp’’ and ‘‘DCN mean amp’’ are superficially respectively more deeply recorded mean MUP amplitudes recorded with the CN cannula. Outliers excluded. Observe the trend for lower mean amplitude for deeply recorded MUPs with the CN cannula compared to the other recordings. Gender, age (years), disease
CNEMG result TA muscle
Note
SM mean amp
DM mean amp
SCN mean amp
DCN mean amp
Female, age 52, S1 radiculopathy Female, age 46, S1 and L4 radiculopathy Female, age 50, L4 radiculopathy Male, age 53, L4 radiculopathy Male, age 67, old polio Male, 45, radiculopathy Female, age 37, L5 radiculopathy Female, 67, L5 radiculopathy Female, 50, L5-S1 radiculopathy Female, 89, L5-S1 Polyneuropathy Male, 68, old polio Male, 74, old polio Male, 74, old polio
Normal Normal Normal Normal Normal Normal Neurogenic Neurogenic Neurogenic Neurogenic Bilateral neurogenic Bilateral neurogenic Neurogenic
Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Bilateral: L/R Bilateral: L/R Unilateral
258 215 218 244 264 234 372 350 197 1577 1097/704 371/1308 1016
304 175 202 303 345 238 409 323 242 1470 1412/616 392/1452 758
347 230 207 222 254 212 348 189 307 1717 1526/313 738/817 776
131 53 101 99 180 190 251 109 171 930 1132/125 473/726 219
Table 2 Results for all motor units summated from all subjects split into two groups. The first group consisting of MUP amplitude from the muscles with normal CNEMG and the second group consisting of the sum of MUP amplitudes from the muscles that showed signs of reinnervation from the CNEMG investigation. CV means coefficient of variation. Observe the greater CV in collateral reinnervation. Normal CNEMG
S macro D macro S CN D CN
Neurogenic CNEMG
N (MUPs)
Mean amplitude uV (sd)
CV
N (MUPs)
Mean amplitude uV (sd)
CV
86 92 87 95
242 258 240 128
0.34 0.45 0.43 0.56
121 124 121 123
783 812 628 452
0.80 0.74 0.94 0.98
(82) (116) (102) (72)
(624) (604) (589) (441)
Fig. 1. MUPs recordings from one muscle with normal CNEMG finding. Observe the low MUP amplitudes for the deep CN cannula recorded MUPs compared to the other recording situations.
409
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
3. Results The eight patients suffering from radiculopathy were investigated in eight muscles. Of these eight muscles, three muscles showed signs of collateral reinnervation from the CNEMG investigation, while the other five muscles showed normal CNEMG. The four patients that suffered from old polio were investigated in six muscles, all except one muscle showed signs of reinnervation. The patient who suffered from polyneuropathy showed pathological CNEMG results from the investigated TA muscle. All in all of these 15 investigated muscles, nine muscles showed different degree of signs of reinnervation in the CNEMG, six muscles showed normal results (Table 1). The mean, SD and coefficient of variation (CV) value for all recorded MUP amplitudes are showed in Table 2. Fig. 1 shows MUPs recorded with all setups from one muscle with normal CNEMG. 3.1. Normal CNEMG group A repeated measures ANOVA (Greenhouse–Geisser correction) showed significant mean MUP amplitude difference between the different recording set-ups (F(2.28, 11.4) = 14.8, P = 0.001). The muscles with normal CNEMG showed significant smaller deeply recorded mean CN MUP amplitude compared to recordings with the macro needle, both for superficial Macro (47%, P = 0.001) and deeply recorded macro MUPs (50%, P = 0.002). The deeply recorded
CN cannula recorded mean MUP amplitude were also significantly lower (46%, P < 0.01) than the superficially recorded CN cannula mean MUP amplitude (Fig. 2A). For the Macro-EMG needle, there was no difference in mean amplitude between superficially and deep recorded MUPs. The CV showed moderate numbers in all setups in this group. 3.2. The group with signs of collateral reinnervation from the CNEMG investigation This group showed in principal similar results as the previous group. The repeated measures ANOVA with Greenhouse–Geisser correction showed significant mean MUP amplitude difference between the different recording set-ups (F(2.02, 16.2) = 5.74,P = 0.013). The deeply recorded CN cannula mean MUP amplitude was lower compared to recordings with the macro needle, both for superficial Macro (42%, P < 0.05) and deeply recorded macro MUPs (44%, P < 0.01). The deeply recorded MUPs showed 28% lower mean MUP amplitude than the superficially recorded MUPs with the CN cannula (P < 0.01), (Fig. 2B). For the Macro-EMG needle, there was no difference in mean amplitude between superficially and deep recorded MUPs. The CV was slightly increased (especially in the CN cannula recordings) relative to the group without reinnervation (Table 2). 3.3. All muscles and MUPs pooled The repeated measures ANOVA with Greenhouse–Geisser correction showed significant mean MUP amplitude difference also for this group between the different recording set-ups (F(2.09, 29.2) = 8.38, P = 0.001). When pooling the mean MUPs for the different muscles in the non-reinnervation group with the ones from the reinnervation group, the deeply recorded mean CN cannula MUP amplitude was significantly lower (P < 0.01) compared to Table 3 Results for all motor units pooled from all subjects. The deeply recorded CN cannula recorded MUPs showed significant different mean MUP amplitude compared to the other recording situations. All MUPs pooled
Superficial macro Deep macro Superficial CN cannula Deep CN cannula
Fig. 2. The mean and 95% confidence interval for the MUP amplitudes recorded with macro and CN cannula. The material is split into those muscles that showed normal CNEMG (A) and those muscles that showed signs of reinnervation regarding the CNEMG investigation (B). Observe that the deeply recorded CN cannula recorded MUPs showed significantly lower mean amplitude compared to macro.
N
Mean MUP ampl (sd)
207 216 208 218
558 576 466 312
(549) (538) (492) (371)
Fig. 3. The mean and 95% confidence interval for the MUP amplitudes recorded with macro and CN cannula. Results from all muscles (those muscles with normal and those with reinnervation) are pooled.
410
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
the tree other recording set ups (deeply and superficially recorded macro, superficially recorded CN cannula MUPs) see Table 3 and Fig. 3. The deeply recorded mean CN cannula MUP amplitude was 33% lower (P = 0.001) than the superficially recorded mean CN cannula MUP amplitude. The superficial and deeply recorded macro MUPs showed no difference regarding mean amplitude. 4. Discussion In this study with a normal or a condition with reinnervation the CN cannula amplitude in the TA muscle shows a significantly lower amplitude for the deeply recorded MUPs compared to standard Macro EMG and to superficially recorded MUPs in this setup using the CNEMG cannula as macro electrode. This was not the case for deep and superficially recorded MUs recorded with the standard Macro EMG method, which showed similar macro amplitude results for deep and superficially recorded MUs. These results from the investigated TA muscle are comparable with the results obtained by Stålberg regarding the biceps brachii muscle (Stålberg, 1980). The difference must be either be due to differences in the techniques or differences in the recorded MU populations. What is the reason for this discrepancy? One factor may be the so called shunting effect. Large electrode surfaces (as the macro surface and CN cannula) tends to attenuate the electrical activity from the closest muscle fibers while small electrode surfaces have lesser shunting effect, resulting in greater contribution to the amplitude from the closest muscle fibers vs. the distant fibers (Stålberg et al., 2010; Falck et al., 1995). The difference in MUP amplitude in this study, the order of 30–50% drop in MUP amplitude for the CNEMG cannula deeply recorded MUPs may correspond to the approximately twice as long exposed recording cannula length for the CNEMG cannula than for the Standard Macro. Another factor may be the spatial relationship of the recording electrode to the MU territory under study. The triggering is performed from the tip of the CN needle and that is a drawback compared with the original Macro EMG according to the fact that the triggering of the CN cannula signal is at one extreme made from the most superficially located muscle fibers resulting in the situation that the recording cannula is in that case almost outside the MU under study. The other extreme is triggering on the most inward located muscle fibers resulting in covering of the whole MU with the cannula which is the most desirable physiologic situation. However, the possible dominating situation is probable in-between these extremes resulting in a partial coverage by the cannula of the MU under study. In this study there was a tendency of a greater spread of the CNEMG Macro amplitude for the reinnervated MUPs compared to the non-reinnervated MUPs with a dominance in the CN cannula recorded group. This may indicate that the ‘‘physiologically correct’’ trigger in macro located presumably in the middle of the MU under study may also have a practical advantage over the trigger located at the end of the recording surface. This variation of the cannula length covering the MU territory may however not explain the whole difference of CNEMG amplitude recorded from deep vs. superficially recorded MUs since the variation of electrode length would be in the same magnitude for both superficial and the deeply performed recordings. The MU location in the muscle may have implications on the MU size. There is one report showing that more superficial MUs are larger than deeper located ones in the vastus lateralis muscle (Knight and Kamen, 2005). If this is the case in the TA muscle is not known, but the similar results from the deep and superficially recordings with standard Macro EMG indicates that this is not the case.
A reason for the lower macro MUP amplitude in deeply located CN cannula recordings compared to standard macro could be due to different MU populations recorded. According to the size principle (Olson et al., 1968), MU size has a positive correlation with increasing degree of activation, even in neurologic conditions (Milner-Brown et al., 1974). Theoretically, the SFEMG trigger which is more selective tolerates more activation before interference of the trigger from different muscle fibers from different MUs disturbing the triggering process. This fact may contribute to recording higher threshold MUs resulting in higher MUP amplitude than the concentric trigger larger uptake area which favorites a lower degree of activation to get a clean trigger compared to the SF. I have no indication of a greater degree of activation of the SF triggered standard Macro EMG compared to the CN cannula recordings, since the baseline otherwise should be more unstable in the standard Macro recordings, which is not the case. This doesn’t however explain the different results from different depths regarding the CNEMG macro recording. This study indicates that the difference in MUP amplitude at different recording depths in the TA muscle performed with the CNEMG cannula probably results from technical reasons, e.g. variable exposition of the length of the cannula which result in greater shunting of the recorded MUPs when recording more deeply. In addition, the suboptimal location of the trigger for the CN cannula recordings may also have some influence. This is in line with Stålbergs earlier results from the biceps brachii muscle indicated in the original report (Stålberg, 1980) regarding the macro method. The conclusion must be that the bare cannula recordings with variable length due to different recording depths in the muscle performed with the CNEMG cannula are not sufficient as a replacement for the standard Macro EMG electrode. Acknowledgement The author thanks Margareta Grindlund for technical assistance. Conflict of interest: None. References Bauermeister W, Jabre JF. The spectrum of concentric macro EMG correlations. Part I. Normal subjects. Muscle Nerve 1992;15:1081–4. Ertas M, Baslo MB, Yildiz N, Yazici J, Oge AE. Concentric needle electrode for neuromuscular jitter analysis. Muscle Nerve 2000;23:715–9. Falck B, Stålberg E, Bischoff C. Influence of recording site within the muscle on motor unit potentials. Muscle Nerve 1995;18:1385–9. Jabre JF. Concentric macro electromyography. Muscle Nerve 1991;14:820–5. Jabre JF. The spectrum of concentric macro EMG correlations. Part II. Patients with diseases of muscle and nerve. Muscle Nerve 1992;15:1085–8. Knight CA, Kamen G. Superficial motor units are larger than deeper motor units in human vastus lateralis muscle. Muscle Nerve 2005;31:475–80. Milner-Brown HS, Stein RB, Lee RG. Pattern of recruiting human motor units in neuropathies and motor neurone disease. J Neurol Neurosurg Psychiatry 1974;37:665–9. Nix WA, Scherer A. Single fiber macro versus concentric trigger macro EMG: a comparison of methods. Muscle Nerve 1992;15:193–8. Olson CB, Carpenter DO, Henneman E. Orderly recruitment of muscle action potentials. Motor unit threshold and EMG amplitude. Arch Neurol 1968;19: 591–7. Sandberg A, Hansson B, Stålberg E. Comparison between concentric needle EMG and macro EMG in patients with a history of polio. Clin Neurophysiol 1999;110: 1900–8. Sandberg A, Nandedkar SD, Stålberg E. Macro electromyography and motor unit number index in the tibialis anterior muscle: differences and similarities in characterizing motor unit properties in prior polio. Muscle Nerve 2011;43: 335–41. Stålberg E, Macro EMG. A new recording technique. J Neurol Neurosurg Psychiatry 1980;43:475–82. Stålberg E. Macroelectromyography in reinnervation. Muscle Nerve 1982;5(9S): S135–8. Stålberg EV, Trontelj JV, Sanders DB. Selective recording. In: Stålberg EV, Trontelj JV, Sanders DB, editors. Single fiber EMG. 3rd ed. Sweden, Fiskebäckskil: Edshagen Publishing House; 2010. p. 7–10.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
The standard concentric needle cannula cannot replace the Macro EMG electrode Arne Sandberg ⇑ Department of Neuroscience, Clinical Neurophysiology, Uppsala University, Uppsala 75185, Sweden
a r t i c l e
i n f o
Article history: Accepted 31 July 2013 Available online 13 September 2013 Keywords: EMG CNEMG Macro EMG
h i g h l i g h t s Concentric needle EMG needle cannula cannot be used as a substitute for the costly standard Macro
EMG needle because there were different motor unit amplitudes between superficially and more deeply recorded motor units with the concentric needle cannula. There was no deep dependency when using the standard Macro electrode regarding the motor unit amplitudes. This deep dependency when using the concentric needle EMG cannula for recording is probable due to different degree of so called shunting effect which depends on the length of the exposed concentric needle cannula.
a b s t r a c t Objective: To establish the usefulness of the single use and affordable standard concentric EMG electrode as a substitute for the expensive standard macro electrode. Methods: Macro EMG performed with macro electrode is compared with recordings from the uninsulated cannula of a standard EMG electrode at two different recording depths in the tibialis anterior muscle. This was performed both in muscles with signs of collateral reinnervation and without. Results: The amplitude of the motor units recorded with the uninsulated concentric needle cannula were lower for the deeply recorded motor units compared to motor unit potential (MUP) amplitudes recorded with the standard macro electrode. The deeply recorded concentric needle (CN) cannula recorded MUPs amplitudes were also lower than superficially recorded CN cannula MUPs. The standard Macro EMG signals show no difference between deeply and superficially recorded motor units. Conclusion: The uninsulated cannula of the concentric needle electrode cannot replace the standard Macro EMG electrode due to technical reasons, probably from different effects of shunting of the bare cannula in deep vs. superficially recorded motor units. Significance: The standard CN electrode could not be used as substitute for the standard Macro EMG needle. Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The standard Macro EMG method developed by Stålberg (1980) is a well known and accepted method especially in estimating the resulting electrical size of the motor unit in reinnervation (Stålberg, 1982). The Macro EMG is a single fiber EMG triggered averaging technique which uses in this context a large recording surface electrode approximately covering the activity from the whole motor unit (MU) under study. It had been shown for instance that the method shows a higher degree of pathology in a condition that ⇑ Tel.: +46 186113440; fax: +46 18504768. E-mail address: [email protected]
commonly shows a great degree of collateral sprouting (e.g. prior polio) than concentric needle EMG (CNEMG) (Sandberg et al., 1999) and the new motor unit estimation technique of motor unit number index (MUNIX) (Sandberg et al., 2011). These differences between the techniques were in the case of the Macro/CNEMG comparison due to technical reasons, different degree of so called phase cancellation in the MUP for these two electrodes. In the Macro/MUNIX comparison the greater degree of pathology shown with the Macro was due to possible difference in motor unit populations studied with the different techniques. The Macro EMG is performed with a special Macro EMG needle electrode, which in practice is a modified standard single fiber EMG (SFEMG) needle with the SFEMG surface exposed in the center of
1388-2457/$36.00 Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinph.2013.07.021
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
the uninsulated cannula. The proximal part of the cannula is insulated with Teflon except the most distal 15 mm, from which recording of the macro signal is performed. These Macro EMG needle electrodes are commercially available. However, the standard macro electrodes are only available at a significant cost, mostly resulting in re-using the electrodes. It has not been possible so far to develop single use macro electrodes at a reasonable price due to technical reasons. It is therefore desirable to explore the possibility to replace the standard macro needle electrode with a single use inexpensive needle. There have been efforts in this direction earlier by Jabre (Jabre 1991, 1992; Bauermeister and Jabre, 1992) and Nix (Nix and Scherer, 1992). However, these earlier attempts have used a not commercially available special needle electrode with proximal Teflon insulation which result in the exposure of a fixed distal uptake area. In Stålbergs original report (Stålberg 1980) regarding Macro EMG a bare cannula without insulation was used, but soon replaced with a fixed length of the exposed cannula. Stålberg found that a fixed length of the cannula gave a lesser shunting effect (Stålberg et al., 2010) resulting in lesser difference between superficially and deeply situated motor units (MUs) in addition to an overall greater macro amplitude than with the uninsulated cannula. The results from that report showed a 30% reduction of macro amplitude regarding deeper recorded MUs compared to superficially recorded ones. That material was recorded from the biceps brachii, 20 motor unit potentials (MUPs) were recorded. I am not aware of any other study that’s confirming that this is also the case in other muscle in a reasonable quantitative study. The aim of this study is to investigate if the standard concentric needle is versatile or not regarding the reported difference in macro amplitude between superficially and deeply located MUs in another muscle than biceps brachii, namely the tibialis anterior (TA) muscle in the sense to replace the standard Macro-EMG electrode. 2. Materials and methods Thirteen subjects (mean 59 years of age, 7 female and 6 male) with the suspicion of different conditions where some degree of collateral reinnervation is expected, e.g. 4 prior polio patients, 8 patients with clinically suspected lumbal radiculopathy and one patient suffering from polyneuropathy were included in the study. The patients were investigated unilaterally in the TA muscle, except two polio patients that were investigated bilaterally. These underwent standard Macro EMG in the TA muscle in addition to use the CNEMG needle non- insulated cannula as uptake electrode for the macro signal and the concentric surface as a trigger of the MUPs. Recording were performed from the same area in the same TA muscle, first with the standard macro electrode superficially located in the muscle, than more deeply located MUPs were recorded by relocating the macro needle more deeply. Superficial location was defined as a penetration distance of 15–25 mm perpendicular with the skin surface. Deep location of the recordings were made from a penetration depth of 27–37 mm. Similar penetration depth was chosen for the concentric needle (CN) cannula recordings. This setup was chosen mainly because it was in the practical point of view feasible in the clinical recording situation since it was possible to use a simple ruler for measurement. I tested initially a more complicated setup in which I moved the cannula an additional 7.5 mm deeper to secure that the macro cannula and the CN cannula was at least partly located in the MU area under study, this setup was too complicated in the clinical situation and should make an odd modification of the macro technique in the longer perspective. This set up gives with a skin layer around 5 mm of thickness a superficial recording span of 10 mm and a deep recording span of 10 mm, thus separated by a depth of
407
2 mm. Any deeper penetration was not possible because the limited length of 37 mm regarding the penetrable part of the needle. In the macro recordings the TECA 26 G, 37 mm (part No. 17915) Macro EMG needle was used. In the CN cannula recordings the Alpine Biomed DCN 37 (26 G, 37 mm) was used. The needles were of course also moved in different perpendicular positions and at least 2 different insertions to record different motor units. With this recording setup, at least minimum of 10 mm length of the concentric needle CN cannula and macro surface is located in the muscle. The MUs recorded were selected as recommended for the standard Macro EMG. The triggering from the ‘‘active’’ CN surface located at the tip of the needle was in accordance with that reported in earlier studies regarding jitter analysis with CN (Ertas et al., 2000). The reference electrode for both the CN cannula recordings and the standard Macro EMG was located over the quadriceps tendon just proximal to the patella. The MUPs were sampled and analyzed in the Macro EMG program in the Keypoint EMG equipment. Identical or distorted MUPs were removed. At least 10 different MUPs were collected for each for the different recording locations and needles. This setup did not allow sampling of identical MUs for the two techniques. The filter settings for the standard Macro EMG single fiberchannel were 500 Hz–10 kHz, for the Standard Macro EMG channel 5 Hz–5 kHz. The filter settings for the CN triggering channel were 1000 Hz–10 kHz. The setting for the high-pass filter was chosen in accordance with the common setting in SFEMG recordings using the small concentric needle. The higher setting (1 kHz) for the CNEMG surface compared to the lower setting (500 Hz) for the SFEMG surface in the standard macro needle was chosen to get as selective recordings as possible from the CNEMG surface to avoid triggering from distant muscle fibers. The filter settings for the CN cannula recording channel were 5 Hz–5 kHz. The peak–peak amplitude were measured from the Macro and cannula channel recordings. These signals were averaged until an acceptable baseline and a constant MUP was obtained. The FD parameter in the standard Macro EMG investigation was not used in this study. The TA muscle was in addition to the above investigation also investigated with ordinary concentric EMG. The patients were separated into two groups depending of the CNEMG result in the investigated TA muscle, one group with normal CNEMG and the second group with signs of collateral reinnervation measured with CNEMG. Reference values from the laboratory were used. This setup was performed to investigate if the electrical motor unit size affects the EMG results from different recording depth in the muscle. The study was approved by the local ethics committee. All subjects gave their informed consent. 2.1. Statistical analysis ANOVA for repeated measures (rmANOVA) including least significant difference was used to compare the multiple recordings. When the so called test of sphericity indicated P < 0.05 (for the pooled material in the results) the Friedman repeated measurement ANOVA on Ranks were used. However, to get a decent indicator of differences between the different macro recording setups, so called pairwise comparisons in the rmANOVA (in addition to tests of between-subjects effect) were performed for all the different setups. Outliers were indentified separately for the groups with normal CNEMG and pathological CNEMG respectively, and were excluded from the study. All statistics were performed with the commercially available software SPSS. Differences were considered significant at P < 0.05.
408
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
Table 1 Material, CNEMG results and mean amplitude from the recordings; ‘‘SM mean amp’’ is superficially recorded mean MUPs with the Macro EMG needle, ‘‘DM mean amp’’ is the mean MUP amplitudes recorded from more deeply located MUs. ‘‘SCN mean amp’’ and ‘‘DCN mean amp’’ are superficially respectively more deeply recorded mean MUP amplitudes recorded with the CN cannula. Outliers excluded. Observe the trend for lower mean amplitude for deeply recorded MUPs with the CN cannula compared to the other recordings. Gender, age (years), disease
CNEMG result TA muscle
Note
SM mean amp
DM mean amp
SCN mean amp
DCN mean amp
Female, age 52, S1 radiculopathy Female, age 46, S1 and L4 radiculopathy Female, age 50, L4 radiculopathy Male, age 53, L4 radiculopathy Male, age 67, old polio Male, 45, radiculopathy Female, age 37, L5 radiculopathy Female, 67, L5 radiculopathy Female, 50, L5-S1 radiculopathy Female, 89, L5-S1 Polyneuropathy Male, 68, old polio Male, 74, old polio Male, 74, old polio
Normal Normal Normal Normal Normal Normal Neurogenic Neurogenic Neurogenic Neurogenic Bilateral neurogenic Bilateral neurogenic Neurogenic
Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Unilateral Bilateral: L/R Bilateral: L/R Unilateral
258 215 218 244 264 234 372 350 197 1577 1097/704 371/1308 1016
304 175 202 303 345 238 409 323 242 1470 1412/616 392/1452 758
347 230 207 222 254 212 348 189 307 1717 1526/313 738/817 776
131 53 101 99 180 190 251 109 171 930 1132/125 473/726 219
Table 2 Results for all motor units summated from all subjects split into two groups. The first group consisting of MUP amplitude from the muscles with normal CNEMG and the second group consisting of the sum of MUP amplitudes from the muscles that showed signs of reinnervation from the CNEMG investigation. CV means coefficient of variation. Observe the greater CV in collateral reinnervation. Normal CNEMG
S macro D macro S CN D CN
Neurogenic CNEMG
N (MUPs)
Mean amplitude uV (sd)
CV
N (MUPs)
Mean amplitude uV (sd)
CV
86 92 87 95
242 258 240 128
0.34 0.45 0.43 0.56
121 124 121 123
783 812 628 452
0.80 0.74 0.94 0.98
(82) (116) (102) (72)
(624) (604) (589) (441)
Fig. 1. MUPs recordings from one muscle with normal CNEMG finding. Observe the low MUP amplitudes for the deep CN cannula recorded MUPs compared to the other recording situations.
409
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
3. Results The eight patients suffering from radiculopathy were investigated in eight muscles. Of these eight muscles, three muscles showed signs of collateral reinnervation from the CNEMG investigation, while the other five muscles showed normal CNEMG. The four patients that suffered from old polio were investigated in six muscles, all except one muscle showed signs of reinnervation. The patient who suffered from polyneuropathy showed pathological CNEMG results from the investigated TA muscle. All in all of these 15 investigated muscles, nine muscles showed different degree of signs of reinnervation in the CNEMG, six muscles showed normal results (Table 1). The mean, SD and coefficient of variation (CV) value for all recorded MUP amplitudes are showed in Table 2. Fig. 1 shows MUPs recorded with all setups from one muscle with normal CNEMG. 3.1. Normal CNEMG group A repeated measures ANOVA (Greenhouse–Geisser correction) showed significant mean MUP amplitude difference between the different recording set-ups (F(2.28, 11.4) = 14.8, P = 0.001). The muscles with normal CNEMG showed significant smaller deeply recorded mean CN MUP amplitude compared to recordings with the macro needle, both for superficial Macro (47%, P = 0.001) and deeply recorded macro MUPs (50%, P = 0.002). The deeply recorded
CN cannula recorded mean MUP amplitude were also significantly lower (46%, P < 0.01) than the superficially recorded CN cannula mean MUP amplitude (Fig. 2A). For the Macro-EMG needle, there was no difference in mean amplitude between superficially and deep recorded MUPs. The CV showed moderate numbers in all setups in this group. 3.2. The group with signs of collateral reinnervation from the CNEMG investigation This group showed in principal similar results as the previous group. The repeated measures ANOVA with Greenhouse–Geisser correction showed significant mean MUP amplitude difference between the different recording set-ups (F(2.02, 16.2) = 5.74,P = 0.013). The deeply recorded CN cannula mean MUP amplitude was lower compared to recordings with the macro needle, both for superficial Macro (42%, P < 0.05) and deeply recorded macro MUPs (44%, P < 0.01). The deeply recorded MUPs showed 28% lower mean MUP amplitude than the superficially recorded MUPs with the CN cannula (P < 0.01), (Fig. 2B). For the Macro-EMG needle, there was no difference in mean amplitude between superficially and deep recorded MUPs. The CV was slightly increased (especially in the CN cannula recordings) relative to the group without reinnervation (Table 2). 3.3. All muscles and MUPs pooled The repeated measures ANOVA with Greenhouse–Geisser correction showed significant mean MUP amplitude difference also for this group between the different recording set-ups (F(2.09, 29.2) = 8.38, P = 0.001). When pooling the mean MUPs for the different muscles in the non-reinnervation group with the ones from the reinnervation group, the deeply recorded mean CN cannula MUP amplitude was significantly lower (P < 0.01) compared to Table 3 Results for all motor units pooled from all subjects. The deeply recorded CN cannula recorded MUPs showed significant different mean MUP amplitude compared to the other recording situations. All MUPs pooled
Superficial macro Deep macro Superficial CN cannula Deep CN cannula
Fig. 2. The mean and 95% confidence interval for the MUP amplitudes recorded with macro and CN cannula. The material is split into those muscles that showed normal CNEMG (A) and those muscles that showed signs of reinnervation regarding the CNEMG investigation (B). Observe that the deeply recorded CN cannula recorded MUPs showed significantly lower mean amplitude compared to macro.
N
Mean MUP ampl (sd)
207 216 208 218
558 576 466 312
(549) (538) (492) (371)
Fig. 3. The mean and 95% confidence interval for the MUP amplitudes recorded with macro and CN cannula. Results from all muscles (those muscles with normal and those with reinnervation) are pooled.
410
A. Sandberg / Clinical Neurophysiology 125 (2014) 406–410
the tree other recording set ups (deeply and superficially recorded macro, superficially recorded CN cannula MUPs) see Table 3 and Fig. 3. The deeply recorded mean CN cannula MUP amplitude was 33% lower (P = 0.001) than the superficially recorded mean CN cannula MUP amplitude. The superficial and deeply recorded macro MUPs showed no difference regarding mean amplitude. 4. Discussion In this study with a normal or a condition with reinnervation the CN cannula amplitude in the TA muscle shows a significantly lower amplitude for the deeply recorded MUPs compared to standard Macro EMG and to superficially recorded MUPs in this setup using the CNEMG cannula as macro electrode. This was not the case for deep and superficially recorded MUs recorded with the standard Macro EMG method, which showed similar macro amplitude results for deep and superficially recorded MUs. These results from the investigated TA muscle are comparable with the results obtained by Stålberg regarding the biceps brachii muscle (Stålberg, 1980). The difference must be either be due to differences in the techniques or differences in the recorded MU populations. What is the reason for this discrepancy? One factor may be the so called shunting effect. Large electrode surfaces (as the macro surface and CN cannula) tends to attenuate the electrical activity from the closest muscle fibers while small electrode surfaces have lesser shunting effect, resulting in greater contribution to the amplitude from the closest muscle fibers vs. the distant fibers (Stålberg et al., 2010; Falck et al., 1995). The difference in MUP amplitude in this study, the order of 30–50% drop in MUP amplitude for the CNEMG cannula deeply recorded MUPs may correspond to the approximately twice as long exposed recording cannula length for the CNEMG cannula than for the Standard Macro. Another factor may be the spatial relationship of the recording electrode to the MU territory under study. The triggering is performed from the tip of the CN needle and that is a drawback compared with the original Macro EMG according to the fact that the triggering of the CN cannula signal is at one extreme made from the most superficially located muscle fibers resulting in the situation that the recording cannula is in that case almost outside the MU under study. The other extreme is triggering on the most inward located muscle fibers resulting in covering of the whole MU with the cannula which is the most desirable physiologic situation. However, the possible dominating situation is probable in-between these extremes resulting in a partial coverage by the cannula of the MU under study. In this study there was a tendency of a greater spread of the CNEMG Macro amplitude for the reinnervated MUPs compared to the non-reinnervated MUPs with a dominance in the CN cannula recorded group. This may indicate that the ‘‘physiologically correct’’ trigger in macro located presumably in the middle of the MU under study may also have a practical advantage over the trigger located at the end of the recording surface. This variation of the cannula length covering the MU territory may however not explain the whole difference of CNEMG amplitude recorded from deep vs. superficially recorded MUs since the variation of electrode length would be in the same magnitude for both superficial and the deeply performed recordings. The MU location in the muscle may have implications on the MU size. There is one report showing that more superficial MUs are larger than deeper located ones in the vastus lateralis muscle (Knight and Kamen, 2005). If this is the case in the TA muscle is not known, but the similar results from the deep and superficially recordings with standard Macro EMG indicates that this is not the case.
A reason for the lower macro MUP amplitude in deeply located CN cannula recordings compared to standard macro could be due to different MU populations recorded. According to the size principle (Olson et al., 1968), MU size has a positive correlation with increasing degree of activation, even in neurologic conditions (Milner-Brown et al., 1974). Theoretically, the SFEMG trigger which is more selective tolerates more activation before interference of the trigger from different muscle fibers from different MUs disturbing the triggering process. This fact may contribute to recording higher threshold MUs resulting in higher MUP amplitude than the concentric trigger larger uptake area which favorites a lower degree of activation to get a clean trigger compared to the SF. I have no indication of a greater degree of activation of the SF triggered standard Macro EMG compared to the CN cannula recordings, since the baseline otherwise should be more unstable in the standard Macro recordings, which is not the case. This doesn’t however explain the different results from different depths regarding the CNEMG macro recording. This study indicates that the difference in MUP amplitude at different recording depths in the TA muscle performed with the CNEMG cannula probably results from technical reasons, e.g. variable exposition of the length of the cannula which result in greater shunting of the recorded MUPs when recording more deeply. In addition, the suboptimal location of the trigger for the CN cannula recordings may also have some influence. This is in line with Stålbergs earlier results from the biceps brachii muscle indicated in the original report (Stålberg, 1980) regarding the macro method. The conclusion must be that the bare cannula recordings with variable length due to different recording depths in the muscle performed with the CNEMG cannula are not sufficient as a replacement for the standard Macro EMG electrode. Acknowledgement The author thanks Margareta Grindlund for technical assistance. Conflict of interest: None. References Bauermeister W, Jabre JF. The spectrum of concentric macro EMG correlations. Part I. Normal subjects. Muscle Nerve 1992;15:1081–4. Ertas M, Baslo MB, Yildiz N, Yazici J, Oge AE. Concentric needle electrode for neuromuscular jitter analysis. Muscle Nerve 2000;23:715–9. Falck B, Stålberg E, Bischoff C. Influence of recording site within the muscle on motor unit potentials. Muscle Nerve 1995;18:1385–9. Jabre JF. Concentric macro electromyography. Muscle Nerve 1991;14:820–5. Jabre JF. The spectrum of concentric macro EMG correlations. Part II. Patients with diseases of muscle and nerve. Muscle Nerve 1992;15:1085–8. Knight CA, Kamen G. Superficial motor units are larger than deeper motor units in human vastus lateralis muscle. Muscle Nerve 2005;31:475–80. Milner-Brown HS, Stein RB, Lee RG. Pattern of recruiting human motor units in neuropathies and motor neurone disease. J Neurol Neurosurg Psychiatry 1974;37:665–9. Nix WA, Scherer A. Single fiber macro versus concentric trigger macro EMG: a comparison of methods. Muscle Nerve 1992;15:193–8. Olson CB, Carpenter DO, Henneman E. Orderly recruitment of muscle action potentials. Motor unit threshold and EMG amplitude. Arch Neurol 1968;19: 591–7. Sandberg A, Hansson B, Stålberg E. Comparison between concentric needle EMG and macro EMG in patients with a history of polio. Clin Neurophysiol 1999;110: 1900–8. Sandberg A, Nandedkar SD, Stålberg E. Macro electromyography and motor unit number index in the tibialis anterior muscle: differences and similarities in characterizing motor unit properties in prior polio. Muscle Nerve 2011;43: 335–41. Stålberg E, Macro EMG. A new recording technique. J Neurol Neurosurg Psychiatry 1980;43:475–82. Stålberg E. Macroelectromyography in reinnervation. Muscle Nerve 1982;5(9S): S135–8. Stålberg EV, Trontelj JV, Sanders DB. Selective recording. In: Stålberg EV, Trontelj JV, Sanders DB, editors. Single fiber EMG. 3rd ed. Sweden, Fiskebäckskil: Edshagen Publishing House; 2010. p. 7–10.