- Email: [email protected]
Influence of stimulus intensity on the soleus H-reflex amplitude and modulation during locomotion
Journal of Electromyography and Kinesiology 23 (2013) 438–442
Contents lists available at SciVerse ScienceDirect
Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin
Influence of stimulus intensity on the soleus H-reflex amplitude and modulation during locomotion Erik B. Simonsen ⇑, Tine Alkjær, Peter C. Raffalt Department of Neuroscience and Pharmacology, University of Copenhagen, Denmark
a r t i c l e
i n f o
Article history: Received 4 January 2012 Received in revised form 17 September 2012 Accepted 30 October 2012
Keywords: H-reflex Walking Running EMG
a b s t r a c t Diverging results have been reported regarding the modulation and amplitude of the soleus H-reflex measured during human walking and running. A possible explanation to this could be the use of too high stimulus strength in some studies while not in others. During activities like walking and running it is necessary to use a small M-wave to control the effective stimulus strength during all phases of the movement. This implies that the descending part of the H-reflex recruitment curve is being used, which may lead to an unwanted suppression of the H-reflex due to limitations imbedded within the H-reflex methodology itself. Accordingly, the purpose of the present study was to study the effect on the soleus H-reflex during walking and running using stimulus intensities normally considered too high (up to 45% Mmax). Using M-waves of 25–45% Mmax as opposed to 5–25% Mmax showed a significant suppression of the peak H-reflex during the stance phase of walking, while no changes were observed during running. No differences were observed regarding modulation pattern. So a possible use of too high stimulus intensity cannot explain the differences mentioned. The surprising result in running may be explained by the much higher voluntary muscle activity, which implies the existence of a V-wave influencing the H-reflex amplitude in positive direction. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction It is commonly known that the H-reflex is most sensitive to spinal modulation on the ascending part of the H-reflex recruitment curve (Schieppati, 1987) and recently it was stated by Grosprêtre and Martin (2012) that during motor tasks, changes in spinal excitability should be assessed in the ascending part of the curve, where H-reflex modulations do not depend on background electrical activity of the muscle tested. During dynamic locomotion like walking and running it is a challenge to provide the same stimulus intensity in all phases of the movement cycle. This cannot easily be obtained, as the same stimulus intensity produces a varying effective stimulus strength in different phases of the movement, which means that a varying number of axons are activated. When measuring the soleus H-reflex, this is due to the stimulating cathode moving with respect to the tibial nerve in the popliteal fossa when the knee joint is being flexed and extended (Dyhre-Poulsen and Simonsen, 2002). The solution is to produce a small M-wave and keep the peak to ⇑ Corresponding author. Address: Department of Neuroscience and Pharmacology, University of Copenhagen, Blegdamsvej 3, 2200 Copenhagen N, Denmark. Tel.: +45 28 75 72 30. E-mail address: [email protected] (E.B. Simonsen). 1050-6411/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jelekin.2012.10.019
peak amplitude as constant as possible during all phases of the movement. The rationale is that if a constant number of efferents are stimulated it is assumed that also a constant number of afferents are stimulated, which then makes it possible to study spinal modulation of the H-reflex during movement. It is, however, a drawback that stimulating for a small M-wave means that the descending part of the recruitment curve is being used. Another problem concerning measurement of the soleus H-reflex during gait is that the amplitude of the evoked potentials varies as the muscle fibers move with respect to the recording EMG electrodes, i.e. the amplitude of the M-wave and the H-reflex varies as a function of the ankle joint angle (Dyhre-Poulsen and Simonsen, 2002). This phenomenon was also shown during static conditions (Gerilovsky et al., 1989) and by Frigon et al. (2007) who found a signicantly higher Mmax at the shortest muscle lengths in the soleus. The solution to this problem is to measure the maximal M-wave in all phases of the movement by a supramaximal stimulus 60 ms after eliciting the test stimulus. Then each M-wave may be expressed relative to the variations in Mmax during the gait cycle (Dyhre-Poulsen and Simonsen, 2002). Due to this phenomenon, the ascending part of the recruitment curve cannot be used during locomotion as a low stimulus producing no M-wave could result in an H-reflex ranging between zero and maximum as the effective stimulus intensity would be unknown.
439
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
Accordingly, it is imperative to ensure that a small M-wave on the beginning of the descending part of the recruitment curve is being used and corrected for fluctuations in Mmax. Especially if too large M-waves and thereby excessive stimulus strength is being used the measured modulation of the H-reflex may be impaired severely. In the gait studies of Simonsen and Dyhre-Poulsen (1999), Simonsen et al. (2002) and Ferris et al. (2001) M-waves of 25 ± 10% Mmax were used and corrected for fluctuations in Mmax. However in all other studies of the soleus H-reflex modulation during human gait no such corrections were applied (e.g. Capaday and Stein, 1986; Capaday and Stein, 1987; Crenna and Frigo, 1987; Edamura et al., 1991; Ethier et al., 2003; Hodapp et al., 2009). Discrepancies in the amplitude of the soleus H-reflex were seen in these studies. Especially during running, where Capaday and Stein (1987) reported a lower soleus H-reflex during running compared with walking, whereas Simonsen and Dyhre-Poulsen (1999) found either a similar excitability during walking and running or a higher excitability during faster running speed. Also during the swing phase differences have been reported as Simonsen et al. (2002) found a gradually increasing H-reflex during the swing phase of walking in about 40% of their subjects and an almost completely suppressed H-reflex in the rest. In all other studies the H-reflex was suppressed during the swing phase. Theoretically, if too high stimulus intensity was used in the latter studies it may have depressed the excitability of the H-reflex during the swing phase in all subjects and limited the possible spinal modulation in general. The purpose of the present study was therefore to quantify the expected suppression caused by the use of relatively high stimulus strengths compared to those normally used in studies of H-reflex modulation during human gait. It was hypothesized that the high stimulus strength would produce a significantly lower H-reflex in most phases of the gait cycle. 2. Methods 2.1. Subjects Seven male subjects accustomed to treadmill locomotion volunteered to participate in the experiments. They gave their informed consent to the conditions of the experiments, which were approved by the local ethics committee. Mean age was 28 years (24–35), mean height 1.85 m (1.72–1.95) and mean body weight 76 kg (65–89). 2.2. Procedure When the subjects arrived EMG electrodes were mounted on the soleus muscle and the site of nerve stimulation was located. The maximal M-wave was then measured at rest in the standing position by giving supra-maximal stimuli until the M-wave ceased to increase. The subjects performed walking and running on a motor-driven treadmill (HS-1200; Technogym) at 4.5 (walking) and 8 (running) km h 1 while the soleus H-reflex excitability was measured during the gait cycle. 2.3. Electromyography The electrodes (Medicotest Q-10-A) were positioned dorsally on the right leg two cm below the visual muscle belly of the gastrocnemius muscle. A reference electrode was placed over the tibia. The skin was shaved, lightly abraded with emery-cloth and cleaned with alcohol. The electrodes were connected to small custom-built preamplifiers (input impedance 80 MX) taped to the skin. The EMG signals were then led through 5 m long shielded wires to custom-built amplifiers with a frequency response of 20 Hz to 10 kHz.
2.4. H-reflex The site of stimulation of the tibial nerve was carefully located in the popliteal fossa using a hand held probe electrode. Conditions for the optimum site were that (1) the M-wave and the H reflex should preferably have the same visual shape and (2) it should be possible to evoke the H-reflex without any visible M-wave. After location of the optimum site, the permanent stimulus electrode (Ambu VL-00-A) was placed over the tibial nerve. A 40 mm diameter anode was placed over the patella. The stimulus was a 1 ms square wave pulse delivered by a constant current stimulator (custom-built). One stimulus was given every 2 s during walking and running, which was slightly out of phase with the gait cycle. Therefore, stimuli were dispersed randomly over the gait cycle. A micro-switch was placed under the subjects heel and coupled to an electronic Voltage integrator, which was used to establish the exact timing of each stimulus within the gait cycle (Simonsen and Dyhre-Poulsen, 1999). Another but supra-maximal stimulus elicited a maximal M-wave 60 ms after the first stimulus (Fig. 1). A personal computer was triggered by the first stimulus and sampled the EMG from the soleus muscle for 120 ms at 20 kHz. Thus, each sweep contained the M-wave, the H-reflex and the maximal M-wave (Simonsen and Dyhre-Poulsen, 2011). The experiments were controlled by a computer program, which has earlier been described in detail (Simonsen and Dyhre-Poulsen, 1999; Ferris et al., 2001; Simonsen et al., 2012). The maximal M-wave (Mmax) varies in amplitude during the gait cycle (Dyhre-Poulsen and Simonsen, 2002). These variations are most probably due to cyclical changes in muscle length and fiber pennation angle during locomotion (Gerilovsky et al., 1989) and we have earlier shown that correcting for the variations in Mmax can be used to obtain a more accurate and stable stimulus intensity (Simonsen and Dyhre-Poulsen, 1999). The computer program controlled the stimulator and was set to produce an M-wave 5–45% of Mmax instead of 15–35% Mmax, which we normally use. The gait cycle was divided into 20 time slices of about 55 ms for walking and 40 ms for running. The peak to peak amplitude of the M-wave was used to measure the effective stimulus intensity. The integrator connected to the micro-switch placed under the heel was set to produce a ramp function from 0 to 2 V over 2 s. Knowing the slope of the ramp function allowed the computer to determine the number of the actual time slice just prior to stimulation as the integrator was reset by a foot-switch at each heel strike. During the experiment the computer program ‘taught’ itself to apply stimuli of an appropriate strength in each of the 20 time slices. The algoritm used for this has been described in detail by Simonsen et al. (2012).
s2 Mmax s1
H M
5 mV 60 ms Fig. 1. A single data sweep sampled at 20 kHz. S1: the first stimulus artefact, M: the M-wave, H: the H-reflex, S2: the second stimulus artefact, Mmax: the maximal Mwave elicited by a supramaximal stimulus 60 ms after the first stimulus.
440
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
% of Mmax 5
0
20
40
60
50 80
40
100
%
4
Soleus H-reflex
30 20
H-reflex (mV)
10 3
0 400 300
2
uV 200
M. tibialis anterior
Running
100
1
0 400
0
0
2
4
6
8
M-wave (mV)
300 M. soleus
uV 200 100
Fig. 2. A typical H-reflex recruitment curve for one subject. The vertical lines indicate the low (5–25% Mmax) and the high (25–45% Mmax) stimulus condition used during walking and running in the present study. It is seen that the descending part of the recruitment curve was used.
2.5. Data treatment After the experiments, all stimuli were allocated to one of two levels, which is indicated on an H-reflex recruitment curve in Fig. 2. The low stimuli consisted of M-waves from 5% to 25% of Mmax while the high stimuli were M-waves from 25% to 45% of Mmax. Each M-wave was expressed relative to the following Mmax in the same sweep (Fig. 1). Averaged over subjects, between 123 and 184 stimuli were used to describe the H-reflex modulations at low and high stimulus intensity, respectively. EMG activity of the soleus and the anterior tibial muscle was sampled from 10 gait cycles without stimuli. The signals were highpass filtered at 20 Hz, rectified and lowpass filtered at 15 Hz to form linear envelopes and finally averaged over the 10 gait cycles. Differences between the H-reflexes elicited by low and high stimuli were tested by a paired Student’s t-test. Level of significance was set to 5%.
0 50 40
%
20 10 0
uV
50 40 30 20 10 0
M. tibialis anterior Walking
300
uV
200
M. soleus
100 0
0
20
40
60
80
100
120
% Gait Cycle
3. Results The H-reflex modulation during walking and running is illustrated for one subject in Fig. 3. It is seen that the peak H-reflex is about the same during walking and running while the EMG activity in both m. soleus and m. tibialis anterior is much higher during running. Moreover, there is a marked co-contraction between the two muscles during running. The subject in Fig. 3 showed a gradually increasing H-reflex during the swing phase of walking while others showed almost complete suppression during the entire swing phase. During running all subjects showed almost complete suppression during the swing (and flight) phase but a marked increase just prior to heel strike. From walking to running, the peak soleus EMG amplitude increased significantly from 160 to 346 lV (p < 0.05) and peak tibialis activity increased also significantly from 164 to 237 lV (p < 0.05). When M-waves of 25 ± 10% of Mmax were used (normal stimulus intensity), the peak soleus H-reflex amplitude averaged across all subjects was 41% during walking and 47% of Mmax during running. This difference was not statistically significant, so accordingly the H-reflex was not lower during running as reported by Capaday and Stein (1987). Therefore, the question was now how an increased stimulus intensity would influence this observation. Fig. 4 illustrates
Soleus H-reflex
30
Fig. 3. The figure shows data from one typical subject stimulated by ‘normal’ intensity producing M-waves of 25 ± 10% Mmax. From top: (1) the soleus H-reflex modulation during running 8 km/h. Each circle represents an H-reflex from an accepted stimulus. The full line shows the reflexes averaged in 16 bins. The H-reflex amplitude is in% Mmax. (2 and 3) EMG from m. tibialis anterior and m. soleus averaged from 10 step cycles. (4) The soleus H-reflex modulation during walking at 4.5 km/h. and (5 and 6) averaged EMGs of m. tibialis anterior and m. soleus.
averaged H-reflex modulations of all subjects during walking and running but separately for low and high stimulus intensity. It is seen that the peak H-reflex during walking was significantly lower (p < 0.05) in the high intensity condition, while during running no differences were observed in peak H-reflex amplitude between the two stimulus conditions (Fig. 4). The mean H-reflex during the swing phase was not affected by the stimulus intensity neither during walking nor during running. For the mean H-reflex during the whole walking cycle no differences were observed due to stimulus intensity. 4. Discussion It was expected that the high stimulus intensity condition would decrease the H-reflex amplitude, which was also seen for
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
60
*
50
% Mmax
40 Walking
30 20 10 0
0
20
40
60
80
100
80
100
60 50
% Mmax
40 30
Running
20 10 0
0
20
40
60
% Gait Cycle Fig. 4. H-reflex modulation averaged from seven subjects during two levels of stimulus intensity. The low intensity is represented by circles (M-waves of 5–25% Mmax) and the high intensity by crosses (M-waves of 25–45% Mmax). On top is walking at 4.5 km/h and below running at 8 km/h. Errorbars are SE and statistically significant differences are indicated by asterisks.
the peak H-reflex during walking (Fig. 4). This is most likely due to a considerable difference in nerve conduction velocity between Ia afferents and a-motoneurones, which causes a collision between orthodromic and antidromic action potentials just outside the spinal cord in the axons of the a-motoneurones actually participating in the formation of the M-wave (Schieppati, 1987; Knikou, 2008). The reason why this takes place on the efferent axons is that the nerve conduction velocity of the Ia afferents is approximately 10 m s 1 faster than that of the a-motoneurones in the soleus muscle (Gasser and Grundfest, 1939). Why did this not seem to happen during running? It must be assumed that fewer axons from a-motoneurones are available to participate in the formation of the H-reflex at the highest M-waves due to collision. However, at a relatively high level of voluntary activity during running the Ia afferents are likely to produce a larger H-reflex as the a-motoneurones will be more excitable and closer to threshold. Another effect of the voluntary activity is that some of the action potentials caused by descending voluntary activity will collide with the antidromic potential in the efferent axons and thereby clear the axon for an H-reflex. This phenomenon is known as the V-wave, which normally are measured by eliciting a maximal M-wave during voluntary activity (Sale et al., 1983; Aagaard et al., 2002; Simonsen et al., 2012). However, the V-wave must exist to a certain degree also at submaximal stimulation intensities, which in this case were up to 45% of Mmax. In support for this, it is a common observation that the soleus H-reflex often cannot be extinguished completely (by collision) at Mmax during voluntary activity in the soleus, see e.g. Grosprêtre and Martin (2012). During running the H-reflex was seen to increase quickly in all subjects just prior to heel strike. This was accompanied by a steep
441
increase in EMG activity in the soleus muscle (Fig. 3) but also with considerable co-contraction of the anterior tibial muscle (Fig. 3). This was consistent for all subjects. Normally, a high anterior tibial activity implies a strong reciprocal inhibition on the soleus motoneurones (Crone et al., 1990), so it is likely that removal of presynaptic inhibition is responsible for this H-reflex modulation. The reason why the H-reflex was not lower at the high stimulus intensities is most likely that a high firing frequency often is seen in the beginning of a muscle contraction (Van Cutsem et al., 1998; Duchateau et al., 2006) and high firing frequency increases muscle stiffness (Rack and Westbury, 1974), which is the purpose of this preparatory activity in the soleus muscle. A high firing frequency of motorunits in this phase of the gait cycle will increase the V-wave effect, especially at the highest stimuli. The size of the H-reflex is also influenced by recurrent inhibition (Pierrot-Deseilligny et al., 1977), Ib inhibition (Pierrot-Deseilligny et al., 1979) and reciprocal inhibition as well as presynaptic inhibition (Crone et al., 1990; Knikou, 2008). However, these mechanisms are not likely to have obscured the effect of the high stimulus intensity in the present study. During voluntary activity as in running it is necessary to elicit an M-wave of an appropriate size in order to see the signal on top of the voluntary EMG, although the EMG is characterized by asynchronous firing of motorunits while the M-wave is a highly synchronous and compound signal. An M-wave 10% of Mmax was typically 1 mV in peak to peak amplitude and the EMG was in this case 300–400 lV. It could be argued that sweep averaging in e.g. 16 bins as seen previously (Capaday and Stein, 1987; Edamura et al., 1991) would remove the EMG. However, such a procedure precludes control of the M-wave amplitude during the experiment. Based on the results of the present study it can be concluded that the earlier reported differences in soleus H-reflex amplitude measured during running are unlikely to have been caused by the use of a too high stimulus strength. However, it is important to stress that stimulus strengths corresponding to more than 35% Mmax should be avoided in studies of locomotion as the H-reflex amplitude during walking was significantly suppressed during the stance phase when using stimulus intensities up to 45% Mmax. Conflict of interest There were no conflict of interests. References Aagaard P, Simonsen EB, Andersen J, Magnusson P, Dyhre-Poulsen P. Neural adaptation to resistance training: changes in evoked V-wave and H-reflex responses. J Appl Physiol 2002;92:2309–18. Capaday C, Stein RB. Amplitude modulation of the soleus H-reflex in the human during walking and standing. J Neurosci 1986;6:1308–13. Capaday C, Stein RB. Difference in the amplitude of the human soleus H reflex during walking and running. J Physiol 1987;392:513–22. Crenna P, Frigo C. Excitability of the soleus H-reflex arc during walking and stepping in man. Exp Brain Res 1987;66(1):40–60. Crone C, Hultborn H, Mazières L, Morin C, Nielsen J, Pierrot-Deseilligny E. Sensitivity of monosynaptic test reflexes to facilitation and inhibition as a function of the test reflex size: a study in man and the cat. Exp Brain Res 1990;81:35–45. Duchateau J, Semmler JG, Enoka RM. Training adaptations in the behavior of human motor units. J Appl Physiol 2006;101(6):1766–75. Dyhre-Poulsen P, Simonsen EB. H reflexes recorded during locomotion. Adv Exp Med Biol 2002;508:377–83. Edamura M, Yang JF, Stein RB. Factors that determine the magnitude and time course of human H-reflexes in locomotion. J Neurosci 1991;11:420–7. Ethier C, Imbeault MA, Ung V, Capaday C. On the soleus H-reflex modulation pattern during walking. Exp Brain Res 2003;151(3):420–5. Ferris DP, Aagaard P, Simonsen EB, Farley CT, Dyhre-Poulsen P. Soleus H-reflex gain in humans walking and running under simulated reduced gravity. J Physiol 2001;530(1):167–80. Frigon A, Carroll TJ, Jones KE, Zehr EP, Collins DF. Ankle position and voluntary contraction alter maximal M waves in soleus and tibialis anterior. Muscle Nerve 2007;35:756–66.
442
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
Gasser HS, Grundfest H. Axon diameters in relation to the spike dimensions and the conduction velocity in mammalian A-fibers. Am J Physiol 1939;127:393. Gerilovsky L, Tsvetinov P, Trenkowa G. Peripheral effects on the amplitude of monopolar and bipolar H-reflex potentials from the soleus muscle. Exp Brain Res 1989;76(1):173–81. Grosprêtre S, Martin A. H reflex and spinal excitability: methodological considerations. J Neurophysiol 2012;107:1649–54. Hodapp M, Vry J, Mall V, Faist M. Changes in soleus H-reflex modulation after treadmill training in children with cerebral palsy. Brain 2009;132:37–44. Knikou M. The H-reflex as a probe: pathways and pitfalls. J Neurosci Methods 2008; 171:1–12. Pierrot-Deseilligny E, Morin C, Katz R, Bussel B. Influence of voluntary movement and posture on recurrent inhibition in human subjects. Brain Res 1977;124: 427–36. Pierrot-Deseilligny E, Katz R, Morin C. Evidence for Ib inhibition in human subjects. Brain Res 1979;166:176–9. Rack PMH, Westbury DR. The short range stiffness of active mammalian muscle and its effect on mechanical properties. J Physiol 1974;240:331–50. Sale DG, MacDougall JD, Upton AR, McComas AJ. Effect of strength training upon motoneurone excitability in man. Med Sci Sport Exc 1983;15:57–62. Schieppati M. The Hoffman reflex: a means of assessing spinal reflex excitability and its descending control in man. Prog Neurobiol 1987;28:345–76. Simonsen EB, Dyhre-Poulsen P. Amplitude of the human soleus H reflex during walking and running. J Physiol 1999;513(3):929–39. Simonsen EB, Dyhre-Poulsen P. Test-retest reliability of the soleus H-reflex excitability measured during human walking. Hum Mov Sci 2011;30(2): 333–40. Simonsen EB, Dyhre-Poulsen P, Alkjaer T, Aagaard P, Magnusson SP. Interindividual differences in H reflex modulation during normal walking. Exp Brain Res 2002;142(1):108–15. Simonsen EB, Alkjær T, Raffalt PC. Reflex response and control of the human soleus and gastrocnemius muscles during walking and running at increasing velocity. Exp Brain Res 2012;219(2):163–74. Van Cutsem M, Duchateau J, Hainaut K. Changes in single motor unit behaviour contribute to the increase in contraction speed after dynamic training in humans. J Physiol 1998;513:295–305.
Erik B. Simonsen is an associate professor at The University of Copenhagen. He graduated in 1984 with a master degree in Human Physiology and a Ph.D. in biomechanics and neurophysiology in 1992 from The University of Copenhagen. His research interests include human gait, H-reflex modulation during movement, biomechanics of human gait.
Tine Alkjær graduated in 1998 with a master degree in Human Physiology from The University of Copenhagen. She received her Ph.D. in clinical gait analysis in 2002 from the University of Copenhagen where she is an associate professor at the Department of Neurophysiology. The main topics of her research are biomechanics and motor control of human walking, function of the anterior cruciate ligament and clinical gait analysis.
Peter Christian Raffalt received in 2010 his master degree from Department of Exercise and Sport Sciences at University of Copenhagen. His research areas since 2010 include reflex modulation during locomotion and movement variability. He is currently Ph.D. student at Department of Neuroscience and Pharmacology at University of Copenhagen with his dissertation topic focused on movement variability and motor control in children with different training background.
Contents lists available at SciVerse ScienceDirect
Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin
Influence of stimulus intensity on the soleus H-reflex amplitude and modulation during locomotion Erik B. Simonsen ⇑, Tine Alkjær, Peter C. Raffalt Department of Neuroscience and Pharmacology, University of Copenhagen, Denmark
a r t i c l e
i n f o
Article history: Received 4 January 2012 Received in revised form 17 September 2012 Accepted 30 October 2012
Keywords: H-reflex Walking Running EMG
a b s t r a c t Diverging results have been reported regarding the modulation and amplitude of the soleus H-reflex measured during human walking and running. A possible explanation to this could be the use of too high stimulus strength in some studies while not in others. During activities like walking and running it is necessary to use a small M-wave to control the effective stimulus strength during all phases of the movement. This implies that the descending part of the H-reflex recruitment curve is being used, which may lead to an unwanted suppression of the H-reflex due to limitations imbedded within the H-reflex methodology itself. Accordingly, the purpose of the present study was to study the effect on the soleus H-reflex during walking and running using stimulus intensities normally considered too high (up to 45% Mmax). Using M-waves of 25–45% Mmax as opposed to 5–25% Mmax showed a significant suppression of the peak H-reflex during the stance phase of walking, while no changes were observed during running. No differences were observed regarding modulation pattern. So a possible use of too high stimulus intensity cannot explain the differences mentioned. The surprising result in running may be explained by the much higher voluntary muscle activity, which implies the existence of a V-wave influencing the H-reflex amplitude in positive direction. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction It is commonly known that the H-reflex is most sensitive to spinal modulation on the ascending part of the H-reflex recruitment curve (Schieppati, 1987) and recently it was stated by Grosprêtre and Martin (2012) that during motor tasks, changes in spinal excitability should be assessed in the ascending part of the curve, where H-reflex modulations do not depend on background electrical activity of the muscle tested. During dynamic locomotion like walking and running it is a challenge to provide the same stimulus intensity in all phases of the movement cycle. This cannot easily be obtained, as the same stimulus intensity produces a varying effective stimulus strength in different phases of the movement, which means that a varying number of axons are activated. When measuring the soleus H-reflex, this is due to the stimulating cathode moving with respect to the tibial nerve in the popliteal fossa when the knee joint is being flexed and extended (Dyhre-Poulsen and Simonsen, 2002). The solution is to produce a small M-wave and keep the peak to ⇑ Corresponding author. Address: Department of Neuroscience and Pharmacology, University of Copenhagen, Blegdamsvej 3, 2200 Copenhagen N, Denmark. Tel.: +45 28 75 72 30. E-mail address: [email protected] (E.B. Simonsen). 1050-6411/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jelekin.2012.10.019
peak amplitude as constant as possible during all phases of the movement. The rationale is that if a constant number of efferents are stimulated it is assumed that also a constant number of afferents are stimulated, which then makes it possible to study spinal modulation of the H-reflex during movement. It is, however, a drawback that stimulating for a small M-wave means that the descending part of the recruitment curve is being used. Another problem concerning measurement of the soleus H-reflex during gait is that the amplitude of the evoked potentials varies as the muscle fibers move with respect to the recording EMG electrodes, i.e. the amplitude of the M-wave and the H-reflex varies as a function of the ankle joint angle (Dyhre-Poulsen and Simonsen, 2002). This phenomenon was also shown during static conditions (Gerilovsky et al., 1989) and by Frigon et al. (2007) who found a signicantly higher Mmax at the shortest muscle lengths in the soleus. The solution to this problem is to measure the maximal M-wave in all phases of the movement by a supramaximal stimulus 60 ms after eliciting the test stimulus. Then each M-wave may be expressed relative to the variations in Mmax during the gait cycle (Dyhre-Poulsen and Simonsen, 2002). Due to this phenomenon, the ascending part of the recruitment curve cannot be used during locomotion as a low stimulus producing no M-wave could result in an H-reflex ranging between zero and maximum as the effective stimulus intensity would be unknown.
439
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
Accordingly, it is imperative to ensure that a small M-wave on the beginning of the descending part of the recruitment curve is being used and corrected for fluctuations in Mmax. Especially if too large M-waves and thereby excessive stimulus strength is being used the measured modulation of the H-reflex may be impaired severely. In the gait studies of Simonsen and Dyhre-Poulsen (1999), Simonsen et al. (2002) and Ferris et al. (2001) M-waves of 25 ± 10% Mmax were used and corrected for fluctuations in Mmax. However in all other studies of the soleus H-reflex modulation during human gait no such corrections were applied (e.g. Capaday and Stein, 1986; Capaday and Stein, 1987; Crenna and Frigo, 1987; Edamura et al., 1991; Ethier et al., 2003; Hodapp et al., 2009). Discrepancies in the amplitude of the soleus H-reflex were seen in these studies. Especially during running, where Capaday and Stein (1987) reported a lower soleus H-reflex during running compared with walking, whereas Simonsen and Dyhre-Poulsen (1999) found either a similar excitability during walking and running or a higher excitability during faster running speed. Also during the swing phase differences have been reported as Simonsen et al. (2002) found a gradually increasing H-reflex during the swing phase of walking in about 40% of their subjects and an almost completely suppressed H-reflex in the rest. In all other studies the H-reflex was suppressed during the swing phase. Theoretically, if too high stimulus intensity was used in the latter studies it may have depressed the excitability of the H-reflex during the swing phase in all subjects and limited the possible spinal modulation in general. The purpose of the present study was therefore to quantify the expected suppression caused by the use of relatively high stimulus strengths compared to those normally used in studies of H-reflex modulation during human gait. It was hypothesized that the high stimulus strength would produce a significantly lower H-reflex in most phases of the gait cycle. 2. Methods 2.1. Subjects Seven male subjects accustomed to treadmill locomotion volunteered to participate in the experiments. They gave their informed consent to the conditions of the experiments, which were approved by the local ethics committee. Mean age was 28 years (24–35), mean height 1.85 m (1.72–1.95) and mean body weight 76 kg (65–89). 2.2. Procedure When the subjects arrived EMG electrodes were mounted on the soleus muscle and the site of nerve stimulation was located. The maximal M-wave was then measured at rest in the standing position by giving supra-maximal stimuli until the M-wave ceased to increase. The subjects performed walking and running on a motor-driven treadmill (HS-1200; Technogym) at 4.5 (walking) and 8 (running) km h 1 while the soleus H-reflex excitability was measured during the gait cycle. 2.3. Electromyography The electrodes (Medicotest Q-10-A) were positioned dorsally on the right leg two cm below the visual muscle belly of the gastrocnemius muscle. A reference electrode was placed over the tibia. The skin was shaved, lightly abraded with emery-cloth and cleaned with alcohol. The electrodes were connected to small custom-built preamplifiers (input impedance 80 MX) taped to the skin. The EMG signals were then led through 5 m long shielded wires to custom-built amplifiers with a frequency response of 20 Hz to 10 kHz.
2.4. H-reflex The site of stimulation of the tibial nerve was carefully located in the popliteal fossa using a hand held probe electrode. Conditions for the optimum site were that (1) the M-wave and the H reflex should preferably have the same visual shape and (2) it should be possible to evoke the H-reflex without any visible M-wave. After location of the optimum site, the permanent stimulus electrode (Ambu VL-00-A) was placed over the tibial nerve. A 40 mm diameter anode was placed over the patella. The stimulus was a 1 ms square wave pulse delivered by a constant current stimulator (custom-built). One stimulus was given every 2 s during walking and running, which was slightly out of phase with the gait cycle. Therefore, stimuli were dispersed randomly over the gait cycle. A micro-switch was placed under the subjects heel and coupled to an electronic Voltage integrator, which was used to establish the exact timing of each stimulus within the gait cycle (Simonsen and Dyhre-Poulsen, 1999). Another but supra-maximal stimulus elicited a maximal M-wave 60 ms after the first stimulus (Fig. 1). A personal computer was triggered by the first stimulus and sampled the EMG from the soleus muscle for 120 ms at 20 kHz. Thus, each sweep contained the M-wave, the H-reflex and the maximal M-wave (Simonsen and Dyhre-Poulsen, 2011). The experiments were controlled by a computer program, which has earlier been described in detail (Simonsen and Dyhre-Poulsen, 1999; Ferris et al., 2001; Simonsen et al., 2012). The maximal M-wave (Mmax) varies in amplitude during the gait cycle (Dyhre-Poulsen and Simonsen, 2002). These variations are most probably due to cyclical changes in muscle length and fiber pennation angle during locomotion (Gerilovsky et al., 1989) and we have earlier shown that correcting for the variations in Mmax can be used to obtain a more accurate and stable stimulus intensity (Simonsen and Dyhre-Poulsen, 1999). The computer program controlled the stimulator and was set to produce an M-wave 5–45% of Mmax instead of 15–35% Mmax, which we normally use. The gait cycle was divided into 20 time slices of about 55 ms for walking and 40 ms for running. The peak to peak amplitude of the M-wave was used to measure the effective stimulus intensity. The integrator connected to the micro-switch placed under the heel was set to produce a ramp function from 0 to 2 V over 2 s. Knowing the slope of the ramp function allowed the computer to determine the number of the actual time slice just prior to stimulation as the integrator was reset by a foot-switch at each heel strike. During the experiment the computer program ‘taught’ itself to apply stimuli of an appropriate strength in each of the 20 time slices. The algoritm used for this has been described in detail by Simonsen et al. (2012).
s2 Mmax s1
H M
5 mV 60 ms Fig. 1. A single data sweep sampled at 20 kHz. S1: the first stimulus artefact, M: the M-wave, H: the H-reflex, S2: the second stimulus artefact, Mmax: the maximal Mwave elicited by a supramaximal stimulus 60 ms after the first stimulus.
440
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
% of Mmax 5
0
20
40
60
50 80
40
100
%
4
Soleus H-reflex
30 20
H-reflex (mV)
10 3
0 400 300
2
uV 200
M. tibialis anterior
Running
100
1
0 400
0
0
2
4
6
8
M-wave (mV)
300 M. soleus
uV 200 100
Fig. 2. A typical H-reflex recruitment curve for one subject. The vertical lines indicate the low (5–25% Mmax) and the high (25–45% Mmax) stimulus condition used during walking and running in the present study. It is seen that the descending part of the recruitment curve was used.
2.5. Data treatment After the experiments, all stimuli were allocated to one of two levels, which is indicated on an H-reflex recruitment curve in Fig. 2. The low stimuli consisted of M-waves from 5% to 25% of Mmax while the high stimuli were M-waves from 25% to 45% of Mmax. Each M-wave was expressed relative to the following Mmax in the same sweep (Fig. 1). Averaged over subjects, between 123 and 184 stimuli were used to describe the H-reflex modulations at low and high stimulus intensity, respectively. EMG activity of the soleus and the anterior tibial muscle was sampled from 10 gait cycles without stimuli. The signals were highpass filtered at 20 Hz, rectified and lowpass filtered at 15 Hz to form linear envelopes and finally averaged over the 10 gait cycles. Differences between the H-reflexes elicited by low and high stimuli were tested by a paired Student’s t-test. Level of significance was set to 5%.
0 50 40
%
20 10 0
uV
50 40 30 20 10 0
M. tibialis anterior Walking
300
uV
200
M. soleus
100 0
0
20
40
60
80
100
120
% Gait Cycle
3. Results The H-reflex modulation during walking and running is illustrated for one subject in Fig. 3. It is seen that the peak H-reflex is about the same during walking and running while the EMG activity in both m. soleus and m. tibialis anterior is much higher during running. Moreover, there is a marked co-contraction between the two muscles during running. The subject in Fig. 3 showed a gradually increasing H-reflex during the swing phase of walking while others showed almost complete suppression during the entire swing phase. During running all subjects showed almost complete suppression during the swing (and flight) phase but a marked increase just prior to heel strike. From walking to running, the peak soleus EMG amplitude increased significantly from 160 to 346 lV (p < 0.05) and peak tibialis activity increased also significantly from 164 to 237 lV (p < 0.05). When M-waves of 25 ± 10% of Mmax were used (normal stimulus intensity), the peak soleus H-reflex amplitude averaged across all subjects was 41% during walking and 47% of Mmax during running. This difference was not statistically significant, so accordingly the H-reflex was not lower during running as reported by Capaday and Stein (1987). Therefore, the question was now how an increased stimulus intensity would influence this observation. Fig. 4 illustrates
Soleus H-reflex
30
Fig. 3. The figure shows data from one typical subject stimulated by ‘normal’ intensity producing M-waves of 25 ± 10% Mmax. From top: (1) the soleus H-reflex modulation during running 8 km/h. Each circle represents an H-reflex from an accepted stimulus. The full line shows the reflexes averaged in 16 bins. The H-reflex amplitude is in% Mmax. (2 and 3) EMG from m. tibialis anterior and m. soleus averaged from 10 step cycles. (4) The soleus H-reflex modulation during walking at 4.5 km/h. and (5 and 6) averaged EMGs of m. tibialis anterior and m. soleus.
averaged H-reflex modulations of all subjects during walking and running but separately for low and high stimulus intensity. It is seen that the peak H-reflex during walking was significantly lower (p < 0.05) in the high intensity condition, while during running no differences were observed in peak H-reflex amplitude between the two stimulus conditions (Fig. 4). The mean H-reflex during the swing phase was not affected by the stimulus intensity neither during walking nor during running. For the mean H-reflex during the whole walking cycle no differences were observed due to stimulus intensity. 4. Discussion It was expected that the high stimulus intensity condition would decrease the H-reflex amplitude, which was also seen for
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
60
*
50
% Mmax
40 Walking
30 20 10 0
0
20
40
60
80
100
80
100
60 50
% Mmax
40 30
Running
20 10 0
0
20
40
60
% Gait Cycle Fig. 4. H-reflex modulation averaged from seven subjects during two levels of stimulus intensity. The low intensity is represented by circles (M-waves of 5–25% Mmax) and the high intensity by crosses (M-waves of 25–45% Mmax). On top is walking at 4.5 km/h and below running at 8 km/h. Errorbars are SE and statistically significant differences are indicated by asterisks.
the peak H-reflex during walking (Fig. 4). This is most likely due to a considerable difference in nerve conduction velocity between Ia afferents and a-motoneurones, which causes a collision between orthodromic and antidromic action potentials just outside the spinal cord in the axons of the a-motoneurones actually participating in the formation of the M-wave (Schieppati, 1987; Knikou, 2008). The reason why this takes place on the efferent axons is that the nerve conduction velocity of the Ia afferents is approximately 10 m s 1 faster than that of the a-motoneurones in the soleus muscle (Gasser and Grundfest, 1939). Why did this not seem to happen during running? It must be assumed that fewer axons from a-motoneurones are available to participate in the formation of the H-reflex at the highest M-waves due to collision. However, at a relatively high level of voluntary activity during running the Ia afferents are likely to produce a larger H-reflex as the a-motoneurones will be more excitable and closer to threshold. Another effect of the voluntary activity is that some of the action potentials caused by descending voluntary activity will collide with the antidromic potential in the efferent axons and thereby clear the axon for an H-reflex. This phenomenon is known as the V-wave, which normally are measured by eliciting a maximal M-wave during voluntary activity (Sale et al., 1983; Aagaard et al., 2002; Simonsen et al., 2012). However, the V-wave must exist to a certain degree also at submaximal stimulation intensities, which in this case were up to 45% of Mmax. In support for this, it is a common observation that the soleus H-reflex often cannot be extinguished completely (by collision) at Mmax during voluntary activity in the soleus, see e.g. Grosprêtre and Martin (2012). During running the H-reflex was seen to increase quickly in all subjects just prior to heel strike. This was accompanied by a steep
441
increase in EMG activity in the soleus muscle (Fig. 3) but also with considerable co-contraction of the anterior tibial muscle (Fig. 3). This was consistent for all subjects. Normally, a high anterior tibial activity implies a strong reciprocal inhibition on the soleus motoneurones (Crone et al., 1990), so it is likely that removal of presynaptic inhibition is responsible for this H-reflex modulation. The reason why the H-reflex was not lower at the high stimulus intensities is most likely that a high firing frequency often is seen in the beginning of a muscle contraction (Van Cutsem et al., 1998; Duchateau et al., 2006) and high firing frequency increases muscle stiffness (Rack and Westbury, 1974), which is the purpose of this preparatory activity in the soleus muscle. A high firing frequency of motorunits in this phase of the gait cycle will increase the V-wave effect, especially at the highest stimuli. The size of the H-reflex is also influenced by recurrent inhibition (Pierrot-Deseilligny et al., 1977), Ib inhibition (Pierrot-Deseilligny et al., 1979) and reciprocal inhibition as well as presynaptic inhibition (Crone et al., 1990; Knikou, 2008). However, these mechanisms are not likely to have obscured the effect of the high stimulus intensity in the present study. During voluntary activity as in running it is necessary to elicit an M-wave of an appropriate size in order to see the signal on top of the voluntary EMG, although the EMG is characterized by asynchronous firing of motorunits while the M-wave is a highly synchronous and compound signal. An M-wave 10% of Mmax was typically 1 mV in peak to peak amplitude and the EMG was in this case 300–400 lV. It could be argued that sweep averaging in e.g. 16 bins as seen previously (Capaday and Stein, 1987; Edamura et al., 1991) would remove the EMG. However, such a procedure precludes control of the M-wave amplitude during the experiment. Based on the results of the present study it can be concluded that the earlier reported differences in soleus H-reflex amplitude measured during running are unlikely to have been caused by the use of a too high stimulus strength. However, it is important to stress that stimulus strengths corresponding to more than 35% Mmax should be avoided in studies of locomotion as the H-reflex amplitude during walking was significantly suppressed during the stance phase when using stimulus intensities up to 45% Mmax. Conflict of interest There were no conflict of interests. References Aagaard P, Simonsen EB, Andersen J, Magnusson P, Dyhre-Poulsen P. Neural adaptation to resistance training: changes in evoked V-wave and H-reflex responses. J Appl Physiol 2002;92:2309–18. Capaday C, Stein RB. Amplitude modulation of the soleus H-reflex in the human during walking and standing. J Neurosci 1986;6:1308–13. Capaday C, Stein RB. Difference in the amplitude of the human soleus H reflex during walking and running. J Physiol 1987;392:513–22. Crenna P, Frigo C. Excitability of the soleus H-reflex arc during walking and stepping in man. Exp Brain Res 1987;66(1):40–60. Crone C, Hultborn H, Mazières L, Morin C, Nielsen J, Pierrot-Deseilligny E. Sensitivity of monosynaptic test reflexes to facilitation and inhibition as a function of the test reflex size: a study in man and the cat. Exp Brain Res 1990;81:35–45. Duchateau J, Semmler JG, Enoka RM. Training adaptations in the behavior of human motor units. J Appl Physiol 2006;101(6):1766–75. Dyhre-Poulsen P, Simonsen EB. H reflexes recorded during locomotion. Adv Exp Med Biol 2002;508:377–83. Edamura M, Yang JF, Stein RB. Factors that determine the magnitude and time course of human H-reflexes in locomotion. J Neurosci 1991;11:420–7. Ethier C, Imbeault MA, Ung V, Capaday C. On the soleus H-reflex modulation pattern during walking. Exp Brain Res 2003;151(3):420–5. Ferris DP, Aagaard P, Simonsen EB, Farley CT, Dyhre-Poulsen P. Soleus H-reflex gain in humans walking and running under simulated reduced gravity. J Physiol 2001;530(1):167–80. Frigon A, Carroll TJ, Jones KE, Zehr EP, Collins DF. Ankle position and voluntary contraction alter maximal M waves in soleus and tibialis anterior. Muscle Nerve 2007;35:756–66.
442
E.B. Simonsen et al. / Journal of Electromyography and Kinesiology 23 (2013) 438–442
Gasser HS, Grundfest H. Axon diameters in relation to the spike dimensions and the conduction velocity in mammalian A-fibers. Am J Physiol 1939;127:393. Gerilovsky L, Tsvetinov P, Trenkowa G. Peripheral effects on the amplitude of monopolar and bipolar H-reflex potentials from the soleus muscle. Exp Brain Res 1989;76(1):173–81. Grosprêtre S, Martin A. H reflex and spinal excitability: methodological considerations. J Neurophysiol 2012;107:1649–54. Hodapp M, Vry J, Mall V, Faist M. Changes in soleus H-reflex modulation after treadmill training in children with cerebral palsy. Brain 2009;132:37–44. Knikou M. The H-reflex as a probe: pathways and pitfalls. J Neurosci Methods 2008; 171:1–12. Pierrot-Deseilligny E, Morin C, Katz R, Bussel B. Influence of voluntary movement and posture on recurrent inhibition in human subjects. Brain Res 1977;124: 427–36. Pierrot-Deseilligny E, Katz R, Morin C. Evidence for Ib inhibition in human subjects. Brain Res 1979;166:176–9. Rack PMH, Westbury DR. The short range stiffness of active mammalian muscle and its effect on mechanical properties. J Physiol 1974;240:331–50. Sale DG, MacDougall JD, Upton AR, McComas AJ. Effect of strength training upon motoneurone excitability in man. Med Sci Sport Exc 1983;15:57–62. Schieppati M. The Hoffman reflex: a means of assessing spinal reflex excitability and its descending control in man. Prog Neurobiol 1987;28:345–76. Simonsen EB, Dyhre-Poulsen P. Amplitude of the human soleus H reflex during walking and running. J Physiol 1999;513(3):929–39. Simonsen EB, Dyhre-Poulsen P. Test-retest reliability of the soleus H-reflex excitability measured during human walking. Hum Mov Sci 2011;30(2): 333–40. Simonsen EB, Dyhre-Poulsen P, Alkjaer T, Aagaard P, Magnusson SP. Interindividual differences in H reflex modulation during normal walking. Exp Brain Res 2002;142(1):108–15. Simonsen EB, Alkjær T, Raffalt PC. Reflex response and control of the human soleus and gastrocnemius muscles during walking and running at increasing velocity. Exp Brain Res 2012;219(2):163–74. Van Cutsem M, Duchateau J, Hainaut K. Changes in single motor unit behaviour contribute to the increase in contraction speed after dynamic training in humans. J Physiol 1998;513:295–305.
Erik B. Simonsen is an associate professor at The University of Copenhagen. He graduated in 1984 with a master degree in Human Physiology and a Ph.D. in biomechanics and neurophysiology in 1992 from The University of Copenhagen. His research interests include human gait, H-reflex modulation during movement, biomechanics of human gait.
Tine Alkjær graduated in 1998 with a master degree in Human Physiology from The University of Copenhagen. She received her Ph.D. in clinical gait analysis in 2002 from the University of Copenhagen where she is an associate professor at the Department of Neurophysiology. The main topics of her research are biomechanics and motor control of human walking, function of the anterior cruciate ligament and clinical gait analysis.
Peter Christian Raffalt received in 2010 his master degree from Department of Exercise and Sport Sciences at University of Copenhagen. His research areas since 2010 include reflex modulation during locomotion and movement variability. He is currently Ph.D. student at Department of Neuroscience and Pharmacology at University of Copenhagen with his dissertation topic focused on movement variability and motor control in children with different training background.