| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Prince of Wales Medical Research Institute and University of New South Wales, Sydney, New South Wales 2031, Australia
Submitted 12 December 2003 ; accepted in final form 15 March 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
twitch interpolation; motor cortex; muscle strength; reliability
During maximal voluntary contractions (MVCs), the presence of a superimposed twitch produced by motor nerve stimulation demonstrates that force produced by the muscle is less than maximal and suggests that either the stimulated axons are not all recruited voluntarily or they are discharging at subtetanic rates. Hence, voluntary activation must be less than maximal (e.g., Refs. 4, 9, 14, 18). Voluntary activation is quantified by comparing the amplitude of the superimposed twitch with the force evoked by the same stimulus at rest, immediately after a contraction (termed "resting twitch": e.g., Refs. 3, 5, 7, 23). The prior contraction potentiates the muscle. If voluntary activation is incomplete, then failure to drive the muscle must have occurred at or above the site of stimulation of the motor axons.
More recently, stimulation of the motor cortex has been used to estimate voluntary activation to further localize the site of failure of voluntary drive. Here, the presence of a superimposed twitch implies that output from the motor cortex was not optimal to drive the muscle maximally at the time of stimulation. Thus one site of failure of voluntary drive must be at or above the level of motor cortical output. This method has revealed submaximal voluntary activation of muscles during maximal voluntary efforts of respiratory muscles (12) and subsequently of thenar (13), elbow extensor (24), and elbow flexor muscles (26). It has also revealed central fatigue, a progressive decline in voluntary activation, during maximal efforts (10, 16, 22, 25). Voluntary activation tested by motor cortical stimulation is not necessarily equivalent to that tested by motor nerve stimulation. For example, with pathology of the corticospinal neurons or motoneurons, a subject may be unable to drive the muscle fully. In such a case, motor nerve stimulation distal to the pathology would reveal impaired voluntary activation, whereas motor cortical stimulation would not. Thus comparison of the two measures may help localize the site of impairment of voluntary drive.
Quantification of voluntary activation with motor cortical stimulation is difficult. The motoneuronal output evoked by the cortical stimulus with the muscle at rest is less than that during a contraction because motor cortical and motoneuronal "excitability" increase with activity (Refs. 8, 15, 27; for review, see Ref. 20). Thus the same magnetic stimulus evokes less cortical output, which recruits fewer motor units during rest than during voluntary activity. Therefore, it is inappropriate to normalize the superimposed twitch to that evoked at rest. Other studies have normalized the superimposed twitch to the background voluntary force before the stimulus (10, 16, 22, 25). This method is also problematic because background voluntary force production is affected by changes that occur in the central nervous system and the muscle (e.g., during fatigue). To avoid these difficulties, we recently proposed a method to estimate the resting twitch that would be evoked by motor cortical stimulation if cortical and motoneuronal excitability were maintained during rest. We then used this to measure voluntary activation in a new way (26). This estimation is based on extrapolation of the relationship between the size of the superimposed twitch evoked by motor cortical stimulation and voluntary force (24, 26). This relationship is negative and linear for elbow flexion forces above 50% of maximum whether the muscle is fresh or fatigued by isometric exercise (26). The extrapolation was calculated from data derived from cortical stimulation during 35 contractions. However, the performance of 35 contractions is time consuming and not practical for experimental conditions where voluntary activation may be changing over time, such as during the development of, or recovery, from fatigue or in patient groups who are unable to perform such large numbers of contractions. In the present study, we have refined this method to reduce the number of contractions required to estimate the resting twitch and to improve its practical application. We have shown that the resting twitch can be reliably estimated with as few as two or three contractions and that its variability is similar to the resting twitch evoked by motor nerve stimulation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Force and EMG Recordings
Isometric elbow flexion torque was measured by using a linear strain gauge (Xtran, Melbourne, Australia: linear to 2 kN). Electromyographic (EMG) activity was recorded with surface electrodes (Ag-AgCl, 10-mm diameter) positioned over the tendon and middle of the muscle belly of biceps brachii and the long head of triceps brachii. Surface EMG signals were amplified (100–300x), filtered (16–1,000 Hz), and sampled (2,000 Hz) for later analysis using a data acquisition system (CED 1401 interface with Signal software, Cambridge Electronic Design, Cambridge, UK).
Stimulation
Three forms of stimulation were used. These included stimulation of the brachial plexus, stimulation of the biceps and brachialis motor nerve, and transcranial magnetic stimulation of the motor cortex (TMS).
Stimulation of the brachial plexus. In the experiment setup before voluntary contractions, the size of the resting maximal compound muscle action potential (Mmax) was determined with surface EMG. It was used as a reference for the size of the EMG responses evoked by TMS. Single electrical stimuli of 100-µs duration were delivered to the brachial plexus while the subjects were at rest via a cathode in the supraclavicular fossa (Erb's point) and an anode on the acromion (constant current, DS7, Digitimer, Welwyn Garden City, Hertfordshire, UK, modified to deliver up to 1 A). The stimulus (60–180 mA) produced a resting Mmax with an average amplitude of 16.6 ± 7.0 and 7.4 ± 3.0 mV for the biceps and triceps, respectively.
Stimulation of the biceps brachii and brachialis motor nerve. Single electrical stimuli of 100-µs duration were delivered (constant current, DS7, Digitimer) to biceps and brachialis via a surface cathode located midway between the anterior edge of the deltoid and the elbow crease with the elbow flexed at 90° and via a surface anode positioned over the bicipital tendon. The stimulus intensity (120–480 mA) was set 20% above the level required to produce a resting twitch of maximal amplitude in unpotentiated muscle (i.e., muscle not potentiated by a prior contraction).
TMS. A circular coil (13.5-cm outside diameter) positioned over the vertex evoked motor-evoked potentials (MEPs) recorded from biceps and triceps (Magstim 200, Magstim, Dyfed, UK). The direction of current flow in the coil preferentially activated the left motor cortex. The stimulator output (50–90% of maximum) was set to produce a large MEP in biceps (minimum amplitude 50% of Mmax) during brief MVCs of the elbow flexor muscles and only a small MEP in triceps (usually <10–15% of Mmax). Stimulus intensity remained constant throughout the protocol.
Protocol
Subjects participated in two identical experiments that were performed at least 24 h apart. Each experiment started with up to three brief (1–2 s) control MVCs, from which the peak torque was measured. Two submaximal target torques of 50 and 75% of the largest maximal control effort were displayed on a visual-feedback device. Subjects performed 10 sets of contractions with 1–2 min of rest between sets to avoid fatigue. Each set involved a brief MVC (1–2 s) followed in a pseudorandom order by either a brief 50% MVC or a 75% MVC and then another brief contraction to the other target torque (Fig. 1A). Contractions in a set were separated by rests of 6 s. A single motor cortical stimulus was delivered during each contraction, and a motor nerve stimulus was delivered at rest between the maximal effort and the first submaximal contraction (2 s after completion of the MVC).
|
Voluntary activation was quantified by measurement of the torque responses to motor cortical stimulation. Any twitch increment in elbow flexion torque evoked during a contraction (superimposed twitch; e.g., Fig. 1B) was expressed as a fraction of the estimated amplitude of the response that could be evoked by the same stimulus in relaxed muscle ("estimated resting twitch"). We then quantified the level of drive as a percentage using the following equation: voluntary activation (%) = (1 – superimposed twitch/estimated resting twitch) x 100.
Four methods were used to estimate the amplitude of the resting twitch evoked by motor cortical stimulation for each subject. Each involved linear regression between the amplitude of the superimposed twitch and voluntary torque. The y-intercept was taken as the estimated amplitude of the resting twitch (Fig. 2). Data from different contractions were used to generate the linear regressions for each method. For the first method, a single linear regression was performed on data from all contractions (30 contractions, 1 regression per subject; Fig. 2A). In the second method, a linear regression was performed for each set of three contractions (10 regressions per subject; Fig. 2B). The third and fourth methods involved a linear regression for each pair of contractions, MVC followed by a 50% MVC or MVC followed by a 75% MVC, respectively (5 regressions per subject per method; Fig. 2, C and D).
|
Statistics
In the text and table, group data are presented as means ± SD, whereas in the figures, means ± SE are shown. Statistical analysis involved two-way repeated-measures ANOVA for comparison of the MEP size during the 50, 75, and 100% MVCs performed in the two testing sessions. Two-way repeated-measures ANOVA was also used for comparison between the resting twitch evoked by motor nerve stimulation and the four types of estimated resting twitch evoked by motor cortical stimulation in the two testing sessions. To assess variability within a session, coefficients of variation (CV; SD/mean x 100; in %) were calculated for each subject. Comparison of the variability in size of the resting twitch evoked by motor nerve stimulation and the estimated resting twitches evoked by TMS was made with one-way repeated-measures ANOVA. Post hoc discrimination between means was made with the Student-Newman-Keuls procedure. A Friedman repeated-measures ANOVA on ranks was used for comparison of the variability in voluntary activation derived with the four estimated resting twitches because these data were not normally distributed. Equivalence tests were performed to determine whether the estimated resting twitches calculated with methods 2, 3, and 4 were similar to that calculated with method 1. Each equivalence test used 10% margins and involved two one-sided t-tests on the difference in the estimated resting twitch between method 1 and the other method. The null hypothesis for these tests was that the mean difference was greater than ±10% of the average resting twitch calculated with method 1 (i.e., the original method, Ref. 26). The reproducibility of the resting twitch between sessions was quantified with the intraclass correlation coefficient (ICC2,1) based on a repeated-measures ANOVA with session as the independent variable (21). Statistical significance was set at the 5% level.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Superimposed Twitch and Voluntary Torque
For each subject, the size of the superimposed twitch evoked by motor cortical stimulation decreased linearly with increasing contraction strength for contractions of 50–100% of maximum (day 1: average r2 = 0.92 ± 0.07; day 2: average r2 = 0.95 ± 0.03; P < 0.001). Figure 3A shows this relationship for one subject. In the first session, the average size of the superimposed twitch was 7.4 ± 2.6 N·m (11.3 ± 2.1% MVC) during the 50% MVC and 0.8 ± 0.7 N·m (1.2 ± 1.3% MVC) during maximal efforts (Fig. 3B). Its size was similar in the two testing sessions (P = 0.75; ICC2,1 = 0.98).
|
Four methods were used to estimate the size of the resting twitch evoked by motor cortical stimulation for each subject (see MATERIALS AND METHODS; Fig. 2). Table 1 shows the average estimated resting twitch calculated with each method. Methods 2 and 3 yielded an estimated resting twitch of similar size to method 1 (test of equivalence, P < 0.01 for each t-test), with comparable results between days (P = 0.44; ICC2,1 > 0.98). However, the estimated resting twitch calculated with method 4 (MVC followed by a 75% MVC) was not similar to method 1 (test of equivalence, P > 0.05 for each test). It also had a greater within- and between-session variability (P < 0.05).
|
Voluntary Activation Measured With Motor Cortical Stimulation
As expected, for each subject, voluntary activation measured with the use of motor cortical stimulation increased linearly with increasing voluntary contraction strength regardless of the method used to estimate the size of the resting twitch (method 1: r2 = 0.92 ± 0.07; method 2: r2 = 0.92 ± 0.12; method 3: r2 = 0.92 ± 0.12; method 4: r2 = 0.86 ± 0.09; P < 0.01). Figure 4 shows the average increase in voluntary activation with torque. Across sessions, voluntary activation was high during maximal efforts (method 1: 94.1 ± 6.1%; method 2: 93.9 ± 6.5%; method 3: 94.1 ± 6.0%; method 4: 94.4 ± 4.8%) and was reproducible between sessions for each method of estimation of the resting twitch (ICC2,1 > 0.86). During MVCs, the average variability of voluntary activation within a session was low for methods 1 (CV = 2.9 ± 1.9%), 2 (CV = 3.7 ± 3.9%), and 3 (CV = 2.8 ± 2.0%), but it was significantly higher for method 4 (CV = 8.9 ± 19.6%; P < 0.05).
|
The size of the MEP was normalized to resting Mmax for between-subject comparisons. Figure 5 shows that, for biceps, the largest MEP occurred during the 50% MVC (average area across days: 85.1 ± 27.2% of Mmax) and that its size decreased with increasing contraction strength (100% MVC; average area across days 55.3 ± 11.6% of Mmax; P < 0.001). Its size was similar for both testing sessions (P = 0.38). In contrast, the size of the triceps MEP was greatest during maximal efforts (average area: 10.8 ± 3.6% Mmax).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
The MEP size for biceps and its antagonist triceps muscle was reproducible between sessions for contractions of 50–100% of maximum. This suggests that the stimulus was similar on both occasions. During 50% MVCs, when the superimposed twitch is largest, the size of the MEP in biceps approached that of Mmax. This suggests that under these conditions most motoneurons were activated by the motor cortical stimulus. During maximal efforts, the MEP size is smaller and the superimposed twitch is very small or in some cases completely absent (10, 22, 26). The smaller MEP may reflect the inability of some motoneurons to fire in response to the stimulus. With higher firing rates in the stronger contractions, more motoneurons may be effectively "refractory" because of their intrinsic properties and the trajectory of the afterhyperpolarization (see Refs. 17, 26). The relatively small size of the triceps MEP at all contraction strengths is important for measurement of voluntary activation because activation of triceps produces extension torque at the elbow and would reduce the size of the flexion twitch. This reduction of the superimposed twitch during elbow flexion contractions would result in an overestimation of voluntary activation of elbow flexor muscles. However, because the response in triceps is small, the effect on the measurement of voluntary activation is also small. The propensity of TMS to activate elbow flexors rather than extensors has been noted previously (19).
The superimposed twitch evoked by motor cortical stimulation decreases linearly with voluntary force over the contraction range studied (26). In the present study, we have confirmed this finding and shown that the relationship is reproducible between sessions. We used this robust relationship to estimate the size of the resting twitch by extrapolation of data collected from 2, 3, or 30 contractions. The average estimated resting twitch calculated from 30 contractions in one testing session was similar to the average calculated from 3 contractions or even 2 contractions if a MVC and 50% MVC were used. For methods 1, 2, and 3, the estimated resting twitch was reproducible between sessions with a low between-session variability (<1.5% MVC). However, the estimated resting twitch calculated with method 4 (MVC and a 75% MVC) was not similar to that calculated with method 1 and had considerably greater within- and between-session variabilities than the other methods. The reproducibility and variability between sessions for methods 1, 2, and 3 based on cortical stimulation were similar to the measures for the resting twitch produced by motor nerve stimulation. This occurred although the twitch produced by motor nerve stimulation was smaller in size because it only reflects torque produced by biceps and brachialis, which contribute about one-half of total elbow flexion torque (2). In contrast, the twitch evoked by motor cortical stimulation reflects the net output from all muscles that generate elbow flexion torque. Furthermore, it is the result of multiple descending volleys that can sometimes activate individual motoneurons more than once (6).
To derive voluntary activation using motor cortical stimulation, we compared the size of the superimposed twitch with the estimated resting twitch (26). As expected, voluntary activation increased linearly with increasing contraction strength for contractions between 50 and 100% of maximum. This occurred for all four methods of estimation of the resting twitch (see Fig. 4). For each method, calculated voluntary activation was high during maximal efforts (>93%) and was reproducible between sessions with lower within-session variability for methods 1, 2, and 3 (<4%) than for method 4 (9%). Incomplete voluntary activation during a maximal effort implies that, at the moment of stimulation, voluntary motor cortical output was not sufficient to recruit all motoneurons or to drive them at a sufficiently fast rate to produce maximal force. Furthermore, extra output from the motor cortex was available, as the motor cortical stimulus was able to evoke extra torque that the subject could not access voluntarily (13, 26). In addition to further localizing the site of failure of voluntary drive, measurement of the EMG responses to motor cortical stimulation enables simultaneous measurement of cortical excitability and inhibition.
The present results have practical implications for the assessment of motor performance with motor cortical stimulation. Of the methods tested, voluntary activation is best derived with estimation of the resting twitch based on a large number of contractions (method 1) for assessment of healthy subjects before and after an intervention or to compare two groups of subjects. For patient groups or when voluntary activation is changing rapidly over time, estimation of the resting twitch based on three contractions (a MVC, 50% MVC, and 75% MVC) will provide an accurate and reliable measure of voluntary activation. If there is not enough time (
15 s) to perform three contractions without disruption of the experimental paradigm, then a good estimate of the resting twitch can be derived from only two contractions if a MVC and 50% MVC are used. Our previous studies showed that the submaximal contraction must be at least 50% MVC (26).
In summary, provided at least two contractions that differ widely in strength are used, estimation of the resting twitch evoked by motor cortical stimulation is reproducible between days with low variability. As a result, voluntary activation measured with cortical stimulation is reliable over time. Because the size of the resting twitch can be estimated accurately with small numbers of contractions, motor cortical stimulation can now be used for measurement of voluntary activation in a range of experimental circumstances when neural drive to the muscles may be changing rapidly.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Smith, J. E. Butler, P. G. Martin, R. A. McBain, and J. L. Taylor Increased ventilation does not impair maximal voluntary contractions of the elbow flexors J Appl Physiol, June 1, 2008; 104(6): 1674 - 1682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Smith, P. G. Martin, S. C. Gandevia, and J. L. Taylor Sustained contraction at very low forces produces prominent supraspinal fatigue in human elbow flexor muscles J Appl Physiol, August 1, 2007; 103(2): 560 - 568. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Todd, J. L. Taylor, J. E. Butler, P. G. Martin, R. B. Gorman, and S. C. Gandevia Use of motor cortex stimulation to measure simultaneously the changes in dynamic muscle properties and voluntary activation in human muscles J Appl Physiol, May 1, 2007; 102(5): 1756 - 1766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Z. Ross, N. Middleton, R. Shave, K. George, and A. Nowicky Human, Environmental & Exercise: Corticomotor excitability contributes to neuromuscular fatigue following marathon running in man Exp Physiol, March 1, 2007; 92(2): 417 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Hunter, J. E. Butler, G. Todd, S. C. Gandevia, and J. L. Taylor Supraspinal fatigue does not explain the sex difference in muscle fatigue of maximal contractions J Appl Physiol, October 1, 2006; 101(4): 1036 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Martin, S. C. Gandevia, and J. L. Taylor Output of Human Motoneuron Pools to Corticospinal Inputs During Voluntary Contractions J Neurophysiol, June 1, 2006; 95(6): 3512 - 3518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Sogaard, S. C. Gandevia, G. Todd, N. T. Petersen, and J. L. Taylor The effect of sustained low-intensity contractions on supraspinal fatigue in human elbow flexor muscles J. Physiol., June 1, 2006; 573(2): 511 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O Prasartwuth, J. L Taylor, and S. C Gandevia Maximal force, voluntary activation and muscle soreness after eccentric damage to human elbow flexor muscles J. Physiol., August 15, 2005; 567(1): 337 - 348. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Todd, J. E Butler, J. L Taylor, and S. C Gandevia Hyperthermia: a failure of the motor cortex and the muscle J. Physiol., March 1, 2005; 563(2): 621 - 631. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, session 1; 
, session 2).
and 
) muscles during 50, 75, and 100% MVCs performed on 2 days (