Elbow Flexor Inhibition as a Function of Muscle Length

Transcription

1 46 JOURNAL OF APPLIED BIOMECHANICS, 2002, 18, by Human Kinetics Publishers, Inc. Elbow Flexor Inhibition as a Function of Muscle Length Luciana Brondino, Esther Suter, Hae-Dong Lee, and Walter Herzog The University of Calgary Muscle inhibition (MI) in human knee extensors increases with increasing maximal voluntary force as a function of knee angle. It was speculated that this angle-dependent MI was modulated by force-dependent feedback, likely Golgi tendon organ pathways. Such angle-dependent MI is of clinical and theoretical importance. The purpose of this study was to determine MI in human elbow flexors for maximal voluntary contractions. Muscle inhibition, elbow flexor force, and electromyographic (EMG) activity were measured in 31 volunteers at elbow angles between 30º and 120º. MI and elbow flexor EMG were the same at all elbow angles. Maximal isometric forces were greatest at the 70º angle, and never fell below 70% of the peak force at any of the tested angles. From these results it is concluded that force-dependent modulation of MI did not occur in the elbow flexors, possibly because maximal isometric force remained relatively close (within 30%) to the peak force. In contrast, force-dependent modulation of MI occurred in the knee extensors at the most extended angles, when the average knee extensor force had dropped to 50% or less of the maximal knee extensor force. It is likely that human maximal voluntary contractions are not associated with a given activation. Rather, activation appears to be modulated by force-dependent feedback at force levels below 70% of the absolute peak force, which manifests itself in a change of MI that parallels the level of maximal isometric force in voluntary contractions. Key Words: muscle activation, elbow flexor forces, superimposed twitch technique Introduction The inability to fully activate a muscle during maximal-effort voluntary contraction, referred to as muscle inhibition, is particularly prevalent in patients with joint pain, injury, or degenerative joint disease, such as osteoarthritis (Hurley, Jones, & Newham, 1994; Rutherford, Jones, & Newham, 1986; Suter & Herzog, 2000). The authors are with the Faculty of Kinesiology, The University of Calgary, 2500 University Dr. NW, Calgary, AB Canada T2N 1N4. 46

2 Elbow Flexor Inhibition 47 However, even healthy persons are often unable to fully activate specific muscle groups such as the elbow flexors (Dowling, Konert, Ljucovic, & Andrews, 1994), knee extensors (Suter & Herzog, 1997), or ankle plantar flexors (Belanger & McComas, 1981). Muscle inhibition is typically evaluated using the twitch interpolation technique (Allen, McKenzie, & Gandevia, 1998; Belanger & McComas, 1981; Merton, 1954). This technique is based on the assumption that during a truly maximal contraction, all motor units of a muscle are activated at high frequencies, and that an electrical twitch to the muscle or its motor nerve will not produce any additional force. However, for submaximal motor unit activation, the superimposed stimulation will evoke an additional force, the magnitude of which increases with decreasing levels of motor unit activation (Allen et al., 1998; Belanger & McComas, 1981; Dowling et al., 1994). One limitation of the superimposed twitch technique is its variability during repeat measurements (Allen, Gandevia, & McKenzie, 1995), across muscles (Belanger & McComas, 1981), and across muscle lengths (Suter & Herzog, 1997). We have speculated that the variability during repeat measurements is associated with the stochastic nature of the electrical twitch application relative to the pulses of the voluntary excitation trains (Suter & Herzog, 2001). By definition, this variability is random and probably cannot be eliminated. However, the variability across muscles appears to be systematic, but has eluded convincing explanation. In a study on human knee extensors, we observed that muscle inhibition increased with increasing maximal force at different knee angles (Suter & Herzog, 1997). From these results we speculated that negative feedback from the Golgi tendon organs would cause muscle inhibition in a force-dependent manner. If this speculation was correct, and constituted a general mechanism, the same relationship between muscle inhibition and isometric force (as a function of muscle length) would have to be observed in other muscle groups. There is limited data from other muscle groups regarding the relation between muscle inhibition and muscle length. Cresswell, Löscher, and Thorstensson (1995) measured inhibition in the plantar flexors and found that muscle inhibition was independent of joint angle. Gandevia and McKenzie (1988) also did not observe any differences in muscle activation in three different muscle groups; however, muscle inhibition was only measured at two muscle lengths and in a limited number of participants. In order to further elucidate the question of length dependency of muscle activation, we undertook the present study. The study was aimed at determining inhibition in the human elbow flexor muscles at different muscle lengths, and thus, presumably, different isometric forces (Kulig, Andrews, & Hay, 1984). Specifically, we tested the hypothesis that increasing isometric muscle forces were associated with increasing muscle inhibition, thereby lending support to the idea that Golgi tendon pathways were partly responsible for the observed length dependence of muscle inhibition. Methods Elbow flexor forces, electromyographic (EMG) activity of biceps and triceps brachii, and superimposed twitch forces were measured in a convenience sample of healthy persons.

3 48 Participants were 31 healthy individuals (14 M, 17 F; age 28.7 ± 5.0 yrs; height 1.72 ± 0.12 m; body mass 69.3 ± 11.9 kg) with no upper extremity, shoulder, or neck problems. They gave free informed consent to participate in this study. The sample consisted of moderately active people recruited primarily from the Faculty of Kinesiology. The conjoint Ethics Committee at the University of Calgary approved all methods and procedures. Strength measurements: Isometric elbow flexor strength was measured using a custom-built upper extremity device consisting of two aluminum plates with a movable axis at the elbow joint. Participants were seated in a dynamometer chair that allowed the shoulder and upper body to be properly restrained. The left arm was used for testing. The upper arm was placed in an adjustable and slightly concave armrest fixed at 30º of shoulder flexion. The lower arm was held in a supinated position and the wrist was fixed and fastened with velcro straps. For each person, the strap was maintained at the same position for all elbow angles tested in order to keep the moment arm constant. Elbow flexor forces were measured using a strain gauge transducer in a full-wheatstone bridge configuration on an S-shaped element at the distal part of the lower arm. The force transducer was attached to the base-plate and could be moved along the long axis of the lower arm to accommodate different arm lengths. The attachment occurred through two hinge joints, which allowed perpendicular alignment of the transducer with respect to the lower arm, independent of lower arm orientation (i.e., elbow angle). EMG measurements: Bipolar surface EMG (Biovision) was obtained from biceps and triceps brachialis. Surface electrodes (Ag-AgCl) were attached to the mid-belly of both muscles along the estimated fiber direction, after the underlying skin had been shaved and cleansed with alcohol. Interelectrode distance was 20 mm. Spot-checking of skin impedance gave values of less than 10 k. EMG signals were passed through a preamplifier located 10 cm from the electrodes, and a band-pass filter with cutoff frequencies of 10 Hz and 1 khz was used. Biceps torques and EMG signals were sampled at 2000 Hz using an analogue-to-digital board with a resolution of 12 bits, and stored on an IPC 486DX for further signal analysis. Superimposed twitch force: Electrical twitches were applied to the motor point of the biceps brachii using an Ag-AgCl electrode (Allen et al., 1998). A carbon-impregnated rubber electrode was used as anode and placed in the frontal part of the upper arm, proximal to the elbow crease. The motor point was identified using a spot-stimulating electrode. Stimulation parameters were controlled using a Grass S88 stimulator in combination with an isolation unit approved for human use (Quincy, MA). Doublet twitches (Burke, 1975) of 0.8-ms duration and 8-ms interpulse intervals were given. The stimulation voltage was individually determined such that optimal excitation was produced. This was achieved by increasing the voltage systematically until the evoked twitch force reached maximal size. This typically happened at values of 140 to 180 V. Protocol: After a brief training and warm-up session, the appropriate stimulation voltage for the twitch interpolation technique was established, as described above. Then the participants performed isometric elbow flexor contractions at five elbow angles (30º, 45º, 70º, 95º, and 120º from full extension) and five contraction levels at each elbow angle (i.e., maximal voluntary contraction [MVC], and 20%,

4 Elbow Flexor Inhibition 49 40%, 60%, and 80% of MVC). Two trials were conducted for each angle and contraction level. Elbow angles and contraction levels were presented in a random manner, except for the MVC contractions, which were always performed first so that the submaximal levels could be determined. Force feedback was provided using an oscilloscope. For submaximal contractions, participants were required to match a target line displayed on the oscilloscope. The participants were asked to build up to the target force in about 1 2 seconds, and to hold the target force steady for another 3 s. The superimposed twitch was given approximately 2 s after the target force was reached. Immediately following the MVC contractions, three superimposed twitches were given to the relaxed muscle. The evoked force represents the torque produced by the relaxed biceps muscle when stimulated by a double twitch, and is referred to as resting twitch force (RTT). The highest evoked force at each angle was taken and used for normalization of the superimposed twitch force. Data analysis: Muscle inhibition was calculated as the percent ratio between the superimposed twitch force and the RTT. Root mean square (RMS) values of the EMG signals were calculated for the 0.5-s period preceding the superimposed twitch. Force-angle diagrams were compared across participants by normalizing the elbow flexor forces to the peak value for each person. Similarly, EMG-angle diagrams were evaluated across participants by normalizing the RMS values of the EMG to the peak value at a given elbow angle. Standard polynomial regression analysis was used to approximate muscle inhibition values as a function of elbow flexor force. Mean and standard deviations, median values of elbow flexor forces, and EMG RMS values were calculated as a function of elbow angle. Repeatedmeasures ANOVA was used to compare muscle inhibition, elbow flexor forces, and EMG RMS values across elbow angles. The level of significance was chosen a priori as = Results Muscle inhibition as a function of the level of activation showed a nonlinear behavior at all five elbow angles tested (Figures 1a 1e). The amount of muscle inhibition for a given level of contraction varied widely, particularly for the low activation levels. Second-order polynomial regression appeared to capture the nonlinearity best, and it was found that 52 74% of the variation in muscle inhibition was explained by the variation in muscle force. The nonlinear shape was similar for all elbow angles tested, with the exception of the most extended elbow angle of 30º, which tended to have higher muscle inhibition at the lower activation levels compared to the more flexed elbow angles. For the MVCs, muscle inhibition was the same across all angles tested, with median values ranging from 2.8 to 6.9% (Figure 2). Elbow flexor forces as a function of elbow angle (shown as the normalized median values in Figure 3) showed the typical inverted U-shape described in other studies (Kulig et al., 1984; Leedham & Dowling, 1995), with the peak forces occurring at 70º of elbow flexion. The normalized RMS values of the biceps EMG were similar for all elbow angles tested (i.e., 75 83%) and were not related in any obvious way to elbow angle (Figure 4). The corresponding values for the antagonist triceps EMG were generally small and did not vary systematically across elbow angles.

5 50 a) 30 degrees % Muscle Inhibition % MVC b) 45 degrees % Muscle Inhibition % MVC Figure 1 Muscle inhibition (in %) as a function of the contraction level for elbow angles of: (a) 30º, (b) 45º, (c) 70º, (d) 95º, and (e) 120º of flexion for all 31 participants combined. As the level of contraction increases to maximal voluntary efforts (MVC), muscle inhibition decreases in nonlinear fashion. Note that the variability in muscle inhibition is much greater for weak contractions than for strong ones, i.e., close to MVC. A second-order polynomial regression has been fitted to the data points to describe the nonlinear trend. (Continued on next page)

6 Elbow Flexor Inhibition 51 c) 70 degrees % Muscle Inhibition % MVC d) 95 degrees % Muscle Inhibition % MVC e) 120 degrees % Muscle Inhibition % MVC Figure 1 (Cont.) Muscle inhibition (in %) as a function of the contraction level for elbow angles.

7 52 Figure 2 Muscle inhibition (median values ± SD) as a function of elbow angle. 0º indicates full extension. Muscle inhibition was observed at all elbow angles, but the amount of muscle inhibition was independent of elbow angle. Force (%) Muscle Inhibition (%) Figure 3 Normalized elbow flexor forces (median values ± SD) as a function of elbow angle. 0 indicates full extension. For each participant, forces were normalized relative to the individual peak force achieved at any joint angle. Peak forces were found to depend on elbow angle, reaching an absolute maximum at an elbow angle of 70º. The relationship was similar to the elbow-flexor vs. force-joint angle relationships reported in the literature (e.g., Kulig et al., 1984).

8 Elbow Flexor Inhibition 53 Normalized Bicep EMG RMS (%) Figure 4 Normalized EMG RMS (median values ± SD) for biceps activation as a function of elbow angle. For each participant, EMG RMS values were normalized relative to the individual peak EMG RMS values achieved at any joint angle. EMG values were independent of elbow angle, indicating that the elbow flexor force vs. joint angle relationship was primarily determined by the anatomy of the joint and the mechanical properties of the elbow flexors. Discussion The interpolated twitch technique has been used to evaluate muscle inhibition in dynamic and static force tasks (Guang, Ranganathan, Sieminow, Liu, & Saghal, 2000), between different muscle groups (Belanger & McComas, 1981), in people with pain or injury (Hurley et al., 1994; Rutherford et al., 1986; Suter & Herzog, 2000), and in older adults (DeSerres & Enoka, 1998). Although the twitch interpolation technique has found wide application, angle dependency of muscle inhibition measurements has not been considered in a systematic way. A previous study on the knee extensor muscles revealed that muscle inhibition changes as a function of knee angle, and therefore as a function of muscle length and force (Suter & Herzog, 1997). This finding has clinical importance if results from different populations tested in different laboratories are compared. There are limited data on the angle/force dependency of muscle inhibition in muscle groups other than the knee extensors (Cresswell et al., 1995; Gandevia & McKenzie, 1988). Cresswell et al. (1995) and Gandevia and McKenzie (1988) found that muscle inhibition and muscle forces were constant across joint angles in plantar flexors as well as the abductor digiti minimi, elbow flexors, and tibialis anterior, respectively. These results are in stark contrast to the changing muscle inhibition found in the knee extensors for varying knee angles and forces (Suter & Herzog, 1997). The purpose of the present study was to assess muscle inhibition as a function of the elbow angle in a large sample of healthy individuals.

9 54 In contrast to muscle inhibition in the knee extensors, muscle inhibition in the elbow flexors was constant across the range of elbow angles tested (Figure 2). This confirms the results of Gandevia and McKenzie (1988), who found no difference in muscle inhibition measured at a short and a resting length of the elbow flexors. In contrast to our study, the 5 participants tested by Gandevia and McKenzie were able to maximally activate their elbow flexors at both elbow angles. We tested a much larger sample (N = 31), and although some were able to maximally recruit their elbow flexors, on average a small inhibition was present at any of the measured muscle lengths. Similar to the findings of Leedham and Dowling (1995), elbow flexor EMG was constant across elbow angles, and antagonist triceps activity was generally small, as found by Allen et al. (1995). For the knee extensors, the amount of muscle inhibition observed as a function of knee angle paralleled the isometric force at the different knee angles. Therefore it was speculated that muscle inhibition was influenced by muscle force rather than muscle length. It is well known that signals from Golgi tendon organs increase with increasing force (e.g., Houk & Henneman, 1967). These signals are thought to have an inhibitory effect during isometric force production (Houk & Henneman, 1967). Thus it had been speculated that the angle dependence of muscle inhibition in the knee extensors was, at least in part, caused by afferent signals from the force-sensing Golgi tendon organs. This speculation is supported by the observation that activation of the knee extensors was elevated at short muscle lengths, i.e., near full extension (Hasler, Denoth, Stacoff, & Herzog, 1994), where force production is small compared to long muscle lengths because of the forcelength relationship. For the elbow flexors, muscle inhibition and activation were constant across elbow angles. This result might be explained by the fact that elbow flexor forces were similar across joint angles. The smallest average force was achieved at the most extended elbow angle tested (30º) and was greater than 70% of the absolute peak isometric force measured at the 70º elbow angle. In contrast, in our previous study (Suter & Herzog, 1997), knee extensor forces were less than 50% of the maximal knee extensor force for the most extended knee angles. Therefore, it appears that muscle inhibition remained constant for the elbow flexors because the isometric forces remained relatively high across all angles. This results would be in agreement with findings from the plantar flexors (Cresswell et al., 1995), tibialis anterior, digitalis minimi, and elbow flexors (Gandevia & McKenzie, 1988), where the forces across the range of muscle length never dropped below 60% of the maximal isometric muscle force. Reanalyzing our knee extensor data, we found that muscle inhibition did not differ across knee angles if only knee angles were considered at which the average isometric force was 70% or greater than the peak isometric force. We conclude from these results that muscle inhibition as a function of joint angle is modified by a force-dependent mechanism, likely the Golgi tendon pathway. When the isometric forces are small at a given joint configuration, presumably because of the force-length relationship (e.g., Gordon, Huxley, & Julian, 1966), muscle activation (EMG) is increased and, correspondingly, muscle inhibition is decreased. However, this mechanism does not come into play for forces that are within approximately 30% of the peak isometric forces achieved at the optimal joint angle. When forces become very low, less than 50% of the peak isometric

10 Elbow Flexor Inhibition 55 force, the nervous system appears to compensate for this mechanical disadvantage by upregulating how much the muscle can be activated during maximal-effort contractions and thereby decreasing muscle inhibition. Thus it appears that muscle inhibition is a natural phenomenon during maximal voluntary contraction. Muscle inhibition occurs because the maximal voluntary -motoneuron drive is in competition with inhibiting pathways (presumably by force-dependent feedback systems) that limit activation. When muscle force production is severely restricted because of a mechanical disadvantage, such as being at an unfavorable muscle length, maximal force is low and the force-dependent feedback is low as well, which allows for a greater activation that, in turn, reduces muscle inhibition. We observed this phenomenon in the knee extensors, where forces become low relative to the absolute maximal isometric knee extensor forces, toward extended knee angles. We did not observe this phenomenon for the elbow flexors, presumably because elbow flexor forces remained, on average, within 30% of the absolute peak force achieved at the optimal elbow angle. Although it seems reasonable to assume that muscle activation is modulated via Golgi tendon pathways, factors that may also contribute to the differences in muscle activation between different muscle groups must be kept in mind. First of all, the quadriceps are an extensor group, while the biceps brachii are flexors. There are differences in innervation and joint afferent signals between flexors and extensors, which may be reflected by a different sensitivity of muscle activation to joint angle (Nichols, Cope, & Abelew, 1999). The same might be true for upper and lower extremity muscle groups, which may display different muscle activation patterns as a result of differences in use, innervation, and presynaptic inhibition (Rossi-Durand, Jones, Adams, & Bawa, 1999). Also, different motor unit recruitment strategies may contribute to the phenomenon observed. For example, it has been found that in small muscles of the hand and foot, such as the adductor pollicis and the 1st dorsal interosseus, all motor units are recruited below 50% of maximal voluntary contraction. In contrast, in large muscles of the limbs, such as the biceps brachii, motor units are recruited at least up to 80% or 85% maximal voluntary contraction, and possibly throughout the full range of voluntary force (Basmajan & De Luca, 1985). However, there is no evidence that motor unit activation patterns are different between the knee extensors and the biceps brachii. Another point to consider is that the knee extensors are by far the strongest muscle group of those discussed here, and the modulation in muscle inhibition may be related to the absolute level of force. References Allen, G.M., Gandevia, C.S., & McKenzie, D.K. (1995). Reliability of measurements of muscle strength and voluntary activation using twitch interpolation. Muscle & Nerve, 18, Allen, G.M., McKenzie, D.K., & Gandevia, S.C. (1998). Twitch interpolation of the elbow flexor muscles at high forces. Muscle & Nerve, 21, Basmajan, J.V., & De Luca, C.J. (1985). Muscles alive. Their functions revealed by electromyography (5th ed.). Baltimore: Williams & Wilkins. Belanger, A.Y., & McComas, A.J. (1981). Extent of motor unit activation during effort. Journal of Applied Physiology: Respiratory, Environmental & Exercise Physiology, 51,

11 56 Burke, R.E. (1975). Motor unit properties and selective involvement in movement. Exercise and Sport Sciences Reviews, 3, Cresswell, A.G., Löscher, W.N., & Thorstensson, A. (1995). Influence of gastrocnemius muscle length on triceps surae torque development and electromyographic activity in man. Experimental Brain Research, 105, DeSerres, S.J., & Enoka, R.M. (1998). Older adults can maximally activate the biceps brachii muscle by voluntary command. Journal of Applied Physiology, 84, Dowling, J.J., Konert, E., Ljucovic, P., & Andrews, D.M. (1994). Are humans able to voluntarily elicit maximum muscle force? Neuroscience Letters, 179, Gandevia, B.C., & McKenzie, D.K. (1988). Activation of human muscles at short muscle lengths during maximal static efforts. Journal of Physiology, 407, Gordon, A.M., Huxley, A.F., & Julian, F.J. (1966). The variation in isometric tension with sarcomere length in vertebrate muscle fibres. Journal of Physiology, 184, Guang, H.Y., Ranganathan, V.K., Sieminow, V., Liu, J.Z., & Saghal, V. (2000). Evidence of inability to fully activate human limb muscle. Muscle & Nerve, 23, Hasler, E.M., Denoth, J., Stacoff, A., & Herzog, W. (1994). Influence of hip and knee joint angles on excitation of knee extensor muscles. Electromyography and Clinical Neurophysiology, 34, Houk, J.C., & Henneman, E. (1967). Responses of Golgi tendon organs to active contraction of the soleus muscle of the cat. Journal of Neurophysiology, 30, Hurley, M.V., Jones, D.W., & Newham, D.J. (1994). Arthrogenic quadriceps inhibition and rehabilitation of patients with extensive traumatic knee injuries. Clinical Science, 86, Kulig, K., Andrews, J.G., & Hay, J.G. (1984). Human strength curves. Exercise and Sport Sciences Reviews, 12, Leedham, J.S., & Dowling, J.J. (1995). Force-length, torque-angle and EMG-joint angle relationships of the human in vivo biceps brachii. European Journal of Applied Physiology, 70, Merton, P.A. (1954). Voluntary strength and fatigue. Journal of Physiology, 123, Nichols, T.R., Cope, T.C., & Abelew, T.A. (1999). Rapid spinal mechanisms of motor coordination. Exercise and Sport Sciences Reviews, Vol. 27?, Rossi-Durand, C., Jones, K.E., Adams, S., & Bawa, P. (1999). Comparison of the depression of H-reflexes following previous activation in upper and lower limb muscles in human subjects. Experimental Brain Research, 126, Rutherford, O.M., Jones, D.A., & Newham, D.J. (1986). Clinical and experimental application of the percutaneous twitch superimposition technique for the study of human muscle activation. Journal of Neurology, Neurosurgery, and Psychiatry, 49, Suter, E., & Herzog, W. (1997). Extent of muscle inhibition as a function of knee angle. Journal of Electromyography and Kinesiology, 7, Suter, E., & Herzog, W. (2000). Does muscle inhibition following knee injury increase risk of osteoarthritis? Exercise and Sport Sciences Reviews, 28, Suter, E., & Herzog, W. (2001). Effect of number of stimuli and timing of twitch application on variability in interpolated twitch torque. Journal of Applied Physiology, 90,