Mechanomyography (MMG) records the low-frequency lateral oscillations of contracting skeletal muscle fibers (1,22). It has been suggested that these lateral oscillations are due to: a) the gross lateral movement of a muscle at the initiation of contraction, b) smaller subsequent lateral oscillations occurring at the resonant frequency of the muscle, and c) dimensional changes of the active muscle fibers (1).
The time and/or frequency domain of the MMG signal has been used to study various aspects of muscle function including neuromuscular fatigue (26), muscle fiber type distribution patterns (31), and training adaptations (12). Clinically, MMG has been used to study a number of neuromuscular disorders in pediatric, adult, and geriatric populations such as myotonic dystrophy (23), and low-back pain (32).
According to Stokes (28), the potential uses and limitations of MMG must be continually reassessed as knowledge of muscle sounds increases and the development of more appropriate methodology occurs. One promising application of MMG involves use of the amplitude and frequency characteristics to examine differences in motor control strategies (motor unit recruitment and firing rate) in various muscles and under a variety of conditions (22). Specifically, it has been suggested that the amplitude of the MMG signal is related to motor unit recruitment, whereas the frequency domain may provide information about motor unit firing rate (22). Thus, simultaneous examination of the torque-related patterns for MMG amplitude and mean power frequency (MPF) may be useful for identifying muscle-specific motor control strategies during isometric and dynamic muscle actions. Furthermore, there is evidence that the motor control strategies used to modulate force during isometric muscle actions differ from those during dynamic muscle actions (3,5,15). For example, we found that MMG amplitude plateaued at 80% MVC during incremental isometric muscle actions of the biceps brachii (3) and vastus medialis (5), but increased to 100% of peak torque (PT) during concentric isokinetic contractions of the same muscles. These findings suggested that motor unit recruitment continued to 100% PT for the concentric isokinetic, but not the isometric muscle actions. There may also be differences in motor control strategies between muscles for the same type of muscle action (3,5). We found that the MMG MPF increased to 100% PT during concentric isokinetic muscle actions for the vastus medialis (5) but did not change for the biceps brachii (3). These findings suggest that variations in motor unit firing rate may contribute to dynamic force gradation in the vastus medialis but not the biceps brachii.
Previous investigations have indicated that there may be differences between isometric, concentric, and eccentric muscle actions for the motor unit activation strategies that control force production. For example, it has been suggested that the normal pattern of motor unit recruitment may be reversed during eccentric muscle actions (21) and that, although maximal force is greater, muscle activation is lower during eccentric than concentric muscle actions (15). This may reflect an inability to achieve maximal muscle activation during eccentric muscle actions (14). Examination of the time and frequency domains of the MMG signals may contribute to our understanding of the specific pattern of motor unit recruitment and firing rate that modulates eccentric force production (24). Eccentric muscle actions are an important component of the activities of daily living, and MMG may provide a noninvasive method for determining how torque is modulated during these important muscle actions. Therefore, the purpose of this study was to examine the MMG amplitude and MPF versus torque relationships during isometric and eccentric isokinetic muscle actions.
Approach to the problem and experimental design.
MMG is a measure of the mechanical activity (oscillations) of activated muscle. Analysis of the amplitude and frequency characteristics of the MMG signal may contribute to our understanding of the relative contributions of motor unit recruitment and firing rate to torque production during various types of muscle actions. There is evidence that the motor control strategies used to modulate torque during isometric muscle actions differs from that used during concentric and eccentric muscle actions. No previous studies have compared the time and frequency domain responses of the MMG signal during submaximal to maximal isometric and eccentric muscle actions. Subjects in the present study were asked to perform submaximal to maximal isometric and eccentric muscle actions, and the patterns for MMG amplitude and MPF were then compared using polynomial regression. The results were then used to gain insight into the motor control strategies used to modulate torque for the two types of muscle actions.
Eight women and three men (mean age ± SD = 20.8 ± 1.4 yr) volunteered to be subjects for this investigation. All procedures were approved by the University Institutional Review Board for Human Subjects and the subjects signed informed consent before any testing.
Each subject visited the laboratory on three occasions separated by at least 24 h: 1) an orientation session that was used to describe the study protocol and familiarize the subjects with the testing equipment. During the orientation session, the subjects performed maximal as well as submaximal trials of the isometric and isokinetic testing at five randomly selected percentages of their maximal values; 2) an isometric testing session; and 3) an isokinetic testing session. The order of the isometric and isokinetic testing sessions was randomized for each subject. Each session began with a warm-up consisting of 5 min of cycle ergometry at a power output of 50 W.
Eccentric, isokinetic leg extension PT of the dominant limb (based on kicking preference) was measured using a calibrated Cybex 6000 isokinetic dynamometer (CYBEX Division of LUMEX, Inc., Ronkonkoma, NY) at 30°·s−1. The subjects were in a seated position in accordance with the Cybex 6000 testing manual (6000 Testing and Rehabilitation System: User’s Guide, CYBEX Division of LUMEX, Inc.). Two separate maximal eccentric muscle actions were performed, with the highest value selected as PT. All other muscle actions were single effort submaximal trials, and the subjects were instructed to produce a torque which, to the best of their subjective ability, corresponded to a specific percentage of PT, according to the procedures of Weir et al. (30). That is, the subjects were asked to provide, in random order, eccentric isokinetic muscle actions at 10, 20, 30, 40, 50, 60, 70, 80, and 90% of eccentric PT. A submaximal trial was repeated if the actual torque produced was not within ±5% of the desired torque value. The subjects were given verbal feedback to produce the desired amount of torque. A 2-min rest was allowed between muscle actions.
Maximal isometric torque was determined at a leg flexion angle of 0.785 rad (45°) below the horizontal plane. Two separate maximal muscle actions were performed, with the highest value selected as MVC. Single effort, submaximal isometric muscle actions at approximately 10, 20, 30, 40, 50, 60, 70, 80, and 90% of MVC were then performed at the same leg flexion angle (30). The order of the submaximal muscle actions was randomly determined. Each isometric muscle action was held for 6 s. A submaximal trial was repeated if the actual torque produced was not within ±5% of the desired torque value. A 2-min rest was allowed between muscle actions. In addition to verbal feedback, the subjects were able to read their torque production directly off the dynamometer.
The MMG signal was detected by a piezoelectric crystal contact sensor (Hewlett-Packard, 21050A, bandwidth 0.02–2000 Hz). The MMG sensor was placed over the belly of the vastus medialis muscle (Fig. 1). A stabilizing ring, double-sided foam tape, and microporous tape were used to ensure consistent contact pressure of the MMG sensor.
The raw MMG signal was stored on a personal computer and expressed as root mean square amplitude values by software (AcqKnowledge III, Biopac Systems Inc., Santa Barbara, CA). The sampling frequency was 1000 Hz. The MMG signal was bandpass filtered (fourth-order Butterworth filter) at 5–100 Hz. For the isokinetic muscle actions, the MMG amplitude and frequency values were calculated for a 1-s time period that corresponded to a 30° range of motion from approximately 150 to 120° of leg flexion (Fig. 2). This range of motion was selected to avoid the acceleration and deceleration phases that are typical of isokinetic dynamometry (4). For the isometric muscle actions, the middle 2 s of the 6-s muscle action were used for the MMG analyses (Fig. 2). All frequency analyses were performed with custom programs written with LabVIEW software (version 6.1, National Instruments, Austin, TX). Each MMG data segment was processed with a Hamming window and discrete Fourier transformation (DFT). We chose DFT, as opposed to fast Fourier transformations, because the DFT is not constrained to 2N number of data points. Therefore, DFT analyses were performed without having to truncate the data segments or resort to zero padding. Frequency data were expressed as MPF. Previous test-retest reliability from our laboratory for eccentric PT, MMG amplitude, and MMG MPF indicated that for eight male subjects measured 48 h apart, the intraclass correlation coefficients (R) exceeded 0.88, 0.97, and 0.76, respectively, with no significant differences (P > 0.05) between mean values for test versus retest at a variety of angular velocities.
Before the statistical analyses, each subject’s data were normalized to their highest recorded value (percent max) for MMG amplitude and MMG MPF. Polynomial regression analyses (SPSS 11.5 for Windows software package, Chicago, IL) were used to determine the relationships for MMG amplitude and MMG MPF versus torque production for the isometric and isokinetic muscle actions. Post hoc statistical power was 1.000 for isometric MMG amplitude, isometric MMG MPF, and eccentric isokinetic MPF. For eccentric isokinetic MMG amplitude, power was 0.483.
Isometric and isokinetic MMG amplitude and MPF.
Figures 3 and 4 describe the relationships for normalized MMG amplitude and MPF versus normalized torque for the isometric muscle actions. For the isometric muscle actions, the relationships for MMG amplitude (r2 = 0.984) and MPF (r2 = 0.989) versus torque were best fit with linear models. Figures 5 and 6 describe the relationships for normalized MMG amplitude and MPF versus normalized torque for the eccentric isokinetic muscle actions. For the eccentric muscle actions, the relationships for MMG amplitude (r2 = 0.580) and MPF (r2 = 0.961) versus torque were also linear.
Isometric muscle actions.
Typically, the MMG amplitude versus isometric force (or torque) relationship: a) increases linearly to MVC (8), b) increases curvilinearly to MVC (19), or c) increases to approximately 75–80% MVC and then plateaus or decreases to MVC (8,10,18,25). The pattern of the MMG amplitude versus isometric force relationship can be influenced by muscle stiffness, intramuscular fluid pressure, and motor unit activation strategies (10,18,24). Linear or curvilinear increases in MMG amplitude to MVC have been attributed to the recruitment of fast-twitch motor units (8), whereas the plateau and/or decrease in MMG amplitude has been attributed to increased muscle stiffness, decreased compliance, or the fusion of motor unit twitches at high firing frequencies (8).
For the isometric muscle actions in the present study, the MMG amplitude versus isometric torque relationship for the vastus medialis at 45° of leg flexion was linear (r2 = 0.984) to MVC. This pattern was consistent with the findings of Ebersole et al. (8) at 50° of leg flexion and Shinohara et al. (27) at 90° of leg flexion for the vastus medialis, vastus intermedius, and rectus femoris muscles. Matheson et al. (18), however, reported a plateau in MMG amplitude between 80 and 100% MVC for the rectus femoris at a leg flexion angle of 60°. Ebersole et al. (8) demonstrated that the pattern for MMG amplitude versus isometric torque depended on the muscle tested and the joint angle examined. For example, Ebersole et al. (8) found that MMG amplitude increased to MVC for the rectus femoris and vastus medialis muscles at 25 and 50° of leg flexion and for the vastus lateralis at 50° of leg flexion. At 75° of leg flexion, however, the MMG amplitude for each muscle (rectus femoris, vastus lateralis, and vastus medialis) increased to 75% MVC, then plateaued between 75 and 100% MVC. It was hypothesized that the plateau at 75° of leg flexion, but not at 25° and 50° of leg flexion, may have been due to the greater isometric torque production, and thus muscle stiffness, at 75° of leg flexion. Muscle stiffness is related to the number of attached cross bridges (11) and increases with isometric torque. This torque-related increase in stiffness may reflect the length-tension relationship of skeletal muscle, which suggests that there is an optimal length at which there is maximum overlap between the myosin cross-bridges and the actin binding sites, leading to maximum muscle tension (13). At high levels of isometric force production, muscle stiffness may reduce the ability of muscle fibers to oscillate and attenuate the amplitude of the MMG signal (22). Thus, the linear MMG amplitude versus isometric torque relationships to MVC for the vastus medialis in the present study (at 45° of leg flexion) as well as those of Ebersole et al. (8) (at 50° of leg flexion) and Shinohara et al. (27) (at 90° of leg flexion), may indicate that the torque-related increase in muscle stiffness was not sufficient to cause a plateau and/or decrease in MMG amplitude.
It is also possible that the conflicting results from previous studies regarding the patterns for the MMG amplitude versus isometric torque relationship may have been due to differences in the activation strategies that modulate force production at different joint angles (9). Specifically, there may be differences in the contributions of motor unit recruitment and firing rate to force production at varying degrees of leg flexion (29), which may, in turn, influence the pattern of the MMG amplitude versus isometric torque relationship. It has been suggested that high motor unit firing rates above 80% MVC may result in the fusion of motor unit twitches, which would limit muscle fiber oscillations and cause a plateau and/or decrease in MMG amplitude (24). In addition, Orizio et al. (24) reported that during voluntary, incremental step and ramp isometric muscle actions, MMG amplitude increased to the end of motor unit recruitment. Thus, with regard to the present study, it is possible that at a leg flexion angle of 45°: a) the motor unit firing rate was not sufficient to cause fusion of twitches of the activated motor units, and thus MMG amplitude increased to MVC; and/or b) the recruitment of additional motor units continued to MVC (24).
The linear (r2 = 0.989) torque-related increase in MMG MPF found in the present study for the vastus medialis was similar to findings reported by Shinohara et al. (27). Shinohara et al. (27), however, found that the patterns of the MMG median frequency (MF) versus isometric torque relationships differed between the three superficial muscles of the quadriceps femoris (at a leg flexion angle of 90°). MMG amplitude increased then plateaued for rectus femoris, remained constant then increased for vastus lateralis, and continuously increased for vastus medialis. Other studies have also reported nonlinear MMG MPF or MF versus isometric torque patterns for the rectus femoris (7,18) and vastus medialis (5) muscles. Dalton and Stokes (7) found that MMG MPF from the rectus femoris increased quadratically with isometric force, with a plateau above 80–90% MVC (at 90° leg flexion). Matheson et al. (18) dichotomized subjects into “high-force” and “low-force” groups, and reported differences in the patterns of the MMG MPF and MF versus isometric force relationships for the rectus femoris between the two groups of subjects (at 60° leg flexion). For both groups, MMG MPF and MF increased to 80% MVC but decreased in the high force group and increased in the low force group between 80 and 100% MVC (18). Recently, we (5) examined the vastus medialis during incremental isometric muscle actions from 10 to 100% MVC at 45° of leg flexion and found that the MMG MPF versus isometric torque relationship was cubic. There was no change in MMG MPF from 10 to 40% MVC, a rapid increase from 40 to 80% MVC, and a continued increase from 80 to 100% MVC, but at a lesser rate than from 40 to 80%. Like the torque-related differences in patterns for MMG amplitude described by Ebersole et al. (8), differences between studies (5,27) for the MMG MPF versus isometric torque relationships may be due to joint angle and/or muscle-specific differences in motor control strategies (29). During isometric muscle actions, torque is modulated by concurrent increases in motor unit recruitment and firing rate to approximately 50–90% MVC (depending on the muscle involved), and then by increases in firing rate alone to MVC (16). Although it has not been directly verified, it has been suggested that the MPF of the MMG signal may qualitatively reflect the global firing rate of the unfused activated motor units (22). Thus, the linear increase in MMG MPF to MVC in the present study may reflect an increase in the global motor unit firing rate across the entire range of torque levels.
Eccentric muscle actions.
No previous study has examined the MMG amplitude and MPF versus torque relationships during eccentric muscle actions of the quadriceps femoris. Recent studies (6,7,17) have, however, described these relationships for the biceps brachii and first dorsal interosseus. Dalton and Stokes (6,7) found that MMG amplitude increased linearly (r = 0.90), but MPF did not change during eccentric muscle actions of the biceps brachii at force outputs ranging from 0 to 8.5 kg. Madeleine et al. (17), however, found no differences in MMG amplitude or MPF values between 0, 25, 50, 75, and 100% PT during eccentric muscle actions of the first dorsal interosseus muscle. In the present study, MMG amplitude and MPF from the vastus medialis increased linearly to 100% PT. The amplitude of the MMG signal is related to motor unit recruitment, whereas the MPF of the MMG signal may qualitatively reflect the global firing rate of the unfused activated motor units (22). Thus, like the isometric muscle actions, the increases in MMG amplitude and MPF in the present study suggested that eccentric torque was modulated by concurrent increases in motor unit recruitment and firing rate from 10 to 100% PT.
Recent studies (5,15,21) have described differences among isometric, concentric, and eccentric muscle actions (within an individual muscle as well as between different muscles) for various aspects of motor control including motor unit recruitment thresholds, initial firing rates, levels of activation, and the relative contributions of motor unit recruitment and firing rate to force production. It has been suggested (21) that during eccentric muscle actions, the normal order of motor unit recruitment (low to high threshold) is reversed. For example, Nardone et al. (20) found differences in the activity levels of the plantar flexor muscles (medial and lateral gastrocnemius, soleus, peroneus) for eccentric versus concentric muscle actions. Furthermore, differences were found in the populations of active motor units between these two types of muscle actions, and it was concluded that eccentric muscle actions were controlled by high threshold motor units (20). Kossev and Christova (15), however, found that decreasing force during eccentric muscle actions was accomplished through a reduction in the firing rate of activated motor units, rather than decruitment. Conversely, Kossev and Christova (15) reported that, unlike the typical pattern for isometric muscle actions (16), during concentric muscle actions of the biceps brachii, gradations in muscle force were influenced more by motor unit recruitment than alterations in firing rate.
We have recently compared the torque-related patterns for MMG amplitude and MPF during isometric versus concentric muscle actions of the biceps brachii (3) and vastus medialis (5) muscles. For incremental isometric muscle actions, MMG amplitude increased to 80% MVC, and then plateaued (3) or decreased (5) to MVC. It was hypothesized (3,5) that the force-related increase in MMG amplitude during submaximal isometric muscle actions (below approximately 80% MVC) reflected increases in motor unit recruitment and firing rates (2,24), whereas the plateau (or decrease) was due to high levels of muscle stiffness or fusion of motor units twitches resulting from high firing rates (8). For the concentric isokinetic muscle actions, MMG amplitude increased linearly to 100% PT for both the biceps brachii (3) and vastus medialis (5) muscles, whereas the MPF did not change for the biceps brachii (3) but increased linearly for the vastus medialis (5). These findings suggested that torque production during concentric muscle actions of the biceps brachii may be modulated by recruitment alone, whereas the vastus medialis modulates torque production through a combination of recruitment and increases in firing rate to 100% PT. The results of the present study, in conjunction with previous studies (3,5), suggest that there may be muscle-specific motor unit activation strategies during both isometric and dynamic muscle actions. Furthermore, for the vastus medialis, the linear increases in MMG amplitude and MPF suggested that motor unit recruitment and firing rate increased to maximal voluntary torque production during isometric as well as concentric and eccentric isokinetic muscle actions (5).
1. Barry, D. T., and N. M. Cole. Fluid mechanics of muscle vibrations. Biophys. J.
2. Barry, D. T., and N. M. Cole. Muscle sounds are emitted at the resonant frequencies of skeletal muscle. IEEE Trans. Biomed. Eng.
3. Beck, T. W., T. J. Housh, G. O. Johnson, et al. Mechanomyographic amplitude and mean power frequency versus torque relationships during submaximal to maximal isokinetic and isometric muscle actions of the biceps brachii. J. Electromyogr. Kinesiol.
4. Brown, L. E., M. Whitehurst, R. Gilbert, and D. N. Buchalter. The effect of velocity and gender on load range during knee extension and flexion exercise on an isokinetic device. J. Orthop. Sports Phys. Ther.
5. Coburn, J. W., T. J. Housh, J. T. Cramer, et al. Mechanomyographic time and frequency domain responses of the vastus medialis muscle during submaximal to maximal isometric and isokinetic muscle actions. Electromyogr. Clin. Neurophysiol.
6. Dalton, P. A., and M. J. Stokes. Acoustic myography reflects force changes during dynamic concentric and eccentric contractions of the human biceps brachii muscle. Eur. J. Appl. Physiol. Occup. Physiol.
7. Dalton, P. A., and M. J. Stokes. Frequency of acoustic myography during isometric contraction of fresh and fatigued muscle and during dynamic contractions. Muscle Nerve
8. Ebersole, K. T., T. J. Housh, G. O. Johnson, T. K. Evetovich, D. B. Smith, and S. R. Perry. MMG and EMG responses of the superficial quadriceps femoris muscles. J. Electromyogr. Kinesiol.
9. Eloranta, V. Coordination of the thigh muscles in static leg extension. Electromyogr. Clin. Neurophysiol.
10. Esposito, F., D. Malgrati, A. Veicsteinas, and C. Orizio. Time, and frequency domain analysis of electromyogram and sound myogram in the elderly. Eur. J. Appl. Physiol. Occup. Physiol.
11. Ettema, G. J., and P. A. Huijing. Skeletal muscle stiffness in static and dynamic contractions. J. Biomech.
12. Evetovich, T. K., T. J. Housh, D. J. Housh, G. O. Johnson, D. B. Smith, and K. T. Ebersole. The effect of concentric isokinetic strength training of the quadriceps femoris on electromyography and muscle strength in the trained and untrained limb. J Strength Cond. Res.
13. Gordon, A. M., A. F. Huxley, and F. J. Julian. The variation in isometric tension with sarcomere length in vertebrate muscle fibres. J. Physiol.
14. Kellis, E., and V. Baltzopoulos. Muscle activation differences between eccentric and concentric isokinetic exercise. Med. Sci. Sports Exerc.
15. Kossev, A., and P. Christova. Discharge pattern of human motor units during dynamic concentric and eccentric contractions. Electroencephalogr Clin Neurophysiol.
16. Lawrence, J. H., and C. J. de Luca. Myoelectric signal versus force relationship in different human muscles. J. Appl. Physiol.
17. Madeleine, P., P. Bajaj, K. Sogaard, and L. Arendt-Nielsen. Mechanomyography and electromyography force relationships during concentric, isometric and eccentric contractions. J. Electromyogr. Kinesiol.
18. Matheson, G. O., L. Maffey-Ward, M. Mooney, K. Ladly, T. Fung, and Y. T. Zhang. Vibromyography as a quantitative measure of muscle force production. Scand. J. Rehabil. Med.
19. Maton, B., M. Petitjean, and J. C. Cnockaert. Phonomyogram, and electromyogram relationships with isometric force reinvestigated in man. Eur. J. Appl. Physiol. Occup. Physiol.
20. Nardone, A., C. Romano, and M. Schieppati. Selective recruitment of high-threshold human motor units during voluntary isotonic lengthening of active muscles. J. Physiol.
21. Nardone, A., and M. Schieppati. Shift of activity from slow to fast muscle during voluntary lengthening contractions of the triceps surae muscles in humans. J. Physiol.
22. Orizio, C. Muscle sound: bases for the introduction of a mechanomyographic signal in muscle studies. Crit. Rev. Biomed. Eng.
23. Orizio, C., F. Esposito, V. Sansone, G. Parrinello, G. Meola, and A. Veicsteinas. Muscle surface mechanical and electrical activities in myotonic dystrophy. Electromyogr. Clin. Neurophysiol.
24. Orizio, C., M. Gobbo, B. Diemont, F. Esposito, and A. Veicsteinas. The surface mechanomyogram as a tool to describe the influence of fatigue on biceps brachii motor unit activation strategy: historical basis and novel evidence. Eur. J. Appl. Physiol.
25. Orizio, C., R. Perini, and A. Veicsteinas. Muscular, sound and force relationship during isometric contraction in man. Eur. J. Appl. Physiol. Occup. Physiol.
26. Perry-Rana, S. R., T. J. Housh, G. O. Johnson, A. J. Bull, and J. T. Cramer. MMG and EMG responses during 25 maximal, eccentric, isokinetic muscle actions. Med. Sci. Sports Exerc.
27. Shinohara, M., M. Kouzaki, T. Yoshihisa, and T. Fukunaga. Mechanomyogram from the different heads of the quadriceps muscle during incremental knee extension. Eur. J. Appl. Physiol. Occup. Physiol.
28. Stokes, M. J. Acoustic myography: applications and considerations in measuring muscle performance. Isokin. Exerc. Sci.
29. Suter, E., and W. Herzog. Extent of muscle inhibition as a function of knee angle. J. Electromyogr. Kinesiol.
30. Weir, J. P., L. L. Wagner, and T. J. Housh. Linearity and reliability of the IEMG v torque relationship for the forearm flexors and leg extensors. Am. J. Phys. Med. Rehabil.
31. Yoshitake, Y., and T. Moritani. The muscle sound properties of different muscle fiber types during voluntary and electrically induced contractions. J. Electromyogr. Kinesiol.
32. Yoshitake, Y., H. Ue, M. Miyazaki, and T. Moritani. Assessment of lower-back muscle fatigue using electromyography, mechanomyography, and near-infrared spectroscopy. Eur. J. Appl. Physiol.
Keywords:©2004The American College of Sports Medicine
MMG; PHONOMYOGRAPHY; ACOUSTIC MYOGRAPHY; ISOKINETIC; STATIC; QUADRICEPS; MOTOR CONTROL