Impact of Contraction Intensity and Velocity on Vastus Lateralis SEMG Power Spectrum and Amplitude

Transcription

1 Impact of Contraction Intensity and Velocity on SEMG Power Spectrum and Amplitude J.P. Miller 1 R.V. Croce 1 W.J. Smith 2 1 Motor Control and Biomechanics Laboratory, Department of Kinesiology, University of New Hampshire, Durham, NH, USA 2 Department of Electrical Engineering, University of New Hampshire, Durham, NH, USA Corresponding Author: John Miller, PhD Tel. (603) Fax. (603) jm@unh.edu KEY WORDS: EMG,, Force, and Velocity ABSTRACT This study investigated the effect of contraction intensity (100%-, 75%-, 50%-, and 25%- maximum voluntary contraction [MVC]) and movement velocity (00 [isometric], 1000, 2000, 3000, and 4000 s-1) on root mean square amplitude (SEMG-RMS) and median frequency power spectrum (SEMG-MNF) of the surface electromyography of vastus lateralis. Ten healthy female university students (age = 23.5 ± 2.7 years, mean height = ± 6.7 cm, mean weight = 66.2 ± 6.3 kg) with no known knee pathologies participated in the investigation. Peak torque (PT), average torque (AVT), SEMG-MNF, and SEMG-RMS were analyzed using separate repeated measures analysis of variance. Results, significant at p < 0.05 or better, were: (1) an inverse relationship between PT and AVT and movement velocity, which was consistent with the literature and the force-velocity curve; (2) greater SEMG-MNF values during all isovelocity conditions compared to isometric, with highest values occurring at 500 s-1 and during 100%- and 75%- MVC; and, (3) an interactive effect of contraction intensity and velocity on muscle SEMG-RMS such that at all contraction intensities highest values were found at 2000 s-1, with a penchant for higher SEMG-RMS values during dynamic movements compared to isometric. It would appear that dynamic muscular contractions facilitated a greater recruitment of fast-twitch fibers -- as determined by increased SEMG-MNF values -- compared to that found during isometric contractions, and that this difference was intensified at slow isokinetic velocities (50 s-1), whereas greater overall muscle activation was found at 2000 s-1. INTRODUCTION Surface electromyography (SEMG) is a noninvasive procedure available to researchers for identifying strategies used by the central nervous system (CNS) to control 80 Vol.12, No. 2, 2012 The Journal of Applied Research.

2 volitional movements (Wakeling, 2009). The SEMG waveform is typically resolved into its corresponding amplitude (SEMG- AMP) and frequencies (SEMG-PSD). Signal amplitude (SEMG-AMP) is frequently used as an indirect measure of overall muscle activity and is usually reported as some form of integrated (SEMG-INT) or root mean square (SEMG-RMS) of the SEMG signal. The SEMG frequency or power spectrum density (SEMG-PSD) is typically measured as either the median (SEMG-MNF) or mean frequency (SEMG-MN), and is associated with the conduction velocity of the respective motor units (Kupa et al., 1995). Purportedly, a shift in the power spectrum towards higher frequencies is indicative of an increase in the average conduction velocity of active muscle fibers, indicating recruitment of larger-sized motor units, whilst a shift toward lower frequencies is indicative of a decrease in the average conduction velocity and de-recruitment (Wakeling, 2009). Although still unsettled, there is reasonable evidence to suggest faster fiber-types (fasttwitch [FT] motor units) generate higher SEMG frequencies and slower fiber-types (slow-twitch [ST] motor units) generate lower SEMG frequencies (von Tscharner & Nigg, 2008). Presently, the interactive effect of force and velocity on SEMG parameters is both inconsistent and equivocal. For example, previous researchers have reported that as contraction velocity increases, SEMG-AMP can either increase (plateauing at about 3000 s-1) (Coburn et al., 2005) or decrease (Croce & Miller, 2006); and, as muscle force increases, SEMG-PSD can either increase (Karlsson & Gerdle, 2001), decrease (Komi & Viitsalo, 1976), does not show any particular relationship against force (Masuda et al., 2001), or is muscle fiber-type dependent (Bilodeau et al., 2003). Overall, documented force- and velocity-specific alterations in SEMG profiles of the QF muscle have been parsimonious and relatively inconsistent, with no research directed toward a comprehensive analysis across a variety of forces and velocities. Accordingly, this research investigated the effect of contraction intensity (100%-, 75%-, 50%-, and 25%-maximum voluntary contraction [MVC]) and movement velocity (00 [isometric], 500, 1000, 2000, and 4000 s-1) on SEMG-MNF and SEMG-RMS of vastus lateralis muscle. As gender differences have been shown to exist for torque and SEMG profiles (Pincivero et al., 2004), this study investigated female participants only. METHODS Subjects Eight healthy female university students (age, ± standard deviation = 23.1 ± 2.6 years, mean height = ± 6.8 cm, mean weight = 64.6 ± 6.1 kg) with no known knee pathologies participated in the investigation. Subjects were verbally informed of the procedures and potential risks, and they read and signed an informed consent prior to participation. The Institutional Review Board of the University of New Hampshire approved this study. Dynamometer Set Up Subjects were tested on a HUMAC-NORM Isokinetic Dynamometer (Computer Sports Medicine, Inc. [CSMI], Stoughton, MA, 2009) using the dominant limb, which was defined as the leg with which the subject would kick a ball (Croce & Miller, 2006). Subjects were seated with stabilization straps placed around the chest, waist, and distal femur. Torque values were recorded in Nm, with real-time torque and position (ROM) data downloaded from the dynamometer into a BIOPAC MP 100 Data Acquisition and Analysis System (BIOPAC Systems Inc., Santa Barbara, CA, 2009) by way of external output channels, thus allowing for simultaneous recording and display of torque, position, and SEMG data. Experimental Protocol Dynamometer, torque, and position output signals were calibrated according to the appropriate systems manuals (CSMI, 2009; BIOPAC Inc., 2009). Subjects were tested at five angular velocities (00 s-1 [isometric], The Journal of Applied Research Vol.12, No. 2,

3 500 s-1, 1000 s-1, 2000 s-1, and 4000 s-1) and four contraction intensities (100%-, 75%-, 50%-, and 25%-MVC) through kneejoint angles of 00 (straight leg) and 900. At each test velocity subjects performed quadriceps contractions at the desired intensity (extension) followed by a passive flexion of the hamstrings. Peak torque (PT) and average torque (AVT) were determined by the repetition at which maximum torque was achieved during a six-repetition trial. Order of contraction intensities and movement velocities tested was counterbalanced over subjects through sampling without replacement procedure. For maximum voluntary contraction (MVC) tests, subjects were instructed to push into extension as hard as possible using strong verbal encouragement. An MVC was operationally defined as a maximal contraction that a subject accepts as maximal and that is produced with appropriate continuous feedback. Isometric MVCs were obtained with the knee flexed at 600 (Croce & Miller, 2006). All other muscle contractions were submaximal, and subjects were instructed to produce a torque corresponding to a specific percentage of their maximum at the corresponding velocity. For submaximal tests, subjects viewed a line graph of torque output and were instructed to reach, but not to exceed, the percent of maximum being tested (torque values produced were within 5% of the desired torque value). A 3-min rest was given between muscle actions to avoid fatigue. Each repetition from which data were analyzed, moment and SEMG recordings were calculated between 500-to-700 of knee flexion. This portion of the range of motion was selected as: (1) frequencies at the extremes of muscle length might be less representative of actual spectral frequencies represented in the muscle (Kaman & Caldwell, 1996), and (2) this range represented only a 100 difference above and below the angle at which isometric measures were made, thereby reducing muscle length discrepancies between isometric and isovelocity conditions. Recording of SEMG Bipolar SEMG was used to determine the electrical activity of the VL during testing. Silver/silver chloride pre-gelled surface electrodes (Moore Medical Corporation) were placed 2.5 cm apart according to points recommended by Criswell (2011). A common reference electrode was placed over the head of the fibula. Skin preparation for electrode placement included removal of dead epithelial cells with a razor, isopropyl alcohol, and an abrasive pre-gel (Nuprep gel) to achieve a skin impedance of < 5 kω. The SEMG signal was amplified, filtered, and analog-to-digitally converted (BIOPAC Systems Inc., MP 150 system) on-line with a sampling rate of 4000 Hz. Raw SEMG signals were monitored on-line, stored, and processed through a Dell Optiplex computer with high and low pass filters of 20 and 500 Hz, respectively. Gain was set at 1,000 with a common mode rejection ratio of 90 db. The SEMG signal was filtered (AcqKnowledge 4.1 software, BIOPAC Inc.,) and SEMG-MNF and SEMG-RMS were calculated for the repetition in which peak torque occurred and the repetition at which the desired percent MVC occurred for each test velocity. The SEMG-RMS data were normalized against the SEMG-RMS obtained for each muscle during the isometric MVC and was used as a measure of overall muscular activity (Criswell, 2011). The SEMG-MNF was processed using Fast Fourier Transformation with a Hamming window and was used to determine potential changes in motor-unit recruitment strategies (von Tscharner & Nigg, 2008; Wakeling, 2009). Statistical Analysis Data were analyzed using the StatView statistical program (SAS Institute, 1998). Peak torque and AVT isometric (00) and isovelocity (500,1000, 2000, and 4000 s-1) values were compared using separate repeated measures analysis of variance (ANOVA). The interaction of movement velocity and contraction intensity on VL SEMG-MNF 82 Vol.12, No. 2, 2012 The Journal of Applied Research.

4 Table 1. Means and standard deviations (in parentheses) of knee extensor peak torque (PT) and average torque (AVT) (in Nm) between 50- and 70-degrees of knee-joint motion as a function of movement velocity. Peak Torque (PT) Average Torque (AVT) Movement Velocity 0 0. s s s s s (+31.81) (+19.26) (+13.84) (+7.13) (+15.03) (+31.91) (+22.31) (+18.93) (+11.18) (+13.30) *Note: Isometric PT was obtained at 600 knee flexion and AVT was the average value over a 5-s contraction.; AVT for isovelocity movements was the mean value between 50- and 70-degrees of knee-joint motion. and SEMG-RMS were analyzed via separate 5 (velocity) x (4) (contraction intensity) repeated measures ANOVA. The conservative Greenhouse-Geisser factor was used to evaluate observed within-group F ratios. Criterion level for significance was p < RESULTS Torque Means and standard deviations for PT and AVT for maximum contractions at each velocity tested are shown in Table 1. Consistent with previous research, both PT and AVT of the quadriceps were greatest when tested isometrically, followed by increasing isovelocity contractions (F4, 36= , p < and F4, 36 = , p < , respectively). EMG Activity SEMG- MNF values of the VL for movement velocity and contraction intensity are displayed in Table 2. Significant velocity (F4, 36 = 6.950, p < 0.001) and contraction intensity (F3, 27 = 3.029, p < 0.05) main effects were found. Post-hoc analyses indicated: (1) greater SEMG-MNF values during all isovelocity conditions compared to isometric condition, with highest values at 500 s-1 compared to 1000, 2000 and 4000 s-1; and, (2) greater SEMG-MNF at 100%- MVC compared to 50%-, and 25%-MVC, with no significant differences between 100%-MVC and 75%-MVC. Normalized SEMG- RMS values of the VL for movement velocity and contraction intensity are displayed in Table 3. Significant main effects for movement velocity (F4, 36 = 8.723, p < 0.001) and contraction intensity (F3, 27 = , p < ), and a significant movement velocity x contraction intensity interaction effect (F12, 108 = 4.218, p < 0.01) were found. Post-hoc analyses indicated the following results of consequence: (1) a linear increase in SEMG amplitude as contraction intensity and force increased from 25%- to 100%-MVC across all isovelocities; and, (2) a significant increase in SEMG-RMS across all isovelocity conditions compared to isometric condition and greater values at 2000 s-1 compared to 500 s-1, with no significant differences amongst 1000, 2000, and 4000 s-1. DISCUSSION Contraction Intensity (Force) and Velocity Relationships for PT and AVT Results are consistent with the classical inverse relationship between torque (force) and velocity in that both PT and AVT were greatest when tested isometrically, followed by dynamic contractions at 500, 1000, 2000, and 4000 s-1 (Table 1). The literature is replete with data confirming this relationship and will not be discussed at any length here (Lieber, 2002). As anticipated, our data confirmed this relationship. Contraction Intensity (Force) and Velocity Relationships for SEMG-MNF Our results indicated greater SEMG-MNF values under all isovelocity conditions compared to isometric condition, and significantly higher SEMG-MNF values at 500 s-1 compared to those found at 1000, 2000 The Journal of Applied Research Vol.12, No. 2,

5 Table 2. Means and standard deviations (in parentheses) of vastus lateralis surface electromyography-median frequency power spectrum (SEMG-MNF) between 50- and 70-degrees of knee-joint motion as a function of movement velocity and contraction intensity. Movement Velocity 0 0. s s s s s (+ 6.52) (+8.79) (+9.43) (+8.66) (+8.96) Contraction Intensity Maximum Seventy-Five % Fifty % Twenty-Five % (+7.42) (+7.72) (+8.13) (+11.40) and 4000 s-1 (Table 2). These results are comparable to results reported by Croce et al. (unpublished data, 2011), who observed similar velocity-related spectral shifts in the vastus medialis (VM) and rectus femoris (RF) muscles, but are in opposition to results of Masuda et al (2001), who observed higher spectral frequencies under isometric contractions compared to that observed under isovelocity contractions between 600 and 2400 s-1; Aagaard et al (2000) and Croce et al (2000), who reported no significant relationship between velocity and SEMG-PSD; and, Gerdle et al (1988) and Cramer et al (2002), who reported spectral shifts based on the particular QF muscle investigated. Although results differ somewhat from the aforementioned investigations, discrepancies can be attributed to differences in the particular QF muscle investigated, differences in electrode placement, differences in isovelocities tested, and differences in the range of motion over which muscles were tested (the latter point being of most significance). Overall, our data indicated maximum recruitment of FT fibers during slow velocity (500 s-1) contractions compared to isometric or faster velocities. To date, the literature is unclear as to the effect of force on QF spectral frequencies. For example, Croce et al. (unpublished data, 2011), Gerdle and Karlsson (1994), and Karlsson and Gerdle (2001), all reported positive, torque-dependent relationships with SEMG-PSD, whilst other researchers reported either no such relationship (eg, Coburn et al, 2005) or a relationship based on muscle fiber-type morphology (eg, Bilodeau et al, 2003). Our results concur with that of Croce et al. (unpublished data, 2011), Gerdle and Karlsson (1994), and Karlsson and Gerdle (2001), who likewise demonstrated increasing SEMG-MNF values with increasing contraction intensities and torque levels (Table 2). Observed increases in SEMG-MNF from our data provide some evidence of specific fiber-type activation during increasing contraction intensities as espoused by von Tscharner and Nigg (2008) and Wakeling (2009). It would make intuitive sense that during a higher contraction intensity of the QF muscle, there would be a greater recruitment of FT fibers than during lower contraction intensities. Contraction Intensity (Force) and Velocity Relationships for SEMG-RMS Results on contraction intensity and muscle amplitude (increased SEMG-RMS with increasing contraction intensities) were consistent with the literature for both isometric (Fukuda et al., 2010) and dynamic muscular contractions (Coburn et al., 2005). In our investigation, SEMG amplitude of active motor units increased significantly across increasing levels of contraction intensity (Table 3). Prior studies investigating force and isometric contractions found mostly a curvilinear increase in SEMG amplitude with increasing muscle force because of concurrent increases in motor unit recruit- 84 Vol.12, No. 2, 2012 The Journal of Applied Research.

6 Table 3. Means and standard deviations (in parentheses) of vastus lateralis surface electromyography-root mean square (SEMG-RMS) as a percent of maximum voluntary contraction between 50- and 70-degrees of knee-joint motion as a function of movement velocity and contraction intensity. Movement Velocity 0 0. s s s s s (+6.52) 0.68 (+0.25) 0.72 (+0.29) 0.77 (+0.24) 0.73 (+0.27) Contraction Intensity Maximum Seventy-Five % Fifty % Twenty-Five % 1.02 (+0.13) 0.79 (+0.14) 0.59 (+0.17) 0.39 (+0.17) ment and firing rate up to 80%-MVC, after which increases are due primarily to increases in firing rate (Coburn et al, 2005). Coburn et al (2005) in a similar, but more limited, investigation involving velocity (00 and 300 s-1) and contraction intensity (20%, 40%, 60%, 80%, and 100%) reported a high correlation between SEMG amplitude and force in both static and dynamic contractions. Our data confirm this relationship as a nearly linear increase in SEMG-RMS was observed with increasing contraction intensities. With regards to the impact of movement velocity on VL muscle amplitude, data indicated a significant increase in SEMG-RMS at all isovelocities compared to isometric and greater values at 2000 s-1 compared to 500 s-1 (Table 3). Though precise comparisons between our data with that from other investigations are difficult due to differences in velocities and muscles tested, our data are consistent with previous research indicating lower muscle amplitude at lower isokinetic velocities compared to higher isokinetic velocities (Aagaard et al, 2000; Babault et al, 2002). Furthermore, observed lower SEMG- RMS at low velocities support the concept of reduced neural drive to the muscle during high-load situations espoused by Perrine and Edgerton (1978). According to this theory, high-tension muscle loading leads to markedly lower SEMG-AMP and is most often displayed during maximum eccentric and slow concentric QF contractions. According to Perrine and Edgerton (1978), a threshold of 960 s-1 was the particular threshold at which one finds this increased inhibition to muscle, which most probably acts to limit maximal tension in the knee and thereby preserving musculoskeletal integrity (Croce and Miller, 2006). CONCLUSIONS During voluntary muscular contractions, the CNS controls muscle force by varying number of motor units recruited, types of motor units recruited (FT vs. ST), and motoneuron firing rates. Based on our data, the degree to which FT muscle fibers were recruited -- as determined by higher SEMG-MNF values -- during the diverse levels of contraction intensity and velocity further obfuscates the way in which the central nervous system (CNS) recruits muscle fibers during a specific motor task. Our results indicated that isovelocity movements facilitated a greater recruitment of FT fibers compared to isometric contractions, peaking at at 50 s-1 and not at faster isovelocities (e.g., 1000, 2000 s-1, and 4000 s-1). It, therefore, appears that the degree to which the CNS coordinates muscular functioning is based on the interplay of both contraction intensity and force. This would have implications for resistance training programs wanting to maximize FT-fiber recruitment, where training with relatively slow dynamic movements would intensify the recruitment of FT fibers during the course of training. On the other hand, our data specified maximum muscle ampli- The Journal of Applied Research Vol.12, No. 2,

7 tude occurring at 2000 s-1. Consequently, if one were more interested maximizing muscle activation during resistance training, more moderately paced movements would work best. REFERENCES 1. Aagaard P, Simonsen EB, Andersen JL, Magnusson F, Bojsen-Moller F, Dyhre-Poulsen P. Neural inhibition during maximal eccentric and concentric quadriceps contraction: Effects of resistance training. J Appl Physiol 2000; 89: Babault N, Pousson M, Michaut A, Ballay Y, Van Hoecke J. EMG activity and voluntary activation during knee-extensor concentric torque generation. Eur J Appl Physiol 2002; 86: Bilodeau M, Schindler-Ivens S, Williams DM, Chandran R, Sharma SS. EMG frequency content changes with increasing force and during fatigue in the quadriceps femoris muscle of men and women. J Electrmyogr Kinesiol 2003; 13: BIOPAC Systems, Inc. BIOPAC Systems Acqknowledge 4.1 Software Users Manual. Santa Barbara, CA, 93117, Coburn JW, Housh TJ, Cramer JT, Weir JP, Miller JM, Beck TW, Malek MH, Johnson GO. Mechanomyographic and electromyographic responses of the vastus medialis muscle during isometric and concentric muscle actions. J. Strength Cond Res 2005; 19: Computer Sports Medicine, Inc. (CSMI) (2009). HUMAC NORM Testing and Rehabilitation System Users Manual. Stoughton, MA, Cramer JT, Housh TJ, Weir JP, Johnson GO, Ebersole KT, Perry SR, Bull AJ. Power output, mechanomyographic, and electromyographic responses to maximal, concentric, isokinetic muscle actions in men and women. J Strength Cond Res 2002; 16: Criswell E. Cram s Introduction to Surface Electromyography, 2nd ed. Boston: Jones and Bartlett, Croce RV, Miller JP. Angle- and velocity-specific alterations in torque and semg activity of the quadriceps and hamstrings during isokinetic extension-flexion movements. Electromyogr Clin Neurophysiol 2006; 46: Croce, R. V., Miller, J. M., Horvat, M., Smith, W. (2011). Quadriceps femoris SEMG consequent to alterations in contraction intensity and velocity, unpublished data, University of New Hampshire. 11. Croce RV, Miller JP, St. Pierre P. Effect of ankle position fixation on peak torque and electromyographic activity of the knee flexors and extensors. Electromyogr Clin Neurophysiol 2000; 40: Fukuda TY, Echeimberg JO, Pompeu JE, Lucareli PRG, Garbelotti S, Gimenes RO, Apolinario A. Root mean square value of the electromyographic signal in the isometric torque of the quadriceps, hamstrings and brachial biceps muscle in female subjects. J Appl Res 2010; 10: Gerdle B, Karlsson S. The mean frequency of the EMG of the knee extensors is torque dependent both in the fatigued and fatigued states. Clin Physiol 1994; 14: Gerdle B, Wretling ML, Henriksson-Larsen K. Do the fibre-type proportion and the angular velocity influence the mean power frequency of the electromyogram? Acta Physiol Scand 1988; 134: Kamen G, Caldwell GE. Physiology and interpretation of the electromyogram. J Clin Neurophysiol 1996; 13: Karlsson S, Gerdle B. Mean frequency and signal amplitude of the surface EMG of the quadriceps muscles increases with increasing torque a study using the continuous wavelet transformation. J Electromyogr Kinesiol 2001; 11: Komi PV, Viitsalo JH. Signal characteristics of EMG at different levels of muscle tension. Acta Physiol Scand 1976; 96: Kupa EJ, Roy SH, Kandarian SC, DeLuca CJ. Effects of muscle fiber type and size on EMG median frequency and conduction velocity. J Appl Physiol 1995; 79: Lieber RL. Skeletal muscle structure, function, & plasticity: The physiological basis of rehabilitation, 2nd ed. Baltimore: Lippincott Williams & Wilkins, Masuda T, Kizuka T, Yong Zhe J, Yamada H, Saitou K, Sadoyama T, Okada M. Influence of contraction force and speed on muscle fiber conduction velocity during dynamic voluntary exercise. J Electromyogr Kinesiol 2001; 11: Perrine JJ, Edgerton VR. Muscle force-velocity and power-velocity relationships under isokinetic loading. Med Sci Sports Exerc 1978; 10: Pincivero DM, Salfetnikov Y, Campy RM, Coelho AJ. Angle- and gender-specific quadriceps femoris muscle recruitment and knee extensor torque. J Biomechanics 2004; 37: SAS Institute. StatView Reference Manual, 2nd ed. Cary, NC: SAS Institute, von Tscharner V. & Nigg BN. Point: Spectral properties of the surface EMG can characterize motor unit recruitment strategies and muscle fiber type. J Appl Physiol 2008; 290: Wakeling JM. Patterns of motor unit recruitment can be determined using surface EMG. J Electromyogr Kinesiol 2009; 19: The Journal of Applied Research Vol.12, No. 2,