Reciprocal coactivation patterns of the medial and lateral quadriceps and hamstrings during slow, medium and high speed isokinetic movements

Journal of Electromyography and Kinesiology 10 (2000) 233–239 www.elsevier.com/locate/jelekin

Reciprocal coactivation patterns of the medial and lateral quadriceps and hamstrings during slow, medium and high speed isokinetic movements John P. Miller *, Ronald V. Croce , Ronald Hutchins Department of Kinesiology, University of New Hampshire, Durham, NH 03824, USA Received in revised form 22 February 2000; accepted 29 February 2000

Abstract The effect of movement velocity and fatigue on the reciprocal coactivation of the quadriceps and hamstrings was investigated through analysis of the root mean square (RMS) and the median frequency (MDF) of surface electromyography for the vastus medialis (VM), vastus lateralis (VL), medial hamstrings (MH) and biceps femoris (BF). Fourteen subjects performed six continuous isokinetic knee extension and flexion movements at 60°, 180° and 300° s⫺1, and 30 continuous movements at 300° s⫺1 to examine muscular fatigue patterns. Statistical analyses revealed that the RMS activity of the VM displayed greater coactivation than the VL (P⬍0.01) and the BF displayed greater coactivation than the MH (P⬍0.0001). There was no effect of velocity on the coactivation levels of the VM, the VL, or the MH; however, there was an effect of velocity on the coactivation levels of the BF (P⬍0.0001). Relative to MDF activity, the MH shifted upward as velocity increased (P⬍0.01) while the BF decreased between 180 and 300° s⫺1 (P⬍0.01). Results of the muscular fatigue test indicated that the RMS activity of the VM showed a higher degree of coactivation than the VL (P⬍0.01) and the BF showed approximately three times the coactivation level of the MH (P⬍0.001). The MDF of the VL and MH shifted downward as the repetitions progressed (P⬍0.01) with no changes for the VM or for the BF. Results of this study suggest that during isokinetic testing, both the VM and BF have significantly greater reciprocal coactivation levels when compared to the VL and MH, respectively. In addition, these results suggest that motor unit recruitment patterns of the VM and VL and the MH and BF differ with regard to the effects of velocity and fatigue.  2000 Elsevier Science Ltd. All rights reserved. Keywords: EMG; Reciprocal coactivation; Knee

1. Introduction The effect of quadriceps and hamstrings forces on knee-joint kinematics and stability has been investigated by researchers both in vitro [12,19] and in vivo [46,47]. Numerous studies have also measured quadriceps and hamstrings electromyographic (EMG) activity during knee-joint exercise [5,13,16,29,30,33,37,38,40]. Based on this, researchers have postulated that coactivation of the hamstrings during active knee extension assists the anterior cruciate ligament (ACL) in maintaining kneejoint stability by exerting an opposing force to anterior tibial translation [5,29]. An insufficiency in hamstrings

* Corresponding author. Tel.: +1-603-862-0263; fax: +1-603-8620154. E-mail address: [email protected] (J.P. Miller).

coactivation can lead to a lack of knee-joint stability, which can result in quadriceps muscle contractions creating unwanted stresses on internal joint structures, episodes of joint instability, and atrophy of the surrounding muscles [3,39]. Researchers have suggested that a proprioceptive mechanism exists that arises from the cruciate ligaments, which influences the tone in the thigh musculature and regulates hamstrings coactivation during active knee extension [18,38,40]. For example, Solomonow et al. [40] indicated that direct stress to the ACL produces quadriceps inhibition and hamstrings facilitation. Raunest et al. [34] concluded that the degree of muscle excitation on mechanical ligament loading is modulated by the amount of pre-load in the cruciate ligaments and the quality of load applied. It appears that sensory innervation of the cruciate ligaments is important for controlling movement and for protecting internal soft tissue

1050-6411/00/$ - see front matter  2000 Elsevier Science Ltd. All rights reserved. PII: S 1 0 5 0 - 6 4 1 1 ( 0 0 ) 0 0 0 1 2 - 2

234

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

structures by innervating and inhibiting the appropriate muscles of the knee [17,29]. If there is injury or surgical repair to the ACL, there is a danger that neuromuscular control mechanisms of the knee joint can become compromised. Recently Croce et al. [10] examined coactivation patterns of the quadriceps and hamstrings during slow and moderate speed isokinetic movements. They noted that the coactivation levels of the vastus medialis were greater than coactivation levels of the vastus lateralis. Similarly, the biceps femoris had significantly greater coactivation levels than did the medial hamstrings. One major limitation of the study by Croce et al. [10] was that high-speed isokinetic movements were not investigated nor were the effects of fatigue on the thigh musculature examined. The purpose of this study, therefore, was to (1) compare the effect of low, moderate, and high speed isokinetic movements on reciprocal coactivation patterns of the medial and lateral quadriceps and hamstrings as measured by the Root Mean Square of the EMG (RMS); (2) examine the effect of effect of low, moderate, and high speed isokinetic movements on reciprocal motor unit recruitment patterns of the medial and lateral quadriceps and hamstrings, as measured by the Median Frequency of the EMG (MDF) [8], and; (3) examine fatigue patterns of the medial and lateral quadriceps and hamstrings as measured by the RMS and MDF.

2. Methods 2.1. Subjects Subjects gave informed consent prior to participation. The Institutional Review Board of the University of New Hampshire approved the methods and procedures. The subjects were 14 healthy adult females (age=19.9±1.3 yr, mean height=162.1±7.7 cm, mean weight=64.4±9.6 kg) with no known knee pathologies. 2.2. Dynamometer set up A Cybex II+ Isokinetic Dynamometer [11] was used to measure the strength of the knee extensors and flexors of the dominant limb. Limb dominance was defined as the leg with which the subject would kick a ball [9]. Subjects were seated with stabilization straps placed around the chest, waist, and distal femur to eliminate extraneous movements and provide constant conditions across subjects. The axis of the dynamometer was aligned with the axis of the knee and the tibial pad was placed proximal to the medial malleolus. Torque values were recorded in Nm and corrected for the effect of gravity using the Cybex II+ gravity correction protocol [11] and procedures outlined by Fillyaw et al. [14].

2.3. Experimental protocol Prior to data collection, subjects participated in stretching exercises for the quadriceps and the hamstrings for approximately 10 min. Subjects were tested at angular velocities of 60, 180 and 300° s⫺1 [28,31]. Subjects were allowed 6–10 submaximal warm-up repetitions at each angular velocity to become familiar with procedures. Subjects then performed a maximal effort contraction of the quadriceps (extension) followed by a maximal effort contraction of the hamstrings (flexion) for six continuous repetitions at 60 and 180° s⫺1 and for 30 continuous repetitions at 300° s⫺1. A 10 min rest period was given between each test velocity to minimize the effect of fatigue on torque production and EMG activity [4]. Order of velocities tested was counterbalanced over subjects. Subjects were instructed to push or pull as fast as possible using strong verbal encouragement (‘push fast’ or ‘pull fast’) during the test procedures. Peak torque (PT) was identified as the highest recorded value among the sampled repetitions. 2.4. Recording of EMG Bipolar surface EMG was used to determine the electrical activity of the vastus medialis (VM), vastus lateralis (VL), medial hamstrings (MH) and biceps femoris (BF) during the isokinetic movements. Silver/silver chloride surface electrodes were placed as close as possible to the estimated motor end plates of the muscles according to Warfel [43]. The electrodes were 2.5 cm apart from center to center with a common reference electrode placed over the head of the fibula. The skin was cleaned and abraded to achieve skin impedance of ⬍5 k⍀. The EMG signal was digitized on-line with a sampling frequency of 1024 Hz using a data acquisition card (DAS-16 Metrabyte) processed through a Gateway 2000 486DX/33 computer with high and low pass filters of 20 and 400 Hz, respectively. The gain was set at 1000 with a common mode rejection ratio of 90 dB. The raw EMG signal was stored and the mean amplitude root mean square (RMS) and median frequency (MDF) of the EMG were calculated over the entire repetition (i.e. 90° of motion) in which the peak torque (PT) occurred for the three tested velocities. To determine possible changes with fatigue the repetition in which the peak torque occurred, the 15th and 30th repetition were analyzed. The RMS was used as a measure of muscular activity using the formula of Basmajian [7]. Normalized RMS for the antagonistic musculature was calculated as a percentage of the RMS activity from the same muscle during its agonistic phase [3,23,30]. The MDF was processed using Fast Fourier Transformation with a Hamming window and was used to determine potential changes in motor unit fiber recruitment. To determine the extent of cross-talk we examined the activity of all

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

Table 1 Means and standard error of measurement of the root mean square muscle cocontraction (% of agonist activity) as a function of movement velocity

Vastus medialis Vastus lateralis Medial hamstrings Biceps femoris

60° s⫺1 M SEM

180° s⫺1 M SEM

300° s⫺1 M SEM

21.0 14.0 15.3

2.3 1.3 2.8

20.1 14.1 15.2

2.2 1.2 2.7

20.2 14.5 18.9

1.8 1.2 3.2

37.2

2.8

39.0

5.7

52.1

8.5

tested muscles during an isometric contraction of the quadriceps with the limb in 80° of knee flexion. The ratios of the MH and BF activity during each simultaneous VM and VL measurement ranged between 3 and 5%, indicating the upper threshold of cross-talk [13]. 2.5. Statistical analysis Separate 3×4 (angular velocity×muscle) repeated measures analyses of variance (ANOVAs) were used to determine changes in the RMS and the MDF of the VM, VL, MH and BF at the three movement velocities. Separate 3×4 (repetition×muscle) repeated measures analyses of variance (ANOVAs) were used to determine changes in the RMS and MDF of the VM, VL, MH and BF during the 30 repetition fatigue test performed at 300° s⫺1. Repetitions used in these analyses were the repetition of PT, the 15th, and the 30th repetitions. The conservative Greenhouse–Geisser adjustment factor was used to evaluate observed within-group F ratios. Post hoc comparisons consisted of planned orthogonal contrasts. The criterion level for significant difference was set at P⬍0.05. 3. Results 3.1. Velocity Means and standard deviations for the normalized quadriceps and hamstrings RMS and MDF activity are shown in Tables 1 and 2. Analysis of the RMS indicated Table 2 Means and standard error of measurement of median frequency (Hz) as a function of movement velocity

Vastus medialis Vastus lateralis Medial hamstrings Biceps femoris

60° s⫺1 M SEM

180° s⫺1 M SEM

300° s⫺1 M SEM

59.0 70.2 73.0

1.6 4.6 4.2

63.2 74.7 79.8

1.3 5.0 2.5

60.8 72.7 81.2

2.6 3.5 4.2

77.1

3.0

78.9

3.8

71.9

3.1

235

Table 3 Means and standard error of measurement of the root mean square muscle cocontraction (% of agonist activity) as a function of fatigue

Vastus medialis Vastus lateralis Medial hamstrings Biceps femoris

Peak torque M SEM

15th M

SEM

30th M

SEM

20.2 14.5 18.9

1.8 1.2 3.2

24.5 19.2 18.5

2.5 2.0 2.1

25.7 19.7 22.2

4.9 3.1 4.1

52.1

7.7

60.4

9.0

57.9

8.7

that the VM displayed 1.5 times the coactivation of the VL at 60° (F=13.3, P⬍0.001), 180° (F=12.3, P⬍0.001), and 300° s⫺1 (F=7.8, P⬍0.01) with no velocity effect noted for the individual muscles. In addition the BF displayed approximately 2.5 times the coactivation of the MH at 60° (F=133.2, P⬍0.001), 180° (F=156.8, P⬍0.0001), and 300° s⫺1 (F=238.1, P⬍0.0001). There was no velocity effect for the MH; however, there was an increase in RMS activity for the BF between 60 and 300° s⫺1 (F=33.3, P⬍0.001) and between 180 and 300° s⫺1 (F=23.2, P⬍0.001). For the MDF there was no velocity effect for the VM or VL; however, the MDF of the MH shifted upward between 60 and 180° s⫺1 (F=5.9, P⬍0.01) and between 60 and 300° s⫺1 (F=8.5, P⬍0.01). Conversely, the MDF of the BF decreased between 180 and 300° s⫺1 (F=6.2, P⬍0.01). 3.2. Fatigue Means and standard deviations for the normalized quadriceps and hamstrings RMS and MDF activity are shown in Tables 3 and 4. Analysis of the RMS during the 300° s⫺1 fatigue test indicated that the VM showed a higher degree of coactivation than the VL for the repetition of PT (F=7.8, P⬍0.01), the 15th repetition (F=6.6, P⬍0.01), and the 30th repetition (F=8.4, P⬍0.01). In addition, the BF showed almost three times the coactivation of the MH for the repetition of PT (F=238.1 P⬍0.0001), the 15th repetition (F=420.8, P⬍0.0001), and the 30th repetition (F=304.1, P⬍0.0001). There was no significant difference in RMS Table 4 Means and standard error of measurement of the median frequency (Hz) as a function of fatigue

Vastus medialis Vastus lateralis Medial hamstrings Biceps femoris

Peak torque M SEM

15th M

SEM

30th M

SEM

60.8 72.7 81.2

2.6 3.5 4.2

59.0 61.6 69.8

1.9 2.4 3.3

57.2 62.0 66.4

3.2 5.8 2.9

71.9

3.1

73.8

3.4

68.8

4.0

236

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

activity between the PT, the 15th and the 30th repetitions for any of the muscles tested. For the MDF there was no fatigue effect for the VM; however, the MDF of the VL shifted downward between the repetition of PT and the 15th repetition (F=11.4, P⬍0.01) and the repetition of PT and the 30th repetition (F=10.5, P⬍0.01). The MDF of the MH shifted downward between the repetition of PT and the 15th repetition (F=11.9, P⬍0.001) and the repetition of PT and the 30th repetition (F=20.1, P⬍0.0001). There was no effect of fatigue on the MDF of the BF.

4. Discussion 4.1. Velocity—RMS In the present investigation on the effect of velocity on muscle activation levels for both the medial and lateral muscles of the knee were investigated. To date, most researchers have used the BF and VL as being representative of the hamstrings and quadriceps, respectively [5,16,29,30,37,38,40]. Therefore, comparisons made between data from our investigation with those from previous researchers were based principally on the BF and VL muscles (i.e. lateral muscles of the thigh). In addition, several other factors limit the extent to which comparisons can be made between our research and those of other researchers. These include differences in velocities investigated, whether integrated EMG (IEMG) or RMS analyses were performed, whether EMG analyses were performed on peak torque or mean torque values, and the point in the range of motion at which EMG analyses were performed [10,20,37]. Nonetheless, similar levels of antagonist BF activity during knee extension can be seen. Data from Snow et al. [37] (mean IEMG amplitude in which peak torque occurred=32%), Osternig et al. [29] (mean IEMG amplitude across trials=41%), and Osternig et al. [30] (mean IEMG amplitude across trials=33%) are all consistent with our data (mean RMS amplitude in which peak torque occurred=42%). To date, few investigators have attempted to investigate both BF and MH activity during knee extension [10,13]. Croce et al. [10] reported a mean RMS amplitude across trials of 36 and 10% for the BF and MH, respectively, and Draganich et al. [13] using a quasi-static knee extension movement (moving weights of 1.8, 3.6, 5.4, and 7.2 kg at a rate of 10° s⫺1), reported mean IEMG activity for the BF and MH from 0 to 67.5° of knee flexion for the 7.2 kg weight to be 12 and 4.7% of agonist activity, respectively. At terminal extension IEMG activity was 19% for the BF and 10% for the MH. The data of Croce et al. [10] were similar to those obtained in our investigation (mean BF=43% across velocities tested and mean MH=16% across velocities tested), while those reported

by Draganich et al. [13] were much lower than ours. One must realize, however, that Draganich et al. [13] employed an extremely slow, quasi-static movement. Consequently, the type of movement employed by Draganich et al. [13] would make comparisons with our data tenuous at best. The relatively low levels of quadriceps antagonist activity found in our study (mean VM=20% and VL=14% across velocities tested) contrasted with the correspondingly higher levels of hamstrings antagonist activity. Nevertheless, our data are in concurrence with the work of Croce et al. [10] (mean VM=23.5% and VL=12.5% across velocities tested), as well as previous studies using the VL as representative of quadriceps activity [5,16,30,38]. For example, Snow et al. [38] found that the level of quadriceps cocontraction ranged from 5 to 8% of maximum quadriceps activity between 70 and 15° of knee flexion for velocities tested, and Osternig et al. [30] found a mean of 6% throughout the joint range. Thus, similar to the findings of Croce et al. [10], the RMS antagonist activity of the quadriceps and hamstrings differed in three important ways. First, antagonist quadriceps activity was much lower than antagonist hamstrings activity under the same test conditions: combined BF and MH cocontraction activity was nearly 56% greater than that of the VM and VL. Secondly, antagonist quadriceps activity was slightly less variable between subjects than antagonist hamstrings activity (compare standard error scores in Table 1). Thirdly, other than BF, the other muscles tested did not display velocity dependence in cocontraction. Our data indicated that the laterally based BF had nearly 2.5 times the cocontraction level of the MH across velocities tested, while the medially based VM had about 1.5 times the cocontraction level of the laterally based VL across velocities tested. Altogether, the current data, as well as data from previous research [10,30,37,38], suggest that the knee flexors and extensors respond differently as antagonists during constant velocity, maximum efforts of the agonists. 4.2. Velocity—MDF A number of studies have noted that changes in the EMG power spectrum are associated with the conduction velocity of the respective motor units which, in turn, is related to muscle fiber size and twitch characteristics [2,8,35,41,45]. Solomonow et al. [41] suggested the use of the MDF as a tracking mechanism for motor unit recruitment strategies. Theoretically, 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, while a shift toward lower frequencies is indicative of a decrease in the average conduction velo-

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

city and a derecruitment of these same motor units. In our investigation there was a significant velocity effect for MDF in the MH and BF, but not in the VM and VL, suggesting that motor recruitment strategies used by the hamstrings varied based on movement. It was also noted that the MDF activity of the BF decreased as movement velocity increased, while the MDF activity of the MH increased as movement velocity increased. Other investigators have observed differences in MDF activity of the flexors and extensors of various agonist– antagonist muscle groups. Using surface EMG, the recruitment strategies of the biceps and triceps during elbow flexion and extension in stepwise and linearly increasing contractions were investigated by Sanchez et al. [36]. They reported differences in control strategies when the biceps muscle acted as an agonist, as opposed to when functioning as an antagonist, but not when the triceps acted as an agonist or as an antagonist. Although this seminal work confirmed that differences do exist in the recruitment strategy of a muscle (in this case the biceps) when its role changes from agonist to antagonist, it was not clear why the same pattern was not seen in the triceps muscle. An interesting aspect of our data, was not only the fact that the hamstrings responded to increasing movement velocities with a shift in motor recruitment patterns, but the fact that the MH and BF responded differently, one recruiting more larger-sized motor units (MH), while the other recruited more lower-sized motor units (BF), as movement velocity increased. One possible explanation for this could be that in view of the fact that the BF displayed a velocity effect in RMS activity and the MH did not, the CNS may recruit larger-sized motor units and, hence more fast-twitch muscle fibers, to compensate for the lack of an increase in muscle activation in the MH at higher movement velocities. The end result being a unique mechanism to control knee-extension movements during increased movement velocity: in one scenario one sees an increase in muscle activation (the BF), while in the other scenario one sees an increase in the size of the motor units being activated (the MH). Based on our investigation, and that of Bernardi et al. [8] and Croce et al. [10], different antagonist strategies appear to exist in the quadriceps and hamstrings. The question which needs to be addressed is whether or not these differences are found under varying conditions, such as closed chain movements in which the knee moves into an extended position. 4.3. Fatigue—RMS The reduction in muscle torque during consecutive isokinetic efforts has been investigated extensively [6,15,42]. This decrease in muscle torque has been attributed to both contractile failure [42] and the availability of energy sources within the muscle [22]. Though the

237

studies are fewer in number, the changes in EMG activity have also been monitored during consecutive dynamic muscular contractions as a way in which to describe the effects of fatigue on maximum strength production [24,27,32]. In general, two primary changes in the EMG signal appear to take place during exhaustive, repetitive exercise: (1) the power spectrum of the EMG signal shifts toward the lower band, which is thought to be caused by a decrease in muscle fiber conduction [1,25,32]; and, (2) the amplitude of the EMG increases [21,25,27]. A limitation in the research has been a dearth of studies investigating antagonist EMG activity during fatiguing dynamic exercise [21,33,44]. Weir et al. [44] found that the coactivation of the quadriceps and hamstrings increases during fatigue, but that the rate of increase was independent of contraction velocity. Psek and Cafarelli [33] reported that during fatiguing static leg extensions in a low-intensity, long-duration and a high-intensity, short-duration fatigue protocols, both the agonist muscle (VL) and the antagonist muscle (BF) showed an increase in EMG activity. The close relationship between the agonist and antagonist EMG changes supported the notion of a common drive to both motor neuron pools during the fatiguing bouts of exercise. Conversely, Kellis [21] found that antagonist hamstring EMG activity, as measured by the BF activity, remained relatively constant during an isokinetic concentric fatigue test of the knee extensors, despite of increased EMG activity of the agonists (VM). Neither of these investigations used frequency of the EMG to evaluated muscle fatigue in the thigh muscles. Our results indicated that there was no significant difference in RMS activity as the repetitions progressed for any of the quadriceps and hamstring muscles tested. This is in agreement with the work of Kellis [21] but contrary to the work of Psek and Cafarelli [33], who found a close relationship between agonist and antagonist increases in EMG activity as a result of fatiguing muscular contractions. The paradoxical results of our study and that of Kellis [21] to the research of Psek and Cafarelli [33] can be attributed to two major differences. First, there were differences in testing protocol. For example, our study involved 30 continuous extension–flexion maximal efforts, whereas Psek and Cafarelli [33] used 3 s contractions at 70% of maximal effort, with 7 s rest intervals between each contraction. Secondly, there were major differences in the types of movements performed: Psek and Cafarelli [33] evaluated coactivation patterns during static contractions whereas Kellis [21] and we evaluated coactivation patterns during an isokinetic concentric fatigue test. It is generally well accepted that there are significant differences in mechanisms of torque development and muscle activation between isometric and concentric muscular contractions [21].

238

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

4.4. Fatigue—MDF

Relative to MDF activity with fatiguing contractions, we observed different patterns of activity among the muscles tested. The VM and BF displayed no changes in MDF, whereas the VL and MH displayed a decrease in MDF. These decreased MDF levels would be indicative of fatigue in the muscle and would be contrary to the results observed in RMS activity. However, closer inspection of the data discloses that although there were no statistically significant differences in RMS activity across repetitions (P=0.06), there was, nonetheless, a modest increase in RMS activity from the repetition of peak torque to the 15th and the 30th repetitions. Once again, this would be indicative of some fatigue in the muscle. It appears that, in our investigation, MDF activity was more sensitive to muscle fatigue than was RMS activity. This would support the contention of Solomonow et al. [41] who suggested the use of the power spectrum of the EMG as a tracking mechanism for motor unit recruitment strategies, and that the MDF shift toward lower frequencies during dynamic exercise can be interpreted as a reliable determinant of local muscle fatigue [1].

5. Conclusions

1. The VM showed greater coactivation than the VL at 60 and 300° s⫺1, with no differences noted for the individual muscles between velocities. In the hamstring muscle group, the BF showed greater coactivation than the MH. 2. During the 300° s⫺1 fatigue test, the VM and VL displayed comparative coactivation levels similar to that seen during the PT repetitions. In addition, the BF showed approximately three times the coactivation of the MH. 3. There was no velocity effect for MDF for the VM or the VL. The MDF for the MH shifted upward across velocities tested while the MDF for the BF decreased between 180 and 300° s⫺1. 4. During the 300° s⫺1 fatigue test, the MDF for the VL and MH shifted downward. 5. The results of this study suggest that during isokinetic flexion–extension movements, both the VM and the BF have significantly greater reciprocal coactivation levels when compared to the VL and the MH. In addition, the results presented here suggest that the reciprocal motor unit recruitment patterns of the of the VM and VL and the MH and BF differ with regard to the effects of velocity and fatigue [10,26].

References [1] Ament W, Bonga GJJ, Hof AL, Verkerke GJ. EMG median power frequency in exhausting exercise. J Electromyogr Kinesiol 1993;3:214–20. [2] Ardendt-Nielsen L, Mills K. The relationship between mean power frequency of the EMG spectrum and muscle fibre conduction velocity. Electroenceph Clin Neurophysiol 1985;60:130–4. [3] Arms SW, Pope RH, Johnson RJ, Fischer RA, Arvidsson I, Ericksson E. The biomechanics of anterior cruciate ligament rehabilitation and reconstruction. Am J Sports Med 1984;12:8– 18. [4] Arnold BL, Perrin DH. Effect of repeated isokinetic concentric and eccentric contractions on quadriceps femoris muscle fatigue. Isok Exer Sci 1995;5:81–4. [5] Baratta R, Solomonow M, Zhou BH, Letson D, Chuinard R, D’Ambrosia R. Muscular coactivation: the role of the antagonist musculature in maintaining knee stability. Am J Sports Med 1988;16:113–22. [6] Barnes W. Isokinetic fatigue curves at different contractile velocities. Arch Phys Med Rehab 1981;62:62–9. [7] Basmajan JV, De Luca C. Muscles alive: their functions revealed by electromyography. Baltimore: Williams and Wilkins, 1985. [8] Bernardi M, Solomonow M, Baratta R. Motor unit recruitment strategy of antagonist muscle pair during linearly increasing contraction. Electromyogr Clin Neurophysiol 1997;37:3–12. [9] Bruininks RH. Bruininks–Oseretsky test of motor proficiency. Circle Pines, MN: American Guidance Service, 1978. [10] Croce RV, Miller JP, Confessore R, Vailis J. Reciprocal coactivation patterns of the lateral and medial quadriceps and hamstrings during low- and moderate-speed isokinetic movement. J Sports Rehab 1998;7:182–96. [11] Cybex II handbook: isolated-joint testing and exercise. Ronkonkoma, NY: Lumex Incorporated, 1983. [12] Draganich F, Vahey JW. An in-vitro study of anterior cruciate ligament strain induced by quadriceps and hamstring forces. J Orthop Res 1990;8:57–63. [13] Draganich LF, Jaeger RJ, Kralj AR. Co-activation of the hamstrings and quadriceps during extension of the knee. J Bone Joint Surg 1989;71A:1075–81. [14] Fillyaw MT, Bevins L, Fernandez L. Importance of correcting isokinetic peak torque for the effect of gravity when calculating knee flexor to extensor muscle ratios. Phys Ther 1986;66:23–31. [15] Gray JW, Chandler JM. Percent decline in peak torque production during repeated concentric and eccentric contractions of the quadriceps femoris muscle. J Orthop Sports Phys Ther 1989;10:309–14. [16] Hagood S, Solomonow M, Baratta R, Zhou BH, D’Ambrosia R. The effect of joint velocity on the contribution of the antagonist musculature to knee stiffness and laxity. Am J Sports Med 1990;18:182–7. [17] Harter RA, Osternig LR, Singer KM. Knee joint proprioception following anterior cruciate ligament reconstruction. J Sports Rehab 1992;1:103–10. [18] Johansson H. Role of knee ligaments in proprioception and regulation of muscle stiffness. J Electromyogr Kinesiol 1991;1:158– 79. [19] Kalund S, Sinkjaer T, Arendt-Nielsen L, Simonsen O. Altered timing of hamstring action in anterior cruciate ligament deficient patients. Am J Sports Med 1990;18:245–8. [20] Kellis E. Quantification of quadriceps and hamstring antagonist activity. Sports Med 1998;25:37–62. [21] Kellis E. The effects of fatigue on the resultant joint moment, agonist and antagonist electromyographic activity at different angles during dynamic knee extension efforts. J Electromyogr Kinesiol 1999;9:191–9.

J.P. Miller et al. / Journal of Electromyography and Kinesiology 10 (2000) 233–239

[22] Kellis E, Baltzopoulos V. Isokinetic eccentric exercise. J Sports Med 1995;19:202–22. [23] Kellis E, Baltzopoulos V. The effects of normalization method on antagonistic activity patterns during eccentric and concentric isokinetic knee extension and flexion. J Electromyogr Kinesiol 1996;6:235–45. [24] Komi PV, Tesch P. Frequency spectrum, muslce structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol 1979;42:41–50. [25] Masuda K, Masuda T, Sadoyama T, Inaki M, Katsuta S. Changes in surface EMG parameters during static and dynamic fatiguing conditions. J Electromyogr Kinesiol 1999;9:39–46. [26] McNair PJ, Wood GA. Frequency analysis of the EMG from the quadriceps of anterior cruciate ligament deficient individuals. Electromyogr Clin Neurophysiol 1993;33:43–8. [27] Oberg T. Muscle fatigue and calibration of EMG measurements. J Electromyogr Kinesiol 1995;5:239–43. [28] Osternig LR. Isokinetic dynamometry: implications for muscle testing and rehabilitation. Exer Sports Sci Rev 1986;14:45–80. [29] Osternig LR, Caster BL, James CR. Contralateral hamstring (biceps femoris) coactivation patterns and anterior cruciate ligament dysfunction. Med Sci Sports Exer 1995;27:805–8. [30] Osternig LR, Hamill J, Lander JE, Robertson R. Co-activation of sprinter and distance runner muscles in isokinetic exercise. Med Sci Sports Exer 1986;18:431–5. [31] Perrine JJ, Edgerton VR. Muscle force–velocity relationships under isokinetic loading. Med Sci Sports Exer 1978;10:159–66. [32] Potvin JR, Bent LR. A validation of techniques using surface EMG signals from dynamic contractions to quantify muscle fatigue during repetitive tasks. J Electromyogr Kinesiol 1997;7:131–9. [33] Psek JA, Cafarelli E. Behavior of coactive muscles during fatigue. J Appl Physiol 1993;74:170–5. [34] Raunest J, Sager M, Burgener E. Proprioceptive mechanisms in the cruciate ligaments: an electromyographic study on reflex activity in the thigh muscles. J Trauma: Inj Infect Crit Care 1996;41:488–93. [35] Sadoyama T, Masuda T, Miyata H, Katsuta S. Fiber conduction velocity and fiber composition in human vastus lateralis. Eur J Appl Physiol 1988;57:767–71. [36] Sanchez JH, Solomonow M, Baratta R, D’Ambrosia R. Control strategies of the elbow antagonist muscle pair during two types of increasing isometric contractions. J Electromyogr Kinesiol 1993;3:33–40. [37] Snow CJ, Cooper J, Quanbury AO, Anderson JE. Antagonist cocontraction of knee flexors during constant velocity muscle shortening and lengthening. J Electromyogr Kinesiol 1993;3:78–86. [38] Snow CJ, Cooper J, Quanbury AO, Anderson JE. Antagonist cocontraction of knee extensors during constant velocity muscle

[39]

[40]

[41]

[42]

[43] [44]

[45]

[46]

[47]

239

shortening and lengthening. J Electromyogr Kinesiol 1995;5:185–92. Solomonow M, Baratta R, D’Ambrosia R. The role of the hamstrings in the rehabilitation of the anterior cruciate ligamentdeficient knee in athletes. Sports Med 1989;7:42–8. Solomonow M, Baratta R, Zhou BH, Shoji H, Bose W, Beck C, D’Ambrosia R. The synergistic action of the anterior cruciate ligament in maintaining joint stability. Am J Sports Med 1987;15:207–313. Solomonow M, Baten C, Smit J, Baratta R, Hermens H, D’Ambrosia R, Shoji H. Electromyogram power spectra frequencies associated with motor unit recruitment strategies. J Appl Physiol 1990;68:1117–85. Tesch PA, Dudley GA, Duvoisin MR, Hather BM, Harris RT. Force and EMG signal patterns during repetitive bouts of concentric and eccentric actions. Acta Physiol Scand 1990;138:263–71. Warfel JH. The extremities: muscles and motor points. 5th ed Philadelphia: Lea and Febiger, 1985. Weir JP, Keefe DA, Eaton JF, Augustine RT, Tobin DM. Effect of fatigue on hamstring coactivation during isokinetic knee extensions. Eur J Appl Physiol 1998;78:555–9. Westbury JR, Shaughnessy TG. Associations between spectra representation of the surface electromyogram and fiber type distribution in human masseter muscle. Electromyogr Clin Neurophysiol 1987;27:427–35. Yasuda K, Sasaki T. Muscle exercise after anterior cruciate ligament reconstruction: biomechanics of the simultaneous isometric contraction method of the quadriceps and the hamstrings. Clin Orthop 1987;220:266–74. Yasuda K, Sasaki T. Exercise after anterior cruciate ligament reconstruction: the force exerted on the tibia by the separate contractions of the quadriceps and or the hamstrings. Clin Orthop 1987;220:275–83.

Dr John P. Miller is an Associate Professor of Kinesiology teaching in the athletic program at the University of New Hampshire. His research interests are in biomechanical analysis of exercises commonly used for rehabilitation of athletes and is the coordinator of the Biomechanical Research Laboratory. His work has been published in the Journal of Applied Physiology, the Journal of Sports Rehabilitatio, and the Journal of Athletic Training and Isokinetics and Exercise Science. Dr Ronald V. Croce is a Professor of Kinesiology teaching at the University of New Hampshire. His research interests are in the areas of motor learning and motor control. His work has been published in the Journal of Sports Rehabilitation, the Journal of Athletic Training and Isokinetics and Exercise Science, Pediatric Exercise Science and Clinical Kinesiology. Ronald Hutchins is a graduate of the athletic training program at the University of New Hampshire. He is currently working as the Athletic Trainer at Albany Preparatory School in Albany, NY. He has previously had his research published in Isokinetics and Exercise Science.