- Email: [email protected]
Torque, work and EMG development in a heel-rise test
Clinical Biomechanics13(1998)344-350
ELSEVIER
Torque, work and EMG development in a heel-rise test U. osterberg, U. Svantesson”, H. Takahashi, G. Grimby Department of Rehabilitation
Medicine, Gijteborg Universiry Guldhedsgatan 19, 413 45 Giiteborg, Sweden
Received5 August 1997;accepted3 December1997
Abstract Objective. To study the fatigue process in the calf muscle during a standing heel-rise test. Design. Heel-rises were performed on a force plate measuring development of torque in the ankle joint. Background. The heel-risetest is a commonlyemployedclinical test to evaluatethe function of the calf muscleby counting the
number of heel-rises before exhaustion. Development of fatigue during other eccentric-concentric actions has earlier been studied as decreasesin force, work and EMG. Methods. Ten healthy males (mean age 25 yr) participated in the study. Torque and work were calculated using measurements from a force plate. Mean power frequency and root-mean-square of the EMG in the triceps surae were evaluated separately in the eccentric and concentric phases. Results. Increases of mean torque during both the eccentric and concentric phases were found. Work performed decreased during the concentric phases due to decreased range of motion. No changes in root-mean-square and decreasesin mean power frequency during the concentric phases indicated development of muscular fatigue, whereas decreasesin both root-mean-square and mean power frequency during the eccentric phases indicated decreasing muscular activity. Conclusions. Accordingly, the limiting factor for the heel-rise test was not loss of muscle force at the range of motion used, but rather a failure to maintain the initial range of motion owing to muscle fatigue. Relevance This method of calculating torque development in the ankle joint provides an opportunity to study the fatigue process in terms of performance. The results show that the heel-rise test reflects muscle endurance rather than strength of the calf muscle. 0 1998 Elsevier Science Ltd. All rights reserved. Keywords:
Mean powerfrequency;Root-mean-square; Calf muscle;Fatigue
1. Introduction
The heel-rise test consists of eccentric-concentric muscle actions of plantar flexion. It is often used in clinical conditions for evaluating the function of the calf muscle as these muscle actions are often used in such physical activities as walking, running and jumping.
The
eccentric
pre-activation
renders
an
increase in performance during the concentric phase, partly due to the utilization of the elastic energy stored during the pre-activation [l]. Recently, the heel-rise test has been evaluated by Lundsford and Perry [2] concerning the number of performed heel-rises repre*Correspondenceand reprint requests to: UIIa Svantesson, Departmentof Rehabilitation Medicine, Guldhedsgatan19, 413 45 Giiteborg,Sweden.E-mail: [email protected]
senting normal capacity. Svantesson et al. [3] investigated differences in fatigue development between the m. gastrocnemius medialis and m. soleus during the test, together with an estimation of mechanical work. However, no studies have been made with measurements of the development of torque and work during the heel-rise test to evaluate further the outcome of the test. According to Edwards [4], fatigue during isometric activity is defined as a failure to maintain the required or expected force. Development of fatigue has been evaluated in dynamic conditions using other variables as well, such as decrease in work [5], increase in impact force [6] or decreases in the mean power frequency (MPF) of the electromyographic (EMG) record [7]. In isokinetic studies, Gerdle and LAngstriim [5] found that measurements
0268-0033/98/$19.00 + 0.000 1998ElsevierScienceLtd. All rights reserved. PII: SO268-0033(98)00100-9
of fatigue
should
be made of the
U. &e&erg
et aLlClinical
Biomechanics
decrease in work and not peak torque, whereas differences in motor unit behavior during fatigue are of greater importance for work than for peak torque. The development of fatigue can also reduce the effect of the eccentric pre-activation if the coupling time becomes too long, as this might result in a loss of stored elastic energy by dissipation as heat [8]. One method to evaluate muscular fatigue is to study the alterations of the EMG power spectrum by means of the MPF, as a decrease in MPF is mainly a result of the decrease in the action potential conduction velocity [9, lo]. This decrease in action potential conduction velocity has been found to be related to the lactate concentration in the muscle [ll]. As the conditions in the muscle are similar in both the eccentric and concentric phases, as a concentration of metabolites due to fatigue in one phase also affects the other phase, it might be difficult to observe differences in MPF between the two phases. However, other factors might also influence the MPF, such as muscle activity level, action potential modification, firing rate, synchronization, skin temperature and additional recruitment of motor units at fatigue (for a review, see Hagg [12]). Highly significant linear correlations have been found between the signal amplitude of the EMG and the force developed [13] during fatiguing exercises. Gerdle et al. [13] also found correlations between MPF and signal amplitude of the EMG. In short, a decrease in MPF along with an increase or no change in the signal amplitude of the EMG is considered to be a practical measurement of muscular fatigue. Earlier studies were made on torque development in the knee during static conditions on a force plate by Thomee et al. [14] and Schuldt et al. [15], the latter emphasizing the importance of measuring the torque caused by anterior-posterior forces. In a study by Bobbert et al. [16], torque in the ankle joint during jumping was calculated, omitting factors arising from angular acceleration, using measurements from a force plate. The purpose of this study was to measure the torque influencing the ankle joint during a standing heel-rise test from the force measurements from a force plate and to calculate work during the test. The development of torque and work, as well as EMG, from the calf muscle were also to be investigated for evaluation of the fatigue process during the heel-rise test. 2. Methods 2.1. Subjects
Ten healthy male students of physiotherapy with a mean age of 25 yr (SD 3 yr: range 21-29 yr), a mean weight of 76 kg (SD 7 kg, range 67-89 kg) and a mean
13 (1998)
345
344-350
height of 179 cm (SD 3 cm, range 164-189 cm) participated in the study. Informed consent was obtained from the participants, and the study was approved by the Ethics Committee of the Faculty of Medicine, Goteborg University, Sweden. 2.2 Methods
Before the test, the subjects warmed up for 5 min with submaximal bicycling. The heel-rise test was performed with the subject, who wore standardized shoes and stood on a force plate (AMTI OR6-5, Newton, MA) and a lo” tilted wedge. For balance, the finger tips touched the wall at the height of the shoulders. The subjects were instructed to lift the heel as high as possible, with a straight knee, until no further heel-rises could be performed due to exhaustion. The force signals were then processed by an AMTI SGA6-4. A metronome was set to 92 beats per minute in order to exceed an angle velocity of approximately 60 deg s- 1 (1.05 rad ss’) for at least 0.41 s. For the measurement of range of motion of the ankle joint, a goniometer (Penny and Giles Biometrics Ltd, Gwent, UK) was placed on the lower lateral side of the leg and the lateral side of the foot. 90” in the range of motion was defined as when the foot was perpendicular to the leg, and 80” was in the direction of dorsiflexion. The signals from the amplified force signal and the angle of the ankle joint were sampled with a frequency of 100 Hz by an IBM Personal Computer AT. A study by Bobbert et al. [16] used a formula for calculation of torque in the ankle joint by Elftman [17]. They found that maximal error due to angular acceleration was 7.5 N at an angular velocity of 5.24 rad s ’ and even less at lower velocities; therefore, they omitted the part of the formula depending on angular acceleration. As maximal angular velocity was to be approximately 1.05 rad s -’ and the forces were supposed to exceed approximately 500 N, this part of the formula was omitted in this study as well. Accordingly, the torque development during the test on the ankle joint was calculated with the formula M, = F7Ax+ F,Az
where F, was given from the force plate as the vertical force and F, as the horizontal force in the anteriorposterior direction. The distances 4~ and AZ were calculated as Ax=x,?-
rcos lO”+t sin 1O”+1cos 10”
+ 1cos((/I-- 90”) AZ= -(x0-+,)
tan lO”+tcos lO”+rsin
+ I sin 10” + 1sin(& - 90’)
10”
L! &terberglClinical
346
Biomechanics 13 (1998) 344-350
where ,xtpis the center of pressure on the force plate in the anterior-posterior direction during the test, r is the distance in the anterior-posterior direction between the dorsal part of the heel and the ankle joint, t is the distal-proximal distance between the wedge and the ankle joint, I is the distance in the anterior-posterior direction between the dorsal part of the heel and the metatarsophalangeal joints and x0 is the coordinate in the anterior-posterior direction for the dorsal part of the heel. C$is the angle of the ankle joint given by the goniometer (Figure 1). The total amount of work during the whole test from tstartto hop was calculated for both the eccentric and the concentric phases according to the formula hop W=
X M,o,At r=lU,,t
where M’ is the torque and o, is the angular velocity in each sample (t) and At is the sampling time (0.01 s). The angle and velocity were filtered by a moving average filter of an order of five. According to Bobbert et al. [16], a low pass filter with a cut-off frequency of 5 Hz was not sufficient for measurements of high angular velocities (5.24 rad SC’); therefore, they used a
cut-off frequency of 16 Hz. The sampling frequency of 100 Hz in the present study, with angular velocities of approximately 60 deg s-’ (1.05 rad ss’) and the filter as above, was taken to be sufficient for the purposes of this study. Using the preset beat of 92 metronome beats per minute, an angular velocity of 60 deg SC’was supposed to be exceeded. However, this velocity was kept, as seen in Figure 3, for only short periods of time. To be able to make a frequency analysis of the EMG, it was necessary for the time during which the analysis was made to be longer than these periods. Angular velocities exceeding approximately 40 deg ss’ were thus accepted during the time period during which the frequency analysis was made. The beginning of each phase was defined as when the angular velocity did not change sign until an angular velocity of at least 40 deg so’ was reached, and the end of each phase was defined as when the angular velocity changed sign. This is the definition of a concentric or an eccentric phase used in this study. Total work for both the concentric and eccentric phases for each heel-rise was calculated and plotted. The change in work was then calculated as the Torque
0
(Nm)
li
II
18
Ang 1 e (deg)
Fig. 1. Position of the foot in the heel-rise test. R is the distance in anterior-posterior direction between the heel and the ankle joint, t is the horizontal distance between ground and ankle joint, 1 is the distance in the anterior-posterior direction between heel and metatarsophalangeal joints and .x0 is the coordinate in the anteriorposterior direction for the heel.
Fig. 2. Trace of the torque and of the angle of the ankle joint during the test as a function of time.
l% &e&erg
et al./Clinical
347
Biomechanics 13 (1998) 344-350
percentage difference between the first four and last four heel-rises. The mean torque values for both the eccentric and concentric phases were extracted and plotted. The change in mean torque was then calculated as the percentage difference between the first four and last four heel-rises for both the eccentric and concentric phases. The coupling time was defined as the time elapsed between the eccentric and the concentric phases. The mean coupling time for the first four and last four heel-rises was calculated. EMG activity was recorded with AgiAgCl surface electrode discs with a diameter of 9 mm (Red Dot, 2239 monitoring electrodes, 3M Medica, Borken, Westfalen, Germany). After the skin was shaved and scrubbed with alcohol, two electrodes at a distance of approximately 30 mm were secured on the bellies of the gastrocnemius medialis and the soleus muscle in a distal-proximal direction. The reference electrodes were placed on the bony part of the medial and lateral parts of the knee. The EMG signal was pre-amplified with a gain of 1000 and thereafter band-pass-filtered between 7 and 490 Hz by a KC-EMG (Chattanooga Group, Inc., TN, USA). The angle and the EMG were sampled at a frequency of 1250 Hz on a Macintosh computer with software developed in Lab View (National Instruments Corporation, Austin, TX, USA) by Punos Electronic AB, Goteborg, Sweden. The EMG signal was then additionally band-pass-filtered using a Butterworth filter with cut-off frequencies of 10 and 300 Hz. The part of the EMG signal for which an angular velocity above 40 deg ss’ was obtained was separately derived for the concentric and eccentric phases by observing the angle of the ankle joint from the goniometer. The same range of motion was derived for all heel-rises in the same test. For each such part, the average of the amplitude of the rectified signal was calculated using the root-mean-square (RMS) method. The fast Fourier transform method was used to obtain the MPF. The development of the MPF and RMS was calculated with simple regression analyses and then reported as percentage decrease from the initial value. This was done separately for the concentric and eccentric phases.
3. Results The total number of heel-rises performed was, on average, 36 (SEM, 2). The pattern of the torque and work development during the test are seen from the recordings for one test person (Figures 2 and 3). The maximal angle of the range of motion decreased by 10” (SEM, 2”) during the test starting with a range of motion of 47” (SEM, Zo). The velocity during the test exceeded the expected 60 deg ss ‘, but only for a small range of motion. The mean time that elapsed between the eccentric phase and the concentric phase, i.e. the coupling time, was 300 ms for the first part of the test and 370 ms for the last part of the test, with no significant difference between the beginning and the end of the test. As seen in Figure 4, the mean torques in both the eccentric and the concentric phases showed significant increases. There was an increase in peak torque of 16% (SEM, 3%) in the eccentric phase and of 22% (SEM, 3%) in the concentric phase. The total amount of work performed during the test was 6624 J (SEM, 467 J) during the eccentric phases and 6631 J (SEM, 472 J) during the concentric phases. Power
(WI
t (59 0
6
Velocity
II
IO
24
24
38
42
u
El
q
I8
a
e
16
u
48
51
10
(deg/s)
2.3. Statistics Mean values and standard deviation (SD) were used for the study population and mean values and standard errors of the mean (SEM) were calculated for measurements with conventional methods. Wilcoxon’s one-sample non-parametric test was used for differences. The level of significance was set at 0.05.
0
I
12
Fig. 3. Trace of the power and of the angle velocity of the ankle joint during the test as a function of time.
348
U dsterherglClinica1 Biomechanics 13 (1998) 344-350
250 80.
20 OJ 0
5
10
15
20
25
30
35
40
Ol 0
45
5
10
15
20
25
30
Fig. 4. Mean torques for both eccentric (0) and concentric actions as a function of number of heel-rises.
(0)
The development of work showed no significant change during the eccentric phases (8% (SEM, 6%)), but a significant decrease in the concentric phases of 20% (SEM, 6%) (Figure 5). All decreasesin MPF were significant. In the eccentric phase the decreaseswere 32% (SEM, 6%) for the gastrocnemius muscle and 27% (SEM, 4%) for the soleus muscle. The decreases in the concentric phase were 16% (SEM, 3%) for the gastrocnemius muscle and 10% (SEM, 3%) for the soleus muscle (Figures 6 and 7). All decreases in RMS in the eccentric phases were significant, but there were no changes in the concentric phases (Table 1). In the eccentric phases, the decreases of RMS were 17% (SEM, 7%) for the gastrocnemius muscle and 35% (SEM, 11%) for the soleus muscle (Figure 8).
The present study showed increases in mean torque in both phases, a decrease in work during the concentric phases and no change in work during the eccentric phases. There was no change in RMS and a decrease
45
in MPF during the concentric phases, but decreases in both RMS and MPF in the eccentric phases were found. The torque curves from one single heel-rise showed a similar pattern to that seen in torque curves obtained from a dynamometer with maximal eccentric-concen-
A :
60.
it x
40. 20.
OJ
.
0
5
10
15
20
25
30
35
40
8
45
Number of heel-rises Fig. 7. MPF for the soleus muscle for both eccentric (m) and concentric (0) actions as a function of the number of heel-rises.
Table 1 P-values for changes during the heel-rise test
0
160.
P-value Coupling time Torque Work
60
MPF
40
Ecc Con Ecc Con Ecc Con
0
40
Fig. 6. MPF for the gastrocnemius muscle for both eccentric (0) and concentric (0) actions as a function of the number of heel-rises.
4. Discussion
r F
35
Number of heel-rises
Number of heel-rises
180.
i
5
10
15
20
25
30
35
40
45
Number of heel-rises Fig. 5. Total work for each heel-rise for both eccentric (0) and concentric (0) actions as a function of the number of heel-rises.
RMS
Ecc Con
Gastro Soleus Gastro Soleus Gastro Soleus Gastro Soleus
0.1688 0.0077 0.0051 0.2411 0.0189 0.005 1 0.0050 0.0069 0.0280 0.0364 0.0217 0.7211 0.2619
L! herberg
et al.lClinical
900 800 700 I 6001
0
5
10
15
20
25
30
35
40
45
Number of heel-rises 8. RMS both for the soleus muscle for eccentric (D) and concentric (o) actions and for the gastrocnemius muscle for eccentric (0) and concentric (0) actions as a function of the number of heelrises.
tric actions [18]. Whereas there were high torque values at the end of the eccentric phase and the beginning of the concentric phase, low values were seen at the end of the concentric phase and the beginning of the eccentric phase. This was probably because of the short lever arm in the highest position in the heel-rise test and because there was a reduced performance ability with a shorter muscle in the dynamometer test. Figure 4 shows mean isokinetic torque values during the concentric phases of approximately 220 N m. These torque values appear to be higher than during maximal voluntary concentric peak torques in eccentric-concentric actions with an angular velocity of 120 deg SC’ on a dynamometer of 88 N m, which were found in young men by Svantesson and Grimby [19]. In contrast, de Graaf et al. [20] found peak torques in jumps of 194 N m, and Liischer et al. [21] found isometric torques on a dynamometer of 196 N m. The discrepancy between the different torques of the ankle joint might be explained by the findings of Bobbert and van lngen Schenau [22], who demonstrated that, in simulated isokinetic plantar flexion at high angular velocities, the duration of the movement was too short for the torque to become maximal. However, in the present study, the velocity is rather low and the range of motion high. The range of motion of 47” at the beginning of the test was not far from the maximal range of motion of the ankle joint, which is 61” for males between 40 and 44 yr, as found by Fugl-Meyer et al. [23].
As seen in Figure 2, the mean torque increased during the test. This increase probably occurred as a result of increasing effort to obtain a required range of motion, but might also be due to a slow displacement during the test of the center of pressure of the foot in the anterior direction, leading to a longer lever-arm and, therefore, the higher torque values observed in some subjects. Despite increasing effort, the subjects
Bion~echanics 13 (1998) 344-350
349
could not maintain a positive angular velocity for the whole range of motion during the concentric phases at the end of the test, as may be seen in Figure 2, where the angular velocity frequently changed sign in the higher part of the range of motion. This was not the case during the eccentric phases, where a decreasing resistance towards the gravitation could only increase the velocity. This implied that the work during the concentric phases, defined as the part of the range of motion at which the angular velocity was positive, decreased more than the work during the eccentric phases. Accordingly, in the present study, there were increases in torque during both phases, but a decrease in work during the concentric phases. A more pronounced decrease in work than in peak torque was also found by Fugel-Meyer et al. [23] and Gerdle and Langstrom [5] for pure concentric contractions in the plantar flexors. Verdonck et al. [24] also found a smaller decrease in peak torque than in work for combined eccentric-concentric contractions in the rcctus femoris and vastus lateralis muscles. As the torque required for further rise at a higher angle in the range of motion is lower than the torque that is required at a lower angle in the range of motion, the reduced range of motion was not a consequence of lower torque production than needed for the rise. An explanation might be that, at the end of the test, the eccentric pre-activation might render a reduced effect on the concentric action as the RMS values, and therefore the activity level, decreased. However, no significant increase in coupling time was seen in this test that would lead to reduced capacity to utilize elastic energy from the pre-activation. No change in RMS and a decrease in MPF during the concentric phases indicated development of muscular fatigue in the concentric phase according to the definition by Lindstrom et al. [9]. Decreases in both RMS and MPF in the eccentric phase showed a decreasing activity level [13] during the eccentric phases. Svantesson et al. [3] also found decreases in MPF in the m. gastrocnemius medialis and m. soleus during both phases, together with no changes in RMS during the concentric phases and decreases in RMS in the eccentric phases in a heel-rise test performed by young girls. Differences in RMS levels between the eccentric and concentric phases seen in Figure 8 might be due to different activation levels in the two phases. This is because, especially at the beginning, the test is not maximal, as is seen from the increasing torque level. An explanation for this difference could be that a higher firing rate during the concentric phases might increase the RMS values of the EMG [25]. Decrease in force from a submaximal level did not seem to be a suitable definition of fatigue during dynamic actions with no preset range of motion, as in the heel-rise test. The torque did not decrease, but
350
U. &terberglClinical
Biomechanics 13 (1998) 344-350
reduced performance was seen as a decrease in work, which was also indicated by the decreasesin MPF. 5. Conclusions There is a reduced capacity to maintain the same work output during all cycles, despite an increased torque production. This implies that the heel-rise test does reflect muscle endurance more than it reflects the strength of the calf muscle. Acknowledgements This study was supported by grants from the University of Goteborg, Swedish Medical Research Council (project no. 03888), Medical Faculty at Goteborg University, Swedish Center for Research in Sports, Goteborg College of Health Sciences, IngaBritt and Arne Lundberg’s Foundation and Greta and Einar Asker Foundation. References 111Bosco C, Tarkka I, Komi PV. Effect of elastic energy and
myoelectrical potentiation of triceps surae during stretch-shortening cycle exercise. Int J Sports Med 1982;3:137-140. PI Lunsford BR, Perry J. The standing heel-rise test for ankle plantar flexion: criterion for normal. Physical Therapy 1995;75(8):49-53. 131 Svantesson U, Gsterberg U, Thomee R, Grimby G. Muscle fatigue in a standing heel-rise test. Stand J Rehab Med 1998; in press. [41 Edwards RHS. Human muscle function and fatigue: physiological mechanisms. In: Ciba Foundation Symposium, vol. 82. London: Pitman Medical, 1981:1-18. PI Gerdle B, LIngstrom M. Repeated isokinetic plantar flexions at velocities. Physiol Stand different angular Acta 1987;130:495-500. 161Nicol C, Komi P, Marconnet P. Fatigue effects of marathon running on neuromuscular performance. I. Changes in muscle force and stiffness characteristics. Stand J Med Sci Sports 1991;1:10-17. 171 Komi PV, Tesch P. EMG frequency spectrum, muscle structure and fatigue during dynamic contractions in man. Eur J Appl Physiol 1979;42:41- 50. 181Cavagna G, Dusman B, Margaria R. Positive work done by a previously stretched muscle. J Appl Physiol 1968;24:21-32. [91 Lindstrom L, Magnusson R, Petersen I. Muscular fatigue and action potential conduction velocity changes studied with frequency analysis of EMG signals. Electromyography
1970;4:341-353.
1101Lindstrom L, Kadefors R, Petersen I. An electromyographic index for localized muscle fatigue. J Appl Physiol 1977;43:750-754. WI Tesch P, Komi PV, Jacobs I, Karlsson J, Viitasalo JT. Influence of lactate accumulation of EMG frequency spectrum during repeated concentric contractions. Acta Physiol Scan 1983;119:61-7. WI Hagg G. Interpretation of EMG spectral alterations and alteration indexes at sustained contraction. American Physiological Society 1992;73:1211-1217. u31 Gerdle B, Eriksson N, Brundin L. The behavior of the mean power frequency of the surface electromyogram in biceos brachii with increasing force and during fatigue. With special regard to the electrode distance. Electromyogr Clin Neurophysiol 1990;30:483-489. v41 Thomee R, Grimby G, Svantesson U, Gsterberg U. Quadriceps muscle performance in sitting and standing in young women with patellofemoral pain syndrome and young healthy women. Scan J Med Sci Sports 1996;6:233-241. [151Schuldt K, Ekholm J, Nemeth G, Arborelius U, HarmsRingdahl K. Knee load and muscle activity during exercises in rising. Stand J Rehabil Med Suppl 1983;2:174-188. WI Bobbert M, Huijing P, van Ingen Schenau G. A mode1 of the human triceps surae muscle-tendon complex applied to jumping. J Biomechanics 1986;11:887-898. u71 Elftman H. Forces and energy changes in the leg during walking. Am J Physiol 1939;125:339-356. WI Svantesson U, Grimby G, Thomee R. Potentiation of concentric plantar flexion torque following eccentric and isometric muscle actions. Acta Physiol Stand 1994;152:287-293. [I91 Svantesson U, Grimby G. Stretch-shortening cycle during plantar flexion in young and elderly women and men. Eur J Appl Physiol 1995;71:381-385. 1201de Graaf J, Bobbert M, Tetteroo W, van Ingen Schenau G. Mechanical output about the ankle in counter movement jumps and jumps with extended knee. Human Movement Science 1987;6:333-347. PI Lijscher W, Cresswell A, Thorstensson A. Significance of Ia-afferent input to the a-motoneuron pool for enhancement of tremor during fatigue. Journal of Neurophysiology (submitted). P21Bobbert M, van Ingen Schenau G. Isokinetic plantar flexion: experimental results and model calculations. J Biomechanics 1990;2:105-119. v31 Fugel-Meyer A, Gerdle B, LBngstrom M. Characteristics of repeated isokinetic plantar flexions in middle-aged and elderly subjects with special regard to muscular work. Acta Physiol Stand 1985;124:213-222. [24] Verdonck A, Frobose I, Hardelauf U, Guttge C. Contraction patterns during isokinetic eccentric and concentric contractions after anterior cruciate ligament injury. Int J Sports Med 1994;15(Suppl 1):60-63. [25] Basmaijan J, deLuca C. Muscles Alive. Their Functions Revealed by Electromyography. Baltimore, MD: Williams and Wilkins 1985.
ELSEVIER
Torque, work and EMG development in a heel-rise test U. osterberg, U. Svantesson”, H. Takahashi, G. Grimby Department of Rehabilitation
Medicine, Gijteborg Universiry Guldhedsgatan 19, 413 45 Giiteborg, Sweden
Received5 August 1997;accepted3 December1997
Abstract Objective. To study the fatigue process in the calf muscle during a standing heel-rise test. Design. Heel-rises were performed on a force plate measuring development of torque in the ankle joint. Background. The heel-risetest is a commonlyemployedclinical test to evaluatethe function of the calf muscleby counting the
number of heel-rises before exhaustion. Development of fatigue during other eccentric-concentric actions has earlier been studied as decreasesin force, work and EMG. Methods. Ten healthy males (mean age 25 yr) participated in the study. Torque and work were calculated using measurements from a force plate. Mean power frequency and root-mean-square of the EMG in the triceps surae were evaluated separately in the eccentric and concentric phases. Results. Increases of mean torque during both the eccentric and concentric phases were found. Work performed decreased during the concentric phases due to decreased range of motion. No changes in root-mean-square and decreasesin mean power frequency during the concentric phases indicated development of muscular fatigue, whereas decreasesin both root-mean-square and mean power frequency during the eccentric phases indicated decreasing muscular activity. Conclusions. Accordingly, the limiting factor for the heel-rise test was not loss of muscle force at the range of motion used, but rather a failure to maintain the initial range of motion owing to muscle fatigue. Relevance This method of calculating torque development in the ankle joint provides an opportunity to study the fatigue process in terms of performance. The results show that the heel-rise test reflects muscle endurance rather than strength of the calf muscle. 0 1998 Elsevier Science Ltd. All rights reserved. Keywords:
Mean powerfrequency;Root-mean-square; Calf muscle;Fatigue
1. Introduction
The heel-rise test consists of eccentric-concentric muscle actions of plantar flexion. It is often used in clinical conditions for evaluating the function of the calf muscle as these muscle actions are often used in such physical activities as walking, running and jumping.
The
eccentric
pre-activation
renders
an
increase in performance during the concentric phase, partly due to the utilization of the elastic energy stored during the pre-activation [l]. Recently, the heel-rise test has been evaluated by Lundsford and Perry [2] concerning the number of performed heel-rises repre*Correspondenceand reprint requests to: UIIa Svantesson, Departmentof Rehabilitation Medicine, Guldhedsgatan19, 413 45 Giiteborg,Sweden.E-mail: [email protected]
senting normal capacity. Svantesson et al. [3] investigated differences in fatigue development between the m. gastrocnemius medialis and m. soleus during the test, together with an estimation of mechanical work. However, no studies have been made with measurements of the development of torque and work during the heel-rise test to evaluate further the outcome of the test. According to Edwards [4], fatigue during isometric activity is defined as a failure to maintain the required or expected force. Development of fatigue has been evaluated in dynamic conditions using other variables as well, such as decrease in work [5], increase in impact force [6] or decreases in the mean power frequency (MPF) of the electromyographic (EMG) record [7]. In isokinetic studies, Gerdle and LAngstriim [5] found that measurements
0268-0033/98/$19.00 + 0.000 1998ElsevierScienceLtd. All rights reserved. PII: SO268-0033(98)00100-9
of fatigue
should
be made of the
U. &e&erg
et aLlClinical
Biomechanics
decrease in work and not peak torque, whereas differences in motor unit behavior during fatigue are of greater importance for work than for peak torque. The development of fatigue can also reduce the effect of the eccentric pre-activation if the coupling time becomes too long, as this might result in a loss of stored elastic energy by dissipation as heat [8]. One method to evaluate muscular fatigue is to study the alterations of the EMG power spectrum by means of the MPF, as a decrease in MPF is mainly a result of the decrease in the action potential conduction velocity [9, lo]. This decrease in action potential conduction velocity has been found to be related to the lactate concentration in the muscle [ll]. As the conditions in the muscle are similar in both the eccentric and concentric phases, as a concentration of metabolites due to fatigue in one phase also affects the other phase, it might be difficult to observe differences in MPF between the two phases. However, other factors might also influence the MPF, such as muscle activity level, action potential modification, firing rate, synchronization, skin temperature and additional recruitment of motor units at fatigue (for a review, see Hagg [12]). Highly significant linear correlations have been found between the signal amplitude of the EMG and the force developed [13] during fatiguing exercises. Gerdle et al. [13] also found correlations between MPF and signal amplitude of the EMG. In short, a decrease in MPF along with an increase or no change in the signal amplitude of the EMG is considered to be a practical measurement of muscular fatigue. Earlier studies were made on torque development in the knee during static conditions on a force plate by Thomee et al. [14] and Schuldt et al. [15], the latter emphasizing the importance of measuring the torque caused by anterior-posterior forces. In a study by Bobbert et al. [16], torque in the ankle joint during jumping was calculated, omitting factors arising from angular acceleration, using measurements from a force plate. The purpose of this study was to measure the torque influencing the ankle joint during a standing heel-rise test from the force measurements from a force plate and to calculate work during the test. The development of torque and work, as well as EMG, from the calf muscle were also to be investigated for evaluation of the fatigue process during the heel-rise test. 2. Methods 2.1. Subjects
Ten healthy male students of physiotherapy with a mean age of 25 yr (SD 3 yr: range 21-29 yr), a mean weight of 76 kg (SD 7 kg, range 67-89 kg) and a mean
13 (1998)
345
344-350
height of 179 cm (SD 3 cm, range 164-189 cm) participated in the study. Informed consent was obtained from the participants, and the study was approved by the Ethics Committee of the Faculty of Medicine, Goteborg University, Sweden. 2.2 Methods
Before the test, the subjects warmed up for 5 min with submaximal bicycling. The heel-rise test was performed with the subject, who wore standardized shoes and stood on a force plate (AMTI OR6-5, Newton, MA) and a lo” tilted wedge. For balance, the finger tips touched the wall at the height of the shoulders. The subjects were instructed to lift the heel as high as possible, with a straight knee, until no further heel-rises could be performed due to exhaustion. The force signals were then processed by an AMTI SGA6-4. A metronome was set to 92 beats per minute in order to exceed an angle velocity of approximately 60 deg s- 1 (1.05 rad ss’) for at least 0.41 s. For the measurement of range of motion of the ankle joint, a goniometer (Penny and Giles Biometrics Ltd, Gwent, UK) was placed on the lower lateral side of the leg and the lateral side of the foot. 90” in the range of motion was defined as when the foot was perpendicular to the leg, and 80” was in the direction of dorsiflexion. The signals from the amplified force signal and the angle of the ankle joint were sampled with a frequency of 100 Hz by an IBM Personal Computer AT. A study by Bobbert et al. [16] used a formula for calculation of torque in the ankle joint by Elftman [17]. They found that maximal error due to angular acceleration was 7.5 N at an angular velocity of 5.24 rad s ’ and even less at lower velocities; therefore, they omitted the part of the formula depending on angular acceleration. As maximal angular velocity was to be approximately 1.05 rad s -’ and the forces were supposed to exceed approximately 500 N, this part of the formula was omitted in this study as well. Accordingly, the torque development during the test on the ankle joint was calculated with the formula M, = F7Ax+ F,Az
where F, was given from the force plate as the vertical force and F, as the horizontal force in the anteriorposterior direction. The distances 4~ and AZ were calculated as Ax=x,?-
rcos lO”+t sin 1O”+1cos 10”
+ 1cos((/I-- 90”) AZ= -(x0-+,)
tan lO”+tcos lO”+rsin
+ I sin 10” + 1sin(& - 90’)
10”
L! &terberglClinical
346
Biomechanics 13 (1998) 344-350
where ,xtpis the center of pressure on the force plate in the anterior-posterior direction during the test, r is the distance in the anterior-posterior direction between the dorsal part of the heel and the ankle joint, t is the distal-proximal distance between the wedge and the ankle joint, I is the distance in the anterior-posterior direction between the dorsal part of the heel and the metatarsophalangeal joints and x0 is the coordinate in the anterior-posterior direction for the dorsal part of the heel. C$is the angle of the ankle joint given by the goniometer (Figure 1). The total amount of work during the whole test from tstartto hop was calculated for both the eccentric and the concentric phases according to the formula hop W=
X M,o,At r=lU,,t
where M’ is the torque and o, is the angular velocity in each sample (t) and At is the sampling time (0.01 s). The angle and velocity were filtered by a moving average filter of an order of five. According to Bobbert et al. [16], a low pass filter with a cut-off frequency of 5 Hz was not sufficient for measurements of high angular velocities (5.24 rad SC’); therefore, they used a
cut-off frequency of 16 Hz. The sampling frequency of 100 Hz in the present study, with angular velocities of approximately 60 deg s-’ (1.05 rad ss’) and the filter as above, was taken to be sufficient for the purposes of this study. Using the preset beat of 92 metronome beats per minute, an angular velocity of 60 deg SC’was supposed to be exceeded. However, this velocity was kept, as seen in Figure 3, for only short periods of time. To be able to make a frequency analysis of the EMG, it was necessary for the time during which the analysis was made to be longer than these periods. Angular velocities exceeding approximately 40 deg ss’ were thus accepted during the time period during which the frequency analysis was made. The beginning of each phase was defined as when the angular velocity did not change sign until an angular velocity of at least 40 deg so’ was reached, and the end of each phase was defined as when the angular velocity changed sign. This is the definition of a concentric or an eccentric phase used in this study. Total work for both the concentric and eccentric phases for each heel-rise was calculated and plotted. The change in work was then calculated as the Torque
0
(Nm)
li
II
18
Ang 1 e (deg)
Fig. 1. Position of the foot in the heel-rise test. R is the distance in anterior-posterior direction between the heel and the ankle joint, t is the horizontal distance between ground and ankle joint, 1 is the distance in the anterior-posterior direction between heel and metatarsophalangeal joints and .x0 is the coordinate in the anteriorposterior direction for the heel.
Fig. 2. Trace of the torque and of the angle of the ankle joint during the test as a function of time.
l% &e&erg
et al./Clinical
347
Biomechanics 13 (1998) 344-350
percentage difference between the first four and last four heel-rises. The mean torque values for both the eccentric and concentric phases were extracted and plotted. The change in mean torque was then calculated as the percentage difference between the first four and last four heel-rises for both the eccentric and concentric phases. The coupling time was defined as the time elapsed between the eccentric and the concentric phases. The mean coupling time for the first four and last four heel-rises was calculated. EMG activity was recorded with AgiAgCl surface electrode discs with a diameter of 9 mm (Red Dot, 2239 monitoring electrodes, 3M Medica, Borken, Westfalen, Germany). After the skin was shaved and scrubbed with alcohol, two electrodes at a distance of approximately 30 mm were secured on the bellies of the gastrocnemius medialis and the soleus muscle in a distal-proximal direction. The reference electrodes were placed on the bony part of the medial and lateral parts of the knee. The EMG signal was pre-amplified with a gain of 1000 and thereafter band-pass-filtered between 7 and 490 Hz by a KC-EMG (Chattanooga Group, Inc., TN, USA). The angle and the EMG were sampled at a frequency of 1250 Hz on a Macintosh computer with software developed in Lab View (National Instruments Corporation, Austin, TX, USA) by Punos Electronic AB, Goteborg, Sweden. The EMG signal was then additionally band-pass-filtered using a Butterworth filter with cut-off frequencies of 10 and 300 Hz. The part of the EMG signal for which an angular velocity above 40 deg ss’ was obtained was separately derived for the concentric and eccentric phases by observing the angle of the ankle joint from the goniometer. The same range of motion was derived for all heel-rises in the same test. For each such part, the average of the amplitude of the rectified signal was calculated using the root-mean-square (RMS) method. The fast Fourier transform method was used to obtain the MPF. The development of the MPF and RMS was calculated with simple regression analyses and then reported as percentage decrease from the initial value. This was done separately for the concentric and eccentric phases.
3. Results The total number of heel-rises performed was, on average, 36 (SEM, 2). The pattern of the torque and work development during the test are seen from the recordings for one test person (Figures 2 and 3). The maximal angle of the range of motion decreased by 10” (SEM, 2”) during the test starting with a range of motion of 47” (SEM, Zo). The velocity during the test exceeded the expected 60 deg ss ‘, but only for a small range of motion. The mean time that elapsed between the eccentric phase and the concentric phase, i.e. the coupling time, was 300 ms for the first part of the test and 370 ms for the last part of the test, with no significant difference between the beginning and the end of the test. As seen in Figure 4, the mean torques in both the eccentric and the concentric phases showed significant increases. There was an increase in peak torque of 16% (SEM, 3%) in the eccentric phase and of 22% (SEM, 3%) in the concentric phase. The total amount of work performed during the test was 6624 J (SEM, 467 J) during the eccentric phases and 6631 J (SEM, 472 J) during the concentric phases. Power
(WI
t (59 0
6
Velocity
II
IO
24
24
38
42
u
El
q
I8
a
e
16
u
48
51
10
(deg/s)
2.3. Statistics Mean values and standard deviation (SD) were used for the study population and mean values and standard errors of the mean (SEM) were calculated for measurements with conventional methods. Wilcoxon’s one-sample non-parametric test was used for differences. The level of significance was set at 0.05.
0
I
12
Fig. 3. Trace of the power and of the angle velocity of the ankle joint during the test as a function of time.
348
U dsterherglClinica1 Biomechanics 13 (1998) 344-350
250 80.
20 OJ 0
5
10
15
20
25
30
35
40
Ol 0
45
5
10
15
20
25
30
Fig. 4. Mean torques for both eccentric (0) and concentric actions as a function of number of heel-rises.
(0)
The development of work showed no significant change during the eccentric phases (8% (SEM, 6%)), but a significant decrease in the concentric phases of 20% (SEM, 6%) (Figure 5). All decreasesin MPF were significant. In the eccentric phase the decreaseswere 32% (SEM, 6%) for the gastrocnemius muscle and 27% (SEM, 4%) for the soleus muscle. The decreases in the concentric phase were 16% (SEM, 3%) for the gastrocnemius muscle and 10% (SEM, 3%) for the soleus muscle (Figures 6 and 7). All decreases in RMS in the eccentric phases were significant, but there were no changes in the concentric phases (Table 1). In the eccentric phases, the decreases of RMS were 17% (SEM, 7%) for the gastrocnemius muscle and 35% (SEM, 11%) for the soleus muscle (Figure 8).
The present study showed increases in mean torque in both phases, a decrease in work during the concentric phases and no change in work during the eccentric phases. There was no change in RMS and a decrease
45
in MPF during the concentric phases, but decreases in both RMS and MPF in the eccentric phases were found. The torque curves from one single heel-rise showed a similar pattern to that seen in torque curves obtained from a dynamometer with maximal eccentric-concen-
A :
60.
it x
40. 20.
OJ
.
0
5
10
15
20
25
30
35
40
8
45
Number of heel-rises Fig. 7. MPF for the soleus muscle for both eccentric (m) and concentric (0) actions as a function of the number of heel-rises.
Table 1 P-values for changes during the heel-rise test
0
160.
P-value Coupling time Torque Work
60
MPF
40
Ecc Con Ecc Con Ecc Con
0
40
Fig. 6. MPF for the gastrocnemius muscle for both eccentric (0) and concentric (0) actions as a function of the number of heel-rises.
4. Discussion
r F
35
Number of heel-rises
Number of heel-rises
180.
i
5
10
15
20
25
30
35
40
45
Number of heel-rises Fig. 5. Total work for each heel-rise for both eccentric (0) and concentric (0) actions as a function of the number of heel-rises.
RMS
Ecc Con
Gastro Soleus Gastro Soleus Gastro Soleus Gastro Soleus
0.1688 0.0077 0.0051 0.2411 0.0189 0.005 1 0.0050 0.0069 0.0280 0.0364 0.0217 0.7211 0.2619
L! herberg
et al.lClinical
900 800 700 I 6001
0
5
10
15
20
25
30
35
40
45
Number of heel-rises 8. RMS both for the soleus muscle for eccentric (D) and concentric (o) actions and for the gastrocnemius muscle for eccentric (0) and concentric (0) actions as a function of the number of heelrises.
tric actions [18]. Whereas there were high torque values at the end of the eccentric phase and the beginning of the concentric phase, low values were seen at the end of the concentric phase and the beginning of the eccentric phase. This was probably because of the short lever arm in the highest position in the heel-rise test and because there was a reduced performance ability with a shorter muscle in the dynamometer test. Figure 4 shows mean isokinetic torque values during the concentric phases of approximately 220 N m. These torque values appear to be higher than during maximal voluntary concentric peak torques in eccentric-concentric actions with an angular velocity of 120 deg SC’ on a dynamometer of 88 N m, which were found in young men by Svantesson and Grimby [19]. In contrast, de Graaf et al. [20] found peak torques in jumps of 194 N m, and Liischer et al. [21] found isometric torques on a dynamometer of 196 N m. The discrepancy between the different torques of the ankle joint might be explained by the findings of Bobbert and van lngen Schenau [22], who demonstrated that, in simulated isokinetic plantar flexion at high angular velocities, the duration of the movement was too short for the torque to become maximal. However, in the present study, the velocity is rather low and the range of motion high. The range of motion of 47” at the beginning of the test was not far from the maximal range of motion of the ankle joint, which is 61” for males between 40 and 44 yr, as found by Fugl-Meyer et al. [23].
As seen in Figure 2, the mean torque increased during the test. This increase probably occurred as a result of increasing effort to obtain a required range of motion, but might also be due to a slow displacement during the test of the center of pressure of the foot in the anterior direction, leading to a longer lever-arm and, therefore, the higher torque values observed in some subjects. Despite increasing effort, the subjects
Bion~echanics 13 (1998) 344-350
349
could not maintain a positive angular velocity for the whole range of motion during the concentric phases at the end of the test, as may be seen in Figure 2, where the angular velocity frequently changed sign in the higher part of the range of motion. This was not the case during the eccentric phases, where a decreasing resistance towards the gravitation could only increase the velocity. This implied that the work during the concentric phases, defined as the part of the range of motion at which the angular velocity was positive, decreased more than the work during the eccentric phases. Accordingly, in the present study, there were increases in torque during both phases, but a decrease in work during the concentric phases. A more pronounced decrease in work than in peak torque was also found by Fugel-Meyer et al. [23] and Gerdle and Langstrom [5] for pure concentric contractions in the plantar flexors. Verdonck et al. [24] also found a smaller decrease in peak torque than in work for combined eccentric-concentric contractions in the rcctus femoris and vastus lateralis muscles. As the torque required for further rise at a higher angle in the range of motion is lower than the torque that is required at a lower angle in the range of motion, the reduced range of motion was not a consequence of lower torque production than needed for the rise. An explanation might be that, at the end of the test, the eccentric pre-activation might render a reduced effect on the concentric action as the RMS values, and therefore the activity level, decreased. However, no significant increase in coupling time was seen in this test that would lead to reduced capacity to utilize elastic energy from the pre-activation. No change in RMS and a decrease in MPF during the concentric phases indicated development of muscular fatigue in the concentric phase according to the definition by Lindstrom et al. [9]. Decreases in both RMS and MPF in the eccentric phase showed a decreasing activity level [13] during the eccentric phases. Svantesson et al. [3] also found decreases in MPF in the m. gastrocnemius medialis and m. soleus during both phases, together with no changes in RMS during the concentric phases and decreases in RMS in the eccentric phases in a heel-rise test performed by young girls. Differences in RMS levels between the eccentric and concentric phases seen in Figure 8 might be due to different activation levels in the two phases. This is because, especially at the beginning, the test is not maximal, as is seen from the increasing torque level. An explanation for this difference could be that a higher firing rate during the concentric phases might increase the RMS values of the EMG [25]. Decrease in force from a submaximal level did not seem to be a suitable definition of fatigue during dynamic actions with no preset range of motion, as in the heel-rise test. The torque did not decrease, but
350
U. &terberglClinical
Biomechanics 13 (1998) 344-350
reduced performance was seen as a decrease in work, which was also indicated by the decreasesin MPF. 5. Conclusions There is a reduced capacity to maintain the same work output during all cycles, despite an increased torque production. This implies that the heel-rise test does reflect muscle endurance more than it reflects the strength of the calf muscle. Acknowledgements This study was supported by grants from the University of Goteborg, Swedish Medical Research Council (project no. 03888), Medical Faculty at Goteborg University, Swedish Center for Research in Sports, Goteborg College of Health Sciences, IngaBritt and Arne Lundberg’s Foundation and Greta and Einar Asker Foundation. References 111Bosco C, Tarkka I, Komi PV. Effect of elastic energy and
myoelectrical potentiation of triceps surae during stretch-shortening cycle exercise. Int J Sports Med 1982;3:137-140. PI Lunsford BR, Perry J. The standing heel-rise test for ankle plantar flexion: criterion for normal. Physical Therapy 1995;75(8):49-53. 131 Svantesson U, Gsterberg U, Thomee R, Grimby G. Muscle fatigue in a standing heel-rise test. Stand J Rehab Med 1998; in press. [41 Edwards RHS. Human muscle function and fatigue: physiological mechanisms. In: Ciba Foundation Symposium, vol. 82. London: Pitman Medical, 1981:1-18. PI Gerdle B, LIngstrom M. Repeated isokinetic plantar flexions at velocities. Physiol Stand different angular Acta 1987;130:495-500. 161Nicol C, Komi P, Marconnet P. Fatigue effects of marathon running on neuromuscular performance. I. Changes in muscle force and stiffness characteristics. Stand J Med Sci Sports 1991;1:10-17. 171 Komi PV, Tesch P. EMG frequency spectrum, muscle structure and fatigue during dynamic contractions in man. Eur J Appl Physiol 1979;42:41- 50. 181Cavagna G, Dusman B, Margaria R. Positive work done by a previously stretched muscle. J Appl Physiol 1968;24:21-32. [91 Lindstrom L, Magnusson R, Petersen I. Muscular fatigue and action potential conduction velocity changes studied with frequency analysis of EMG signals. Electromyography
1970;4:341-353.
1101Lindstrom L, Kadefors R, Petersen I. An electromyographic index for localized muscle fatigue. J Appl Physiol 1977;43:750-754. WI Tesch P, Komi PV, Jacobs I, Karlsson J, Viitasalo JT. Influence of lactate accumulation of EMG frequency spectrum during repeated concentric contractions. Acta Physiol Scan 1983;119:61-7. WI Hagg G. Interpretation of EMG spectral alterations and alteration indexes at sustained contraction. American Physiological Society 1992;73:1211-1217. u31 Gerdle B, Eriksson N, Brundin L. The behavior of the mean power frequency of the surface electromyogram in biceos brachii with increasing force and during fatigue. With special regard to the electrode distance. Electromyogr Clin Neurophysiol 1990;30:483-489. v41 Thomee R, Grimby G, Svantesson U, Gsterberg U. Quadriceps muscle performance in sitting and standing in young women with patellofemoral pain syndrome and young healthy women. Scan J Med Sci Sports 1996;6:233-241. [151Schuldt K, Ekholm J, Nemeth G, Arborelius U, HarmsRingdahl K. Knee load and muscle activity during exercises in rising. Stand J Rehabil Med Suppl 1983;2:174-188. WI Bobbert M, Huijing P, van Ingen Schenau G. A mode1 of the human triceps surae muscle-tendon complex applied to jumping. J Biomechanics 1986;11:887-898. u71 Elftman H. Forces and energy changes in the leg during walking. Am J Physiol 1939;125:339-356. WI Svantesson U, Grimby G, Thomee R. Potentiation of concentric plantar flexion torque following eccentric and isometric muscle actions. Acta Physiol Stand 1994;152:287-293. [I91 Svantesson U, Grimby G. Stretch-shortening cycle during plantar flexion in young and elderly women and men. Eur J Appl Physiol 1995;71:381-385. 1201de Graaf J, Bobbert M, Tetteroo W, van Ingen Schenau G. Mechanical output about the ankle in counter movement jumps and jumps with extended knee. Human Movement Science 1987;6:333-347. PI Lijscher W, Cresswell A, Thorstensson A. Significance of Ia-afferent input to the a-motoneuron pool for enhancement of tremor during fatigue. Journal of Neurophysiology (submitted). P21Bobbert M, van Ingen Schenau G. Isokinetic plantar flexion: experimental results and model calculations. J Biomechanics 1990;2:105-119. v31 Fugel-Meyer A, Gerdle B, LBngstrom M. Characteristics of repeated isokinetic plantar flexions in middle-aged and elderly subjects with special regard to muscular work. Acta Physiol Stand 1985;124:213-222. [24] Verdonck A, Frobose I, Hardelauf U, Guttge C. Contraction patterns during isokinetic eccentric and concentric contractions after anterior cruciate ligament injury. Int J Sports Med 1994;15(Suppl 1):60-63. [25] Basmaijan J, deLuca C. Muscles Alive. Their Functions Revealed by Electromyography. Baltimore, MD: Williams and Wilkins 1985.