Electroencephalography and clinical Neurophysiology , 85 (1992) 166-172 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0924-980X/92/$05.00
166
E L M O C O 91661
The influence of muscle length on muscle fibre conduction velocity and development of muscle fatigue Lars A r e n d t - N i e l s e n
a
Nikolai G a n t c h e v b and T h o m a s Sinkja~r a
Laboratory for Motor Control, Department of Medical Informatics, Aalborg University, DK-9220 Aalborg E (Denmark), and b Central Laboratory of Biophysics, Bulgarian Academy of Sciences, Sofia 1113 (Bulgaria) (Accepted for publication: 9 D e c e m b e r 1991)
Summary The influence of muscle (vastus lateralis) length on the muscle fibre conduction velocity (MFCV) and on muscle fatigue was studied in 8 healthy volunteers. In experiment 1, the electromyographic (EMG) responses were evoked by electrical stimulation of the motor point and recorded by a surface electrode array aligned along the muscle fibre direction. The M F C V (determined by cross-correlation) was m e a s u r e d at knee flexions of 5 ° (full extension), 45 °, 90 ° and 120° with 3 different extension torques. The M F C V declined with increasing muscle length and increased with increasing background torque at knee flexions from 5 ° to 90 °. From 90 ° to 120 ° knee flexion of M F C V tended to increase. In experiment 2, the E M G activity at a static fatiguing contraction (80% MVC) was measured at 45 ° and 90 ° knee flexion. The E M G was measured until the subject gave up contracting the muscle (endurance). The largest increase in the RMS amplitude and the fastest decreases in the mean power frequency (MPF) and M F C V were found at 90 ° flexion. The M V C at 45 ° knee flexion was 35% lower than at 90 ° and the time until endurance was approximately twice as long for the 45 ° contraction. The results indicate that muscle length is an important parameter for the propagation velocity of action potentials and for the development of static muscle fatigue.
Key words: Muscle; Length; Fatigue; Conduction velocity; EMG; Propagation
Muscle fibre length in situ can change by more than 100% when a joint is rotated from the fully extended to the fully flexed position (Haines 1932, 1934). Theoretically and experimentally it has been found that the muscle fibre conduction velocity (MFCV) increases when the diameter increases (Katz 1947; Hfikansson 1957; Kossev et al. 1991). Studies in which the MFCV is used clinically have begun to appear (for review see Arendt-Nielsen and Zwarts 1989) and it is therefore important to reach conclusions on the various factors affecting the MFCVo It has been suggested that the MFCV either increases (Morimoto 1986), decreases (Wilska and Varjoranta 1940; H,~kansson 1957) or remains unchanged (Martin 1954; Hodgkin 1954) when the muscle is shortened. The first aim of the present study was to measure the muscle fibre conduction velocity (MFCV) from the human vastus lateralis muscle at different muscle
Correspondence to: Dr. Lars Arendt-Nielsen, Laboratory for Motor Control, D e p a r t m e n t of Medical Informatics, Aalborg University, Fredrik Bajersvej 7D, DK-9220 Aalborg E (Denmark). Tel.: + 4 5 98158522, ext. 4954; Fax: +45 98154008.
lengths to clarify whether the MFCV increases, decreases or remains unchanged when the muscle is shortened. The muscle length also seems to be of importance to the way in which the muscle is fatigued as the endurance is enhanced at short muscle lengths (Aljure and Borrero 1968; Fitch and McComas 1985; McKenzie and Gandevia 1987). The twitch contraction and relaxation times are decreased in short muscles (Gandevia and McKenzie 1988) but it has been shown (Gandevia and McKenzie 1988) that the motoneurones increase their firing rates until fusion is achieved in short muscles. Even though fusion can be obtained, the maximum voluntary force declined by 21-49% at short muscle lengths due to reduced twitch forces (Gandevia and McKenzie 1988). During dynamic muscle load, the muscle length changes continuously, but the importance of length changes as the development of fatigue is unknown. The electromyogram (EMG) can be used to clarify some of these aspects. The second aim of the present study was to monitor EMG parameters (MFCV, mean power frequency, RMS amplitude) during fatiguing voluntary contractions maintained at different muscle lengths.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE Methods and materials
Volunteers Eight healthy male volunteers participated (mean age 23 years, range 20-30 years). Informed consent was obtained, and the Helsinki Declaration was respected.
Experimental set-up During the recording session, the subject was seated in a test chair. The dorsal part of the ankle was firmly fastened against a rubber block with straps around the ankle. The block was mounted on a shaft which again was attached to a strain gauge. The torque generated by the knee extensor muscles was displayed on an oscilloscope and collected on computer. The axis of rotation of the shaft was aligned with the axis of rotation of the knee joint. A safety belt was strapped around the belly and the back of the test chair. The hip joint was flexed 90 ° . All experiments were performed under isometric conditions. The maximum voluntary isometric knee extension torque was measured at 90 ° knee flexion as the largest of 3 brief maximal extensions. The skin t e m p e r a t u r e over the vastus lateralis was maintained at 34.5°C by an infrared heater. It is important to maintain the t e m p e r a t u r e constant as the M F C V is highly affected by changes in temperature. The t e m p e r a t u r e of 34.5°C was normally 2-3°C above the normal skin temperature.
Data collection The E M G electrodes were placed in a bipolar array (centre electrode common), 15 m m apart, and placed distal to the motor point and parallel to the fibre axis of the vastus lateralis muscle. Electrical surface stimulation was used to determine the position of its motor point. The stimulus current was increased until maximum E M G amplitude was reached. The correct electrode alignment was obtained when the maximal delay between the two E M G s evoked by electrical stimulation over the motor point had been reached. Effective stabilization of the electrode was achieved by adhesive tape and external compression. The E M G signals were filtered (20-500 Hz), amplified, sampled (2000 Hz) and stored on disc. Before the experiment, the skin was lightly abraded and cleaned with isopropyl alcohol. Electrode gel was applied sparingly to the 3 Ag-AgC1 surface electrodes. During the experiments, the E M G signal and torque were stored in a computer for later analysis.
Data analysis The E M G signals were divided into 250 msec time intervals. Cross-correlation between the 2 E M G signals was performed (voluntary activity or electrically evoked
167 compound action potentials) and provided the electrodes were positioned correctly, a high (R = 0.8-0.99) correlation between the two signals was obtained. The peak in the correlogram was displaced from time zero by a time lag reflecting the conduction time between the two pairs of electrodes. The mean power frequency (MFP) was calculated as the ratio of the first order and zero order moment of the E M G power spectrum. The RMS amplitude was calculated as the square root of the squared mean amplitude of the E M G epochs. As a function of time, the changes in the parameters were estimated by the slopes of the linear regression lines. Wilcoxon's test was used for statistical analysis and P < 0.05% was regarded as statistically significant.
Protocol Experiment 1.
The M F C V was determined from the compound potential evoked by electrical motor point stimulation at 4 different knee angles (5, 45, 90 and 120 °) and 3 different contraction levels (0, 25, 50% maximum voluntary contraction (MVC) was determined for 90 ° knee flexion). The subject was asked to hit a target line on the oscilloscope and maintain the torque at the given percentage of MVC. Constant extension torques were used at the different knee flexion angles. A knee angle of 5 ° corresponded to full extension. The angles and torques were presented in random order. For each condition, the maximum delay between the electrically evoked compound muscle action potential was determined. When maximum delay was obtained a number (5-10) of compound potentials were immediately collected for cross-correlation analysis. The mean M F C V for the 5 - 1 0 potentials was calculated. Experiment 2. On two occasions, separated by at least 48 h, the E M G and isometric knee extension torque were measured at knee extensions of 90 and 45 °. The initial torque was 80% MVC. Maximum voluntary contractions were determined individually for 45 and 90 ° knee angle. For each condition the electrode array was aligned to give maximum delay between the voluntary E M G signals. For each condition the subjects performed brief contractions. When the contraction was stable at 80% M V C 1 sec of E M G was collected by computer and the MFCV was calculated. This was repeated until correct alignment was obtained for the actual condition. A correct alignment was, however, not crucial as we were interested in measuring the changes in MFCV during contraction as an indicator of fatigue. The order of knee angles was randomized. The subjects were continuously encouraged to maintain the torque as high as possible for as long as possible. Endurance was defined as the time when the subject gave up contracting the muscle due to pain or exhaustion. When the subjects gave up the torque was often reduced to 3 0 - 4 0 % of the initial
168
L. A R E N D T - N I E L S E N E T AL.
found at a knee flexion of 5 ° and a contraction of 50% MVC. The minimum M F C V of 3.6 m / s e e was found at a knee flexion of 90 ° and relaxed muscle. The torque corresponding to 50% M V C at 90 ° knee angle was perceived by the subjects to be close to 100% MVC when the knee was close to full extension. The findings demonstrate that the M F C V can change by approximately 2 m / s e e , dependent on the background force and the muscle length. The changes of 2 m / s e e correspond to a 55% change with respect to the M F C V in a relaxed muscle in its rest length (90°). At a flexion of 120 °, the M F C V increased slightly compared to 90 ° (Fig. 1).
6,0 5,5
E
5,0
0 14.
4,5
~
i'°5
50% MVC 25% MVC 0%MVC
,
5
|
i
45
90
120
Experiment 2
Knee angle (degrees)
The time until endurance (subject could not continue) was 52.1 + 8.0 sec and 24.2 + 5.3 sec at 45 and 90 °, respectively. The M V C at 45 ° was 64% of what could be produced at 90 °. In some cases, the M F C V increased slightly during the first 5 - 1 0 sec of the contraction, followed by a decline (Fig. 2). A linear regression fit was estimated throughout the declining phase. The mean decline rate for the M F C V of 0.047 m / s e e / s e e was significantly higher at the knee angle of 90 ° compared to 45 ° (0.024 m / s e e / s e e ) . The mean initial velocity at 45 ° was 1.5 times (5.45 m / s e e ) the velocity at 90 °. The mean decreases in the M F C V during fatigue were 1.14 m / s e e and 1.25 m / s e e for the contractions of 90 and 45 °, respectively. The mean power frequency decreased by 0.35 H z / s e c on average for the contraction at 45 ° and significantly ( P < 0.05) more (0.61 H z / s e c ) for the contraction at 90 ° (Fig. 2). The mean initial MPFs were similar at 45 and 90 ° (95.4 and 91.6 Hz, respectively). Interestingly, on average the mean power frequency reached the same level (around 77 Hz) after the two fatiguing contractions. As for the RMS value, the mean
Fig. 1. The mean (N = 8) M F C V changes as the knee flexion angle is increased from 5 ° (full extension, short muscle length) to 90 ° . The MFCV was determined from the compound muscle potential evoked at 3 different background contraction levels (0, 25 and 50% of MVC determined at a knee flexion of 90 °) where the highest contraction forces cause the highest velocities.
torque. E M G epochs of 250 msec were collected each 5 sec throughout the experiment and the MFCV, M P F and RMS were calculated.
Results
Experiment 1 At all knee angles, the M F C V increased significantly by approximately 1 m / s e e when the background torque was increased from 0 to 50% MVC. The M F C V decreased significantly by approximately 1 m / s e e when the knee angle was changed from full extension (5 °) to a flexion of 90 °. The maximal M F C V of 5.5 m / s e e was
MFCV
~
1oo.M P F
4. FIMS
5'
0
~
4'
2.
80"
< 3
I 0
20
40
Tlme (sec)
60
70
, 0
20
40
Time (sec)
60
0
20
40
60
Tlme (sec)
Fig. 2. A n example from one person showing how the M F C V (left), RMS (middle), and M P F (right) values changed during fatiguing static contractions at 80% MVC. T h e knee joint angles were 45 ° ( • ) and 90 ° ([]). The muscle was stretched most at 90 °.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE
increase rate was significantly higher (0.21 m V / s e c ) for 90 ° compared to 45 ° (0.029 m V / s e c , Fig. 2). The initial RMS values were identical for the two angles which indicates that the RMS value is related to the percentage of MVC and not to the actual muscle force.
Discussion
It is known that many factors (e.g., temperature, tissue filtering, electrode alignment, force, fatigue) affect the MFCV (Arendt-Nielsen and Zwarts 1989). In both our experiments the electrode array was aligned (1) just prior to data collection and (2) under conditions similar to those present when data were collected. Possible artifacts caused by changes in tissue resistivity, and fat thickness between place of alignment and place of recording can therefore be excluded. Errors due to geometrical arrangement of the muscle can, however, not be excluded. The reason why the MFCV-length relationship is non-monotonic might be that the muscle fibres of the vastus lateralis are slightly curved at short muscle lengths. When the muscle length reaches a certain level, the fibres are straightened under the electrode array which virtually results in a slightly increased conduction velocity. This factor cannot be eliminated when surface recordings are made.
MFCV and muscle length The MFCV declined when the muscle was stretched. When a muscle is stretched, the diameter most probably decreases and the conduction velocity along the fibre changes (H~kansson 1957). The core conductor model suggests that the MFCV increases by the square root of the fibre radius. The biochemical mechanisms related to this relation are unknown. The volume of the individual muscle fibre does not change during passive stretch (Elliott et al. 1963, 1967), which implies a reduction in fibre diameter. The total muscle circumference can be regarded as an indirect estimate of the average fibre diameter. Some (St~lberg 1966; Kereshi et al. 1983; Broman et al. 1986) have found a positive correlation between the MFCV and circumference of the limb whereas others have not found it (Nishizono et al. 1979). It has also been suggested that the resistance of the extracellular volume plays a very important role in the conduction velocity (Buchthal et al. 1955; Buchthal and Sten-Knudsen 1959). An increase in length would increase the resistance and reduce the conduction velocity (Buchthal et al. 1955). When the muscle fibre is stretched, the spatial duration of the compound muscle fibre action potential is prolonged (Gydikov and Kosarov 1973). In the vastus muscles, the fibre length changes from 80 to 185 mm when the knee joint is moved from the fully extended
169
to the fully flexed position (Haines 1934). If the conduction velocity at the same time decreases (due to unknown biochemical changes when the muscle diameter is reduced), the temporal duration of the action potential increases. Stretching the muscle causes thinning of the tissue interposed between the skin and the muscle (Gath and St~lberg 1975), which eventually will change the distance-dependent low-pass filtering effect on the potentials. This will obviously affect the mean power frequency (MPF) and cause an increase as the muscle is stretched. This has been shown for the biceps brachii when the muscle is stretched during a movement from a joint angle of 60 ° to 120° (Shankar et al. 1989). Others (Bazzy et al. 1986; Inbar et al. 1987) have shown that a stretched muscle has a lower MPF than a short muscle. If MPF and MFCV correlate as suggested (Arendt-Nielsen et al. 1984) both findings can fit with our results, as the MFCV decreased as the muscle was stretched from 5 to 90 ° knee angle and tended to increase as the muscle was further stretched to 120 °. For the biceps brachii it has been reported that the MPF change versus length is not a monotonic function but has a peak value when the muscle is stretched corresponding to 15 ° above a flexion of 90 ° (Shankar et al. 1989). In the present study, we found that the MFCV reached a minimum for a muscle stretch corresponding to a flexion of 90 °. These non-monotonic relationships might explain the controversies whether the MFCV increases (Wilska and Varjoranta 1940; H~kansson 1957), decreases (Morimoto 1986) or remains constant (Martin 1954; Hodgkin 1954) at increasing muscle length. Another explanation for the nonmonotonic relationship might be provided by the findings of Eloranta (1989). In that study motor unit activity of the human knee extensor muscles was found to increase towards full knee flexion. An increased firing rate would result in increased MFCV (St~lberg 1966; Nishizono et al. 1989). This might also affect the conduction velocity of the electrically evoked potential.
MFCV and contraction force We observed an increase in the MFCV of the compound muscle potential when the background force was increased. As we aligned the electrode to its maximum delay under each condition, it could not be due to changes in the muscle fibre direction as the muscle contracts. Increased contraction force will shorten the muscle by up to 25% (Hoffer et al. 1990), which again probably will increase the MFCV. This contraction-induced shortening may partly explain the increases in the MFCV (Sadoyama et al. 1983; Arendt-Nielsen et al. 1984; Broman et al. 1985; Sadoyama and Masuda 1987) and the MPF (Hagberg and Ericson 1982; Muro et al. 1982; Nagata et al. 1990), measured from the
170
E M G interference pattern, as the contraction level increases. The MFCV increases have previously been related to the recruitment of motor units with higher forces (Brooke and Engel 1969), higher MFCV (Andreassen and Arendt-Nielsen 1987; Nishizono et al. 1990) and higher recruitment thresholds (Gantchev et al. 1991). Another reason why the MFCV is higher at higher contraction forces may be the change in firing rates. After the passage of an action potential, the excitability of the muscle membrane is changed with results in a refractory period. Farmer et al. (1960) found that refractory periods fall into 2 classes: one varying between 2.2 and 3.1 msec and one varying between 3.4 and 4.6 msec. If a new action potential is elicited after 6-10 msec the conduction velocity will be 10-20% faster. St~tlberg (1966) studied extensively the relationship between inter-spike interval and MFCV, which he called the velocity recovery function. At the end of the refractory period (about 2 - 4 msec) he found that the propagation velocity of the second action potential was about 20% lower than the first. At 6-10 msec interval the velocity was unchanged and from 10 to 50 msec the second pulse propagated 12.5-24% faster than the first pulse. He found also that the conduction velocity increased by 0.03-0.6 m / s e c when the stimulation frequency increased from 5 to 20 Hz. This has been confirmed by Nishizono et al. (1989). They measured MFCV from single motor units elicited by microstimulation. An increase in stimulus frequency from 5 to 40 Hz increased the conduction velocity from 3.7 to 4.8 m / s e c . In the study of Bellemare et al. (1983) it was shown that the maximum firing rate was 30 Hz for the biceps brachii muscle, whereas it was only 10 Hz for soleus. If we assume the maximum firing rate of the vastus lateralis is 20 Hz a change in contraction from 0 to 50% MVC would roughly increase firing rate by up to 10 Hz which, according to Nishizono et al. (1989), would result in a 0.47 m / s e c increase in velocity. We found approximately 1 m / s e c increases.
Muscle fatigue and muscle length Long muscles have been reported to be more difficult to fatigue than short muscles (Aljure and Borero 1968; Fitch and McComas 1985; McKenzie and Gandevia 1987). The changes in the MFCV, MPF and RMS during fatiguing contractions under isometric conditions depend on the contraction force (Arendt-Nielsen and Mills 1988; Arendt-Nielsen et al. 1989). It is well documented that the MFCV (for review see ArendtNielsen and Zwarts 1989) and MPF (Petrofsky and Lind 1980; Eberstein and Beattie 1985; Broman et al. 1985) decrease during static muscle fatigue and that the RMS (Sadoyama and Miyano 1981; H~ikkinen and Komi 1983) increases. As the RMS values were equal at 80% MVC obtained at knee extensions of 45 and
L. ARENDT-NIELSEN ET AL.
90 °, as previously shown (Vredenbregt and Rau 1973), we believe that the recruitment status in the contraction was initially identical in the muscle although the absolute forces were different. Normally, the integrated E M G or RMS increases with increasing contraction force (Bigland and Lippold 1954). We therefore assume that both recruitment and motoneurone firing (Gandevia and McKenzie 1988) conditions in the contractions were equal initially. In all cases we found that the E M G parameters changed most rapidly during the short-lasting contraction at a knee flexion of 90 °. The energy required to create and maintain crossbridges might be less for the stretched muscle, which increases the time to endurance. We found that the maximum force, developed at a knee flexion of 45 °, was 64% of that which could be performed at 90 °, due to the reduced twitch forces in short muscles (Gandevia and McKenzie 1988). This is related to another important factor for the development of muscle fatigue; the intramuscular ischaemia which occurs at high contraction forces (Zwarts and Arendt-Nielsen 1988). If an 80% MVC, performed at a knee angle of 90 °, causes intramuscular ischaemia and if an 80% MVC at 45 ° allows perfusion due to the lower force, this might be an important factor for the time until endurance. An interesting finding was that the MPF value reached the same lower limit at the end of the contraction just prior to endurance. Fuglevand et al. (1991) found that the percentages decreased in MPF during 20%, 35% and 65% MVC contractions were similar when the contractions were maintained until endurance. A possible source of artifact might be that the cross-talk from adjacent muscles changes as the knee angle changes. Cross-talk might in such cases act as 'noise' and affect the cross-correlation function, and hence the accuracy of the MFCV determination. Cross-talk among muscles of the leg might contribute by up to 16% of the E M G amplitude (De Luca and Merletti 1988). This is assumed to be of no importance as the cross-correlation functions we measured during the fatiguing contractions in all cases had a well defined peak. We have made some model simulations (unpublished observations) where we have added 'cross-talk noise' to the two E M G signals and found that the cross-correlation function was easily identified when 50% of the E M G amplitude signals was uncorrelated noise. Det Obelske Familiefond, Aalborg Stifts Julelotteri and Assurandcr Societetet are kindly acknowledged for their financial support.
References Aljure, E.F. and Borrero, L.M. The influence of muscle length on the development of fatigue in toad sartorius. J. Physiol. (Lond.), 1968, 199: 241-252.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE
171
Andreassen, S. and Arendt-Nielsen, L. Muscle fibre conduction velocity in motor units of the human anterior tibial muscle: a new size principle parameter. J. Physiol. (Lond.), 1987, 391: 561-571. Arendt-Nielsen, L. and Mills, K.R. The relationship between mean power frequency of the EMG spectrum and muscle fibre conduction velocity. Electroenceph. clin. Neurophysiol., 1985, 60: 130134. Arendt-Nielsen, L. and Mills, K.R. Muscle fibre conduction velocity, mean power frequency, mean EMG voltage and force during submaximal fatiguing contractions of the human quadriceps. Eur. J. Appl. Physiol., 1988, 58: 20-25. Arendt-Nielsen, L. and Zwarts, M. The measurement of muscle fibre conduction velocity in humans - techniques and applications. J. Clin. Neurophysiol., 1989, 6: 173-190. Arendt-Nielsen, L., Forster, A. and Mills, K.R. The relationship between muscle-fiber conduction velocity and force in human vastus lateralis. J. Physiol. (Lond.), 1984, 353: 6P. Arendt-Nielsen, L., Mills, K.R. and Forster, A. Changes in muscle fibre conduction velocity, mean power frequency and mean EMG voltage during prolonged submaximal contractions. Muscle Nerve, 1989, 2: 493-497. Bazzy, A.R., Korten, J.B. and Haddad, G.G. Increase in electromyogram low-frequency power in non-fatigued contracting skeletal muscle. J. Appl. Physiol., 1986, 61: 1012-1017. Bigland, B. and Lippold, O.C.J. The relation between force, velocity and integrated electrical activity in human muscles. J. Physiol. (Lond.), 1954, 123: 214-224. Bellemare, F., Woods, J.J., Johansson, R. and Bigland-Ritchie, B. Motor-unit discharge rates in maximal voluntary contractions of three human muscles. J. Neurophysiol., 1983, 50: 1983. Broman, H., Bilotto, G. and De Luca, C. Myoelectric signal conduction velocity and spectral parameters: influence of force and time. J. Appl. Physiol., 1985, 58: 1428-1437. Brooke, M.H. and Engel, W.K. The histographic analysis of human muscle biopsies with regard to fibre types. Neurology, 1969, 19: 221-233. Buchthal, F. and Sten-Knudsen, O. Impulse propagation in striated muscle fibers and role of the internal current in activation. Ann. NY Acad. Sci., 1959, 81: 422-445. Buchthal, F., Guld, C. and Rosenfalck, A. Innervation zone and propagation velocity in human muscle. Acta Physiol. Scand., 1955, 35: 174-190. De Luca, C.J. and Merletti, R. Surface myoelectric signal cross-talk among muscles of the leg. Electroenceph. clin. Neurophysiol., 1988, 69: 568-575. Eberstein, A. and Beattie, B. Simultaneous measurement of muscle conduction velocity and EMG power spectrum changes during fatigue. Muscle Nerve, 1985, 8: 768-773. Elliott, G.F., Lowy, J. and Worthington, C.R. An X-ray and light diffraction study of the filament lattice of striated muscle in the living state and rigor. J. Mol. Biol., 1963, 6: 295-305. Elliott, G.F., Lowy, J. and Milman, B.M. Low angle X-ray diffraction studies of living striated muscle during contraction, J. Mol. Biol., 1967, 25: 31-45. Eloranta, V. Coordination of the thigh muscles in static leg extension. Electromyogr. Clin. Neurophysiol., 1989, 29: 227-233. Farmer, T.W., Buchthal, F. and Rosenfalck, P. Refractory period of human muscle after passage of a propagated action potential. Electroenceph. clin. Neurophysiol., 1960, 12: 455-466. Fitch, S. and McComas, A.J. Influence of human muscle length on fatigue. J. Physiol. (Lond.), 1985, 362: 205-213. Fuglevand, A.J., Zackowski, K.M., Huey, K.A. and Enoka, R.M. Inability of humans to fully activate muscle during sustained contractions at different force levels. Neurosci. Soe. Proc. (New Orleans), 1991: abstract 555.1. Gandevia, S.C. and McKenzie, D.K. Activation of human muscles at
short muscle lengths during maximal static efforts. J. Physiol. (Lond.), 1988, 407: 599-613. Gantchev, N., Kossev, A., Gydikov, A. and Gerasimenko, Y. Relation between the motor units recruitment threshold and their potentials propagation velocity at isometric activity. Electromyogr. Clin. Neurophysiol., 1991, 31: in press. Gath, I. and St~lberg, E. Frequency and time domain characteristics of single muscle fiber potentials. Electromyogr. Clin. Neurophysiol., 1975, 39: 371-376. Gydikov, A. and Kosarov, D. The influence of various factors on the shape of the myopotentials in using monopolar electrodes. Electromyogr. Clin. Neurophysiol., 1973, 13: 319-343. Hagberg, M. and Ericson, B.-E. Myoelectric power spectrum dependence on muscular contraction level of elbow flexors. Eur. J. Appl. Physiol., 1982, 48: 147-156. Haines, R.W. The law of muscle and tendon growth. J. Anat., 1932, 66: 578-585. Haines, R.W. On muscle of full and of short action, J. Anat., 1934, 69: 20-24. Hfikansson, C.H. Action potential and mechanical response of isolated cross striated frog muscle fibers at different degrees of stretch. Acta Physiol. Scand., 1957, 39: 199-216. H~ikkinen, K. and Komi, P.V. Electromyographic and mechanical characteristics of human skeletal muscle during fatigue under voluntary and reflex conditions. Electroenceph. clin. Neurophysiol., 1983, 55: 436-444. Hodgkin, A.L. A note on conduction velocity, J. Physiol. (Lond.), 1954, 125: 221-224. Hoffer, J.A., Caputi, A.A., Pose, I.E. and Griffiths, R.I. Roles of muscle activity and load on the relationship between muscle spindle length and whole muscle length in the freely walking cat. In: J.H.J. Allum and M. Hulliger (Eds.), Progress in Brain Research, Vol. 80. Elsevier Science Publishers, Amsterdam, 1990: 75-85. Inbar, G.F., Allin, J. and Kranz, H. Surface EMG spectral changes with muscle length. Med. Biol. Eng. Comput., 1987, 25: 683-689. Katz, B. The effect of electrolyte deficiency on the rate of conduction in single nerve fibre. J. Physiol. (Lond.), 1947, 106: 411-417. Kereshi, S., Manzano, G. and McComas, A.J. Impulse conduction velocities in human biceps brachii muscle. Exp. Neurol., 1983, 80: 652-662. Kossev, A., Gantchev, N., Gydikov, A., Gerasimenko, Y. and Christova, P. The effect of muscle fiber length change on motor units potentials propagation velocity. Electromyogr. Clin. Neurophysiol., 1991, 31: in press. Martin, A.R. The effect of change in length on conduction velocity in muscle. J. Physiol. (Lond.), 1954, 125: 215-220. McKenzie, D.K. and Gandevia, S.C. Influence of muscle length on human inspiratory and limb muscle endurance. Resp. Physiol., 1987, 67: 171-182. Morimoto, S. Effect of length change in muscle fibers on conduction velocity in human motor units. Jpn. J. Physiol., 1986, 36: 773-782. Muro, M., Nagata, A., Murakami, M.D. and Moritani, T. Surface EMG power spectral analysis of neuromuscular disorders during isometric and isotonic contractions. Am. J. Phys. Med., 1982, 61: 244-254. Nagata, S., Arsenault, A.B., Gagnon, D., Smyth, G. and Mathieu, P.-A. EMG power spectrum as a measure of muscular fatigue at different levels of contraction. Med. Biol. Eng. Comput., 1990, 28: 374-378. Nishizono, H., Saito, Y. and Miyashita, M. The estimation of conduction velocity in human skeletal muscles in situ with surface electrodes. Electroenceph. clin. Neurophysiol., 1979, 46: 659-664. Nishizono, H., Kurata, H. and Miyashita, M. Muscle fiber conduction velocity related to stimulation rate. Electroenceph. clin. Neurophysiol., 1989, 72: 529-534.
172 Nishizono, H., Fujimoto, T., Ohtake, H. and Miyashita, M. Muscle fibre conduction velocity and contractile properties estimated from surface electrode arrays. Electroenceph. clin. Neurophysiol., 1990, 75: 75-81. Petrofsky, J.S. and Lind, A.R. Frequency analysis of the surface electromyogram during sustained isometric contractions. Eur. J. Appl. Physiol., 1980, 43: 173-182. Sadoyama, T. and Masuda, T. Changes of the average muscle fiber conduction velocity during a varying force contraction. Electroenceph. clin. Neurophysiol., 1987, 67: 495-497. Sadoyama, T. and Miyano, H. Frequency analysis of surface EMG to evaluation of muscle fatigue. Eur. J. Appl. Physiol., 1981, 47: 239-246. Sadoyama, T., Masuda, T. and Miyano, H. Relationships between muscle fiber conduction velocity and frequency parameters of surface EMG during sustained contraction. Eur. J. Appl. Physiol., 1983, 51: 247-256.
L. ARENDT-NIELSEN ET AL. Shankar, S., Gander, R.E. and Brandell, B.R. Changes in the myoelectrical signal (MES) power spectra during dynamic contractions. Electroenceph. clin. Neurophysiol., 1989, 73: 142-150. St,~lberg, E. Propagation velocity in human muscle fibers in situ. Acta Physiol. Scand., 1966, 287: 3-112. Vredenbregt, J. and Rau, G. Surface electromyography in relation to force muscle length and endurance. In: J.E. Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology, Vol. 1. Karger, Basel, 1973: 607-622. Wilska, A. and Varjoranta, K. Untersuchungen iiber die Geschwindigkeit des geleiteten Aktionspotentials einer Muskelfaser bei verschiedenen Dehnungszustanden des Muskels. Skand. Arch. Physiol., 1940, 83: 82-87. Zwarts, M.J. and Arendt-Nielsen, L. The influence of force and circulation on average muscle fibre conduction velocity during local muscle fatigue. Eur. J. Appl. Physiol., 1988, 58: 278-283.
166
E L M O C O 91661
The influence of muscle length on muscle fibre conduction velocity and development of muscle fatigue Lars A r e n d t - N i e l s e n
a
Nikolai G a n t c h e v b and T h o m a s Sinkja~r a
Laboratory for Motor Control, Department of Medical Informatics, Aalborg University, DK-9220 Aalborg E (Denmark), and b Central Laboratory of Biophysics, Bulgarian Academy of Sciences, Sofia 1113 (Bulgaria) (Accepted for publication: 9 D e c e m b e r 1991)
Summary The influence of muscle (vastus lateralis) length on the muscle fibre conduction velocity (MFCV) and on muscle fatigue was studied in 8 healthy volunteers. In experiment 1, the electromyographic (EMG) responses were evoked by electrical stimulation of the motor point and recorded by a surface electrode array aligned along the muscle fibre direction. The M F C V (determined by cross-correlation) was m e a s u r e d at knee flexions of 5 ° (full extension), 45 °, 90 ° and 120° with 3 different extension torques. The M F C V declined with increasing muscle length and increased with increasing background torque at knee flexions from 5 ° to 90 °. From 90 ° to 120 ° knee flexion of M F C V tended to increase. In experiment 2, the E M G activity at a static fatiguing contraction (80% MVC) was measured at 45 ° and 90 ° knee flexion. The E M G was measured until the subject gave up contracting the muscle (endurance). The largest increase in the RMS amplitude and the fastest decreases in the mean power frequency (MPF) and M F C V were found at 90 ° flexion. The M V C at 45 ° knee flexion was 35% lower than at 90 ° and the time until endurance was approximately twice as long for the 45 ° contraction. The results indicate that muscle length is an important parameter for the propagation velocity of action potentials and for the development of static muscle fatigue.
Key words: Muscle; Length; Fatigue; Conduction velocity; EMG; Propagation
Muscle fibre length in situ can change by more than 100% when a joint is rotated from the fully extended to the fully flexed position (Haines 1932, 1934). Theoretically and experimentally it has been found that the muscle fibre conduction velocity (MFCV) increases when the diameter increases (Katz 1947; Hfikansson 1957; Kossev et al. 1991). Studies in which the MFCV is used clinically have begun to appear (for review see Arendt-Nielsen and Zwarts 1989) and it is therefore important to reach conclusions on the various factors affecting the MFCVo It has been suggested that the MFCV either increases (Morimoto 1986), decreases (Wilska and Varjoranta 1940; H,~kansson 1957) or remains unchanged (Martin 1954; Hodgkin 1954) when the muscle is shortened. The first aim of the present study was to measure the muscle fibre conduction velocity (MFCV) from the human vastus lateralis muscle at different muscle
Correspondence to: Dr. Lars Arendt-Nielsen, Laboratory for Motor Control, D e p a r t m e n t of Medical Informatics, Aalborg University, Fredrik Bajersvej 7D, DK-9220 Aalborg E (Denmark). Tel.: + 4 5 98158522, ext. 4954; Fax: +45 98154008.
lengths to clarify whether the MFCV increases, decreases or remains unchanged when the muscle is shortened. The muscle length also seems to be of importance to the way in which the muscle is fatigued as the endurance is enhanced at short muscle lengths (Aljure and Borrero 1968; Fitch and McComas 1985; McKenzie and Gandevia 1987). The twitch contraction and relaxation times are decreased in short muscles (Gandevia and McKenzie 1988) but it has been shown (Gandevia and McKenzie 1988) that the motoneurones increase their firing rates until fusion is achieved in short muscles. Even though fusion can be obtained, the maximum voluntary force declined by 21-49% at short muscle lengths due to reduced twitch forces (Gandevia and McKenzie 1988). During dynamic muscle load, the muscle length changes continuously, but the importance of length changes as the development of fatigue is unknown. The electromyogram (EMG) can be used to clarify some of these aspects. The second aim of the present study was to monitor EMG parameters (MFCV, mean power frequency, RMS amplitude) during fatiguing voluntary contractions maintained at different muscle lengths.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE Methods and materials
Volunteers Eight healthy male volunteers participated (mean age 23 years, range 20-30 years). Informed consent was obtained, and the Helsinki Declaration was respected.
Experimental set-up During the recording session, the subject was seated in a test chair. The dorsal part of the ankle was firmly fastened against a rubber block with straps around the ankle. The block was mounted on a shaft which again was attached to a strain gauge. The torque generated by the knee extensor muscles was displayed on an oscilloscope and collected on computer. The axis of rotation of the shaft was aligned with the axis of rotation of the knee joint. A safety belt was strapped around the belly and the back of the test chair. The hip joint was flexed 90 ° . All experiments were performed under isometric conditions. The maximum voluntary isometric knee extension torque was measured at 90 ° knee flexion as the largest of 3 brief maximal extensions. The skin t e m p e r a t u r e over the vastus lateralis was maintained at 34.5°C by an infrared heater. It is important to maintain the t e m p e r a t u r e constant as the M F C V is highly affected by changes in temperature. The t e m p e r a t u r e of 34.5°C was normally 2-3°C above the normal skin temperature.
Data collection The E M G electrodes were placed in a bipolar array (centre electrode common), 15 m m apart, and placed distal to the motor point and parallel to the fibre axis of the vastus lateralis muscle. Electrical surface stimulation was used to determine the position of its motor point. The stimulus current was increased until maximum E M G amplitude was reached. The correct electrode alignment was obtained when the maximal delay between the two E M G s evoked by electrical stimulation over the motor point had been reached. Effective stabilization of the electrode was achieved by adhesive tape and external compression. The E M G signals were filtered (20-500 Hz), amplified, sampled (2000 Hz) and stored on disc. Before the experiment, the skin was lightly abraded and cleaned with isopropyl alcohol. Electrode gel was applied sparingly to the 3 Ag-AgC1 surface electrodes. During the experiments, the E M G signal and torque were stored in a computer for later analysis.
Data analysis The E M G signals were divided into 250 msec time intervals. Cross-correlation between the 2 E M G signals was performed (voluntary activity or electrically evoked
167 compound action potentials) and provided the electrodes were positioned correctly, a high (R = 0.8-0.99) correlation between the two signals was obtained. The peak in the correlogram was displaced from time zero by a time lag reflecting the conduction time between the two pairs of electrodes. The mean power frequency (MFP) was calculated as the ratio of the first order and zero order moment of the E M G power spectrum. The RMS amplitude was calculated as the square root of the squared mean amplitude of the E M G epochs. As a function of time, the changes in the parameters were estimated by the slopes of the linear regression lines. Wilcoxon's test was used for statistical analysis and P < 0.05% was regarded as statistically significant.
Protocol Experiment 1.
The M F C V was determined from the compound potential evoked by electrical motor point stimulation at 4 different knee angles (5, 45, 90 and 120 °) and 3 different contraction levels (0, 25, 50% maximum voluntary contraction (MVC) was determined for 90 ° knee flexion). The subject was asked to hit a target line on the oscilloscope and maintain the torque at the given percentage of MVC. Constant extension torques were used at the different knee flexion angles. A knee angle of 5 ° corresponded to full extension. The angles and torques were presented in random order. For each condition, the maximum delay between the electrically evoked compound muscle action potential was determined. When maximum delay was obtained a number (5-10) of compound potentials were immediately collected for cross-correlation analysis. The mean M F C V for the 5 - 1 0 potentials was calculated. Experiment 2. On two occasions, separated by at least 48 h, the E M G and isometric knee extension torque were measured at knee extensions of 90 and 45 °. The initial torque was 80% MVC. Maximum voluntary contractions were determined individually for 45 and 90 ° knee angle. For each condition the electrode array was aligned to give maximum delay between the voluntary E M G signals. For each condition the subjects performed brief contractions. When the contraction was stable at 80% M V C 1 sec of E M G was collected by computer and the MFCV was calculated. This was repeated until correct alignment was obtained for the actual condition. A correct alignment was, however, not crucial as we were interested in measuring the changes in MFCV during contraction as an indicator of fatigue. The order of knee angles was randomized. The subjects were continuously encouraged to maintain the torque as high as possible for as long as possible. Endurance was defined as the time when the subject gave up contracting the muscle due to pain or exhaustion. When the subjects gave up the torque was often reduced to 3 0 - 4 0 % of the initial
168
L. A R E N D T - N I E L S E N E T AL.
found at a knee flexion of 5 ° and a contraction of 50% MVC. The minimum M F C V of 3.6 m / s e e was found at a knee flexion of 90 ° and relaxed muscle. The torque corresponding to 50% M V C at 90 ° knee angle was perceived by the subjects to be close to 100% MVC when the knee was close to full extension. The findings demonstrate that the M F C V can change by approximately 2 m / s e e , dependent on the background force and the muscle length. The changes of 2 m / s e e correspond to a 55% change with respect to the M F C V in a relaxed muscle in its rest length (90°). At a flexion of 120 °, the M F C V increased slightly compared to 90 ° (Fig. 1).
6,0 5,5
E
5,0
0 14.
4,5
~
i'°5
50% MVC 25% MVC 0%MVC
,
5
|
i
45
90
120
Experiment 2
Knee angle (degrees)
The time until endurance (subject could not continue) was 52.1 + 8.0 sec and 24.2 + 5.3 sec at 45 and 90 °, respectively. The M V C at 45 ° was 64% of what could be produced at 90 °. In some cases, the M F C V increased slightly during the first 5 - 1 0 sec of the contraction, followed by a decline (Fig. 2). A linear regression fit was estimated throughout the declining phase. The mean decline rate for the M F C V of 0.047 m / s e e / s e e was significantly higher at the knee angle of 90 ° compared to 45 ° (0.024 m / s e e / s e e ) . The mean initial velocity at 45 ° was 1.5 times (5.45 m / s e e ) the velocity at 90 °. The mean decreases in the M F C V during fatigue were 1.14 m / s e e and 1.25 m / s e e for the contractions of 90 and 45 °, respectively. The mean power frequency decreased by 0.35 H z / s e c on average for the contraction at 45 ° and significantly ( P < 0.05) more (0.61 H z / s e c ) for the contraction at 90 ° (Fig. 2). The mean initial MPFs were similar at 45 and 90 ° (95.4 and 91.6 Hz, respectively). Interestingly, on average the mean power frequency reached the same level (around 77 Hz) after the two fatiguing contractions. As for the RMS value, the mean
Fig. 1. The mean (N = 8) M F C V changes as the knee flexion angle is increased from 5 ° (full extension, short muscle length) to 90 ° . The MFCV was determined from the compound muscle potential evoked at 3 different background contraction levels (0, 25 and 50% of MVC determined at a knee flexion of 90 °) where the highest contraction forces cause the highest velocities.
torque. E M G epochs of 250 msec were collected each 5 sec throughout the experiment and the MFCV, M P F and RMS were calculated.
Results
Experiment 1 At all knee angles, the M F C V increased significantly by approximately 1 m / s e e when the background torque was increased from 0 to 50% MVC. The M F C V decreased significantly by approximately 1 m / s e e when the knee angle was changed from full extension (5 °) to a flexion of 90 °. The maximal M F C V of 5.5 m / s e e was
MFCV
~
1oo.M P F
4. FIMS
5'
0
~
4'
2.
80"
< 3
I 0
20
40
Tlme (sec)
60
70
, 0
20
40
Time (sec)
60
0
20
40
60
Tlme (sec)
Fig. 2. A n example from one person showing how the M F C V (left), RMS (middle), and M P F (right) values changed during fatiguing static contractions at 80% MVC. T h e knee joint angles were 45 ° ( • ) and 90 ° ([]). The muscle was stretched most at 90 °.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE
increase rate was significantly higher (0.21 m V / s e c ) for 90 ° compared to 45 ° (0.029 m V / s e c , Fig. 2). The initial RMS values were identical for the two angles which indicates that the RMS value is related to the percentage of MVC and not to the actual muscle force.
Discussion
It is known that many factors (e.g., temperature, tissue filtering, electrode alignment, force, fatigue) affect the MFCV (Arendt-Nielsen and Zwarts 1989). In both our experiments the electrode array was aligned (1) just prior to data collection and (2) under conditions similar to those present when data were collected. Possible artifacts caused by changes in tissue resistivity, and fat thickness between place of alignment and place of recording can therefore be excluded. Errors due to geometrical arrangement of the muscle can, however, not be excluded. The reason why the MFCV-length relationship is non-monotonic might be that the muscle fibres of the vastus lateralis are slightly curved at short muscle lengths. When the muscle length reaches a certain level, the fibres are straightened under the electrode array which virtually results in a slightly increased conduction velocity. This factor cannot be eliminated when surface recordings are made.
MFCV and muscle length The MFCV declined when the muscle was stretched. When a muscle is stretched, the diameter most probably decreases and the conduction velocity along the fibre changes (H~kansson 1957). The core conductor model suggests that the MFCV increases by the square root of the fibre radius. The biochemical mechanisms related to this relation are unknown. The volume of the individual muscle fibre does not change during passive stretch (Elliott et al. 1963, 1967), which implies a reduction in fibre diameter. The total muscle circumference can be regarded as an indirect estimate of the average fibre diameter. Some (St~lberg 1966; Kereshi et al. 1983; Broman et al. 1986) have found a positive correlation between the MFCV and circumference of the limb whereas others have not found it (Nishizono et al. 1979). It has also been suggested that the resistance of the extracellular volume plays a very important role in the conduction velocity (Buchthal et al. 1955; Buchthal and Sten-Knudsen 1959). An increase in length would increase the resistance and reduce the conduction velocity (Buchthal et al. 1955). When the muscle fibre is stretched, the spatial duration of the compound muscle fibre action potential is prolonged (Gydikov and Kosarov 1973). In the vastus muscles, the fibre length changes from 80 to 185 mm when the knee joint is moved from the fully extended
169
to the fully flexed position (Haines 1934). If the conduction velocity at the same time decreases (due to unknown biochemical changes when the muscle diameter is reduced), the temporal duration of the action potential increases. Stretching the muscle causes thinning of the tissue interposed between the skin and the muscle (Gath and St~lberg 1975), which eventually will change the distance-dependent low-pass filtering effect on the potentials. This will obviously affect the mean power frequency (MPF) and cause an increase as the muscle is stretched. This has been shown for the biceps brachii when the muscle is stretched during a movement from a joint angle of 60 ° to 120° (Shankar et al. 1989). Others (Bazzy et al. 1986; Inbar et al. 1987) have shown that a stretched muscle has a lower MPF than a short muscle. If MPF and MFCV correlate as suggested (Arendt-Nielsen et al. 1984) both findings can fit with our results, as the MFCV decreased as the muscle was stretched from 5 to 90 ° knee angle and tended to increase as the muscle was further stretched to 120 °. For the biceps brachii it has been reported that the MPF change versus length is not a monotonic function but has a peak value when the muscle is stretched corresponding to 15 ° above a flexion of 90 ° (Shankar et al. 1989). In the present study, we found that the MFCV reached a minimum for a muscle stretch corresponding to a flexion of 90 °. These non-monotonic relationships might explain the controversies whether the MFCV increases (Wilska and Varjoranta 1940; H~kansson 1957), decreases (Morimoto 1986) or remains constant (Martin 1954; Hodgkin 1954) at increasing muscle length. Another explanation for the nonmonotonic relationship might be provided by the findings of Eloranta (1989). In that study motor unit activity of the human knee extensor muscles was found to increase towards full knee flexion. An increased firing rate would result in increased MFCV (St~lberg 1966; Nishizono et al. 1989). This might also affect the conduction velocity of the electrically evoked potential.
MFCV and contraction force We observed an increase in the MFCV of the compound muscle potential when the background force was increased. As we aligned the electrode to its maximum delay under each condition, it could not be due to changes in the muscle fibre direction as the muscle contracts. Increased contraction force will shorten the muscle by up to 25% (Hoffer et al. 1990), which again probably will increase the MFCV. This contraction-induced shortening may partly explain the increases in the MFCV (Sadoyama et al. 1983; Arendt-Nielsen et al. 1984; Broman et al. 1985; Sadoyama and Masuda 1987) and the MPF (Hagberg and Ericson 1982; Muro et al. 1982; Nagata et al. 1990), measured from the
170
E M G interference pattern, as the contraction level increases. The MFCV increases have previously been related to the recruitment of motor units with higher forces (Brooke and Engel 1969), higher MFCV (Andreassen and Arendt-Nielsen 1987; Nishizono et al. 1990) and higher recruitment thresholds (Gantchev et al. 1991). Another reason why the MFCV is higher at higher contraction forces may be the change in firing rates. After the passage of an action potential, the excitability of the muscle membrane is changed with results in a refractory period. Farmer et al. (1960) found that refractory periods fall into 2 classes: one varying between 2.2 and 3.1 msec and one varying between 3.4 and 4.6 msec. If a new action potential is elicited after 6-10 msec the conduction velocity will be 10-20% faster. St~tlberg (1966) studied extensively the relationship between inter-spike interval and MFCV, which he called the velocity recovery function. At the end of the refractory period (about 2 - 4 msec) he found that the propagation velocity of the second action potential was about 20% lower than the first. At 6-10 msec interval the velocity was unchanged and from 10 to 50 msec the second pulse propagated 12.5-24% faster than the first pulse. He found also that the conduction velocity increased by 0.03-0.6 m / s e c when the stimulation frequency increased from 5 to 20 Hz. This has been confirmed by Nishizono et al. (1989). They measured MFCV from single motor units elicited by microstimulation. An increase in stimulus frequency from 5 to 40 Hz increased the conduction velocity from 3.7 to 4.8 m / s e c . In the study of Bellemare et al. (1983) it was shown that the maximum firing rate was 30 Hz for the biceps brachii muscle, whereas it was only 10 Hz for soleus. If we assume the maximum firing rate of the vastus lateralis is 20 Hz a change in contraction from 0 to 50% MVC would roughly increase firing rate by up to 10 Hz which, according to Nishizono et al. (1989), would result in a 0.47 m / s e c increase in velocity. We found approximately 1 m / s e c increases.
Muscle fatigue and muscle length Long muscles have been reported to be more difficult to fatigue than short muscles (Aljure and Borero 1968; Fitch and McComas 1985; McKenzie and Gandevia 1987). The changes in the MFCV, MPF and RMS during fatiguing contractions under isometric conditions depend on the contraction force (Arendt-Nielsen and Mills 1988; Arendt-Nielsen et al. 1989). It is well documented that the MFCV (for review see ArendtNielsen and Zwarts 1989) and MPF (Petrofsky and Lind 1980; Eberstein and Beattie 1985; Broman et al. 1985) decrease during static muscle fatigue and that the RMS (Sadoyama and Miyano 1981; H~ikkinen and Komi 1983) increases. As the RMS values were equal at 80% MVC obtained at knee extensions of 45 and
L. ARENDT-NIELSEN ET AL.
90 °, as previously shown (Vredenbregt and Rau 1973), we believe that the recruitment status in the contraction was initially identical in the muscle although the absolute forces were different. Normally, the integrated E M G or RMS increases with increasing contraction force (Bigland and Lippold 1954). We therefore assume that both recruitment and motoneurone firing (Gandevia and McKenzie 1988) conditions in the contractions were equal initially. In all cases we found that the E M G parameters changed most rapidly during the short-lasting contraction at a knee flexion of 90 °. The energy required to create and maintain crossbridges might be less for the stretched muscle, which increases the time to endurance. We found that the maximum force, developed at a knee flexion of 45 °, was 64% of that which could be performed at 90 °, due to the reduced twitch forces in short muscles (Gandevia and McKenzie 1988). This is related to another important factor for the development of muscle fatigue; the intramuscular ischaemia which occurs at high contraction forces (Zwarts and Arendt-Nielsen 1988). If an 80% MVC, performed at a knee angle of 90 °, causes intramuscular ischaemia and if an 80% MVC at 45 ° allows perfusion due to the lower force, this might be an important factor for the time until endurance. An interesting finding was that the MPF value reached the same lower limit at the end of the contraction just prior to endurance. Fuglevand et al. (1991) found that the percentages decreased in MPF during 20%, 35% and 65% MVC contractions were similar when the contractions were maintained until endurance. A possible source of artifact might be that the cross-talk from adjacent muscles changes as the knee angle changes. Cross-talk might in such cases act as 'noise' and affect the cross-correlation function, and hence the accuracy of the MFCV determination. Cross-talk among muscles of the leg might contribute by up to 16% of the E M G amplitude (De Luca and Merletti 1988). This is assumed to be of no importance as the cross-correlation functions we measured during the fatiguing contractions in all cases had a well defined peak. We have made some model simulations (unpublished observations) where we have added 'cross-talk noise' to the two E M G signals and found that the cross-correlation function was easily identified when 50% of the E M G amplitude signals was uncorrelated noise. Det Obelske Familiefond, Aalborg Stifts Julelotteri and Assurandcr Societetet are kindly acknowledged for their financial support.
References Aljure, E.F. and Borrero, L.M. The influence of muscle length on the development of fatigue in toad sartorius. J. Physiol. (Lond.), 1968, 199: 241-252.
MUSCLE LENGTH, CONDUCTION VELOCITY AND FATIGUE
171
Andreassen, S. and Arendt-Nielsen, L. Muscle fibre conduction velocity in motor units of the human anterior tibial muscle: a new size principle parameter. J. Physiol. (Lond.), 1987, 391: 561-571. Arendt-Nielsen, L. and Mills, K.R. The relationship between mean power frequency of the EMG spectrum and muscle fibre conduction velocity. Electroenceph. clin. Neurophysiol., 1985, 60: 130134. Arendt-Nielsen, L. and Mills, K.R. Muscle fibre conduction velocity, mean power frequency, mean EMG voltage and force during submaximal fatiguing contractions of the human quadriceps. Eur. J. Appl. Physiol., 1988, 58: 20-25. Arendt-Nielsen, L. and Zwarts, M. The measurement of muscle fibre conduction velocity in humans - techniques and applications. J. Clin. Neurophysiol., 1989, 6: 173-190. Arendt-Nielsen, L., Forster, A. and Mills, K.R. The relationship between muscle-fiber conduction velocity and force in human vastus lateralis. J. Physiol. (Lond.), 1984, 353: 6P. Arendt-Nielsen, L., Mills, K.R. and Forster, A. Changes in muscle fibre conduction velocity, mean power frequency and mean EMG voltage during prolonged submaximal contractions. Muscle Nerve, 1989, 2: 493-497. Bazzy, A.R., Korten, J.B. and Haddad, G.G. Increase in electromyogram low-frequency power in non-fatigued contracting skeletal muscle. J. Appl. Physiol., 1986, 61: 1012-1017. Bigland, B. and Lippold, O.C.J. The relation between force, velocity and integrated electrical activity in human muscles. J. Physiol. (Lond.), 1954, 123: 214-224. Bellemare, F., Woods, J.J., Johansson, R. and Bigland-Ritchie, B. Motor-unit discharge rates in maximal voluntary contractions of three human muscles. J. Neurophysiol., 1983, 50: 1983. Broman, H., Bilotto, G. and De Luca, C. Myoelectric signal conduction velocity and spectral parameters: influence of force and time. J. Appl. Physiol., 1985, 58: 1428-1437. Brooke, M.H. and Engel, W.K. The histographic analysis of human muscle biopsies with regard to fibre types. Neurology, 1969, 19: 221-233. Buchthal, F. and Sten-Knudsen, O. Impulse propagation in striated muscle fibers and role of the internal current in activation. Ann. NY Acad. Sci., 1959, 81: 422-445. Buchthal, F., Guld, C. and Rosenfalck, A. Innervation zone and propagation velocity in human muscle. Acta Physiol. Scand., 1955, 35: 174-190. De Luca, C.J. and Merletti, R. Surface myoelectric signal cross-talk among muscles of the leg. Electroenceph. clin. Neurophysiol., 1988, 69: 568-575. Eberstein, A. and Beattie, B. Simultaneous measurement of muscle conduction velocity and EMG power spectrum changes during fatigue. Muscle Nerve, 1985, 8: 768-773. Elliott, G.F., Lowy, J. and Worthington, C.R. An X-ray and light diffraction study of the filament lattice of striated muscle in the living state and rigor. J. Mol. Biol., 1963, 6: 295-305. Elliott, G.F., Lowy, J. and Milman, B.M. Low angle X-ray diffraction studies of living striated muscle during contraction, J. Mol. Biol., 1967, 25: 31-45. Eloranta, V. Coordination of the thigh muscles in static leg extension. Electromyogr. Clin. Neurophysiol., 1989, 29: 227-233. Farmer, T.W., Buchthal, F. and Rosenfalck, P. Refractory period of human muscle after passage of a propagated action potential. Electroenceph. clin. Neurophysiol., 1960, 12: 455-466. Fitch, S. and McComas, A.J. Influence of human muscle length on fatigue. J. Physiol. (Lond.), 1985, 362: 205-213. Fuglevand, A.J., Zackowski, K.M., Huey, K.A. and Enoka, R.M. Inability of humans to fully activate muscle during sustained contractions at different force levels. Neurosci. Soe. Proc. (New Orleans), 1991: abstract 555.1. Gandevia, S.C. and McKenzie, D.K. Activation of human muscles at
short muscle lengths during maximal static efforts. J. Physiol. (Lond.), 1988, 407: 599-613. Gantchev, N., Kossev, A., Gydikov, A. and Gerasimenko, Y. Relation between the motor units recruitment threshold and their potentials propagation velocity at isometric activity. Electromyogr. Clin. Neurophysiol., 1991, 31: in press. Gath, I. and St~lberg, E. Frequency and time domain characteristics of single muscle fiber potentials. Electromyogr. Clin. Neurophysiol., 1975, 39: 371-376. Gydikov, A. and Kosarov, D. The influence of various factors on the shape of the myopotentials in using monopolar electrodes. Electromyogr. Clin. Neurophysiol., 1973, 13: 319-343. Hagberg, M. and Ericson, B.-E. Myoelectric power spectrum dependence on muscular contraction level of elbow flexors. Eur. J. Appl. Physiol., 1982, 48: 147-156. Haines, R.W. The law of muscle and tendon growth. J. Anat., 1932, 66: 578-585. Haines, R.W. On muscle of full and of short action, J. Anat., 1934, 69: 20-24. Hfikansson, C.H. Action potential and mechanical response of isolated cross striated frog muscle fibers at different degrees of stretch. Acta Physiol. Scand., 1957, 39: 199-216. H~ikkinen, K. and Komi, P.V. Electromyographic and mechanical characteristics of human skeletal muscle during fatigue under voluntary and reflex conditions. Electroenceph. clin. Neurophysiol., 1983, 55: 436-444. Hodgkin, A.L. A note on conduction velocity, J. Physiol. (Lond.), 1954, 125: 221-224. Hoffer, J.A., Caputi, A.A., Pose, I.E. and Griffiths, R.I. Roles of muscle activity and load on the relationship between muscle spindle length and whole muscle length in the freely walking cat. In: J.H.J. Allum and M. Hulliger (Eds.), Progress in Brain Research, Vol. 80. Elsevier Science Publishers, Amsterdam, 1990: 75-85. Inbar, G.F., Allin, J. and Kranz, H. Surface EMG spectral changes with muscle length. Med. Biol. Eng. Comput., 1987, 25: 683-689. Katz, B. The effect of electrolyte deficiency on the rate of conduction in single nerve fibre. J. Physiol. (Lond.), 1947, 106: 411-417. Kereshi, S., Manzano, G. and McComas, A.J. Impulse conduction velocities in human biceps brachii muscle. Exp. Neurol., 1983, 80: 652-662. Kossev, A., Gantchev, N., Gydikov, A., Gerasimenko, Y. and Christova, P. The effect of muscle fiber length change on motor units potentials propagation velocity. Electromyogr. Clin. Neurophysiol., 1991, 31: in press. Martin, A.R. The effect of change in length on conduction velocity in muscle. J. Physiol. (Lond.), 1954, 125: 215-220. McKenzie, D.K. and Gandevia, S.C. Influence of muscle length on human inspiratory and limb muscle endurance. Resp. Physiol., 1987, 67: 171-182. Morimoto, S. Effect of length change in muscle fibers on conduction velocity in human motor units. Jpn. J. Physiol., 1986, 36: 773-782. Muro, M., Nagata, A., Murakami, M.D. and Moritani, T. Surface EMG power spectral analysis of neuromuscular disorders during isometric and isotonic contractions. Am. J. Phys. Med., 1982, 61: 244-254. Nagata, S., Arsenault, A.B., Gagnon, D., Smyth, G. and Mathieu, P.-A. EMG power spectrum as a measure of muscular fatigue at different levels of contraction. Med. Biol. Eng. Comput., 1990, 28: 374-378. Nishizono, H., Saito, Y. and Miyashita, M. The estimation of conduction velocity in human skeletal muscles in situ with surface electrodes. Electroenceph. clin. Neurophysiol., 1979, 46: 659-664. Nishizono, H., Kurata, H. and Miyashita, M. Muscle fiber conduction velocity related to stimulation rate. Electroenceph. clin. Neurophysiol., 1989, 72: 529-534.
172 Nishizono, H., Fujimoto, T., Ohtake, H. and Miyashita, M. Muscle fibre conduction velocity and contractile properties estimated from surface electrode arrays. Electroenceph. clin. Neurophysiol., 1990, 75: 75-81. Petrofsky, J.S. and Lind, A.R. Frequency analysis of the surface electromyogram during sustained isometric contractions. Eur. J. Appl. Physiol., 1980, 43: 173-182. Sadoyama, T. and Masuda, T. Changes of the average muscle fiber conduction velocity during a varying force contraction. Electroenceph. clin. Neurophysiol., 1987, 67: 495-497. Sadoyama, T. and Miyano, H. Frequency analysis of surface EMG to evaluation of muscle fatigue. Eur. J. Appl. Physiol., 1981, 47: 239-246. Sadoyama, T., Masuda, T. and Miyano, H. Relationships between muscle fiber conduction velocity and frequency parameters of surface EMG during sustained contraction. Eur. J. Appl. Physiol., 1983, 51: 247-256.
L. ARENDT-NIELSEN ET AL. Shankar, S., Gander, R.E. and Brandell, B.R. Changes in the myoelectrical signal (MES) power spectra during dynamic contractions. Electroenceph. clin. Neurophysiol., 1989, 73: 142-150. St,~lberg, E. Propagation velocity in human muscle fibers in situ. Acta Physiol. Scand., 1966, 287: 3-112. Vredenbregt, J. and Rau, G. Surface electromyography in relation to force muscle length and endurance. In: J.E. Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology, Vol. 1. Karger, Basel, 1973: 607-622. Wilska, A. and Varjoranta, K. Untersuchungen iiber die Geschwindigkeit des geleiteten Aktionspotentials einer Muskelfaser bei verschiedenen Dehnungszustanden des Muskels. Skand. Arch. Physiol., 1940, 83: 82-87. Zwarts, M.J. and Arendt-Nielsen, L. The influence of force and circulation on average muscle fibre conduction velocity during local muscle fatigue. Eur. J. Appl. Physiol., 1988, 58: 278-283.
