Five volunteer subjects held isometric handgrip contractions at specific submaximal tensions until the required tension could no longer be maintained. At the start of those contractions, the amplitude of the surface electromyogram (EMG) was linearly related to the tension exerted; the amplitude of the EMG increased linearly throughout these sustained contractions by a constant amount-about 30% of the maximum. During sustained contractions, brief, intermittent maximal efforts showed that strength declined linearly at all tensions. At 25% maximal voluntary contraction (MVC), there was a linear fall in the EMG amplitude associated with the brief maximal efforts, but the fall in strength was more rapid than the fall in EMG amplitude. At 70% MVC, there was no fall in the EMG amplitude in response to the brief maximal efforts, while the muscle strength fell linearly. MUSCLE & NERVE
2:257-264
1979
AMPLITUDE OF THE SURFACE ELECTROMYOGRAM DURING FATIGUING ISOMETRIC ALEXANDER R. UND, DPhil, DSc, and JERROLD S. PETROFSKY, PhD
A clear-cut linear relationship has been established between the force of brief isometric contractions and the average rectified surface electromyogram (EMG) as long as the muscles are kept at a constaiiLlength.5~1u*’4,18 Also, continued strcnuous isometric activity has beer s h o ~ .to ! ~result in an increase in the amplitude of the integrate-: surface EMG.””’’ This increase has been attributed to recruitment of additional motor units or to higher discharge frequencies in recruited units as the muscle f a t i g ~ e sSo . ~ far, no systematic-study of these phenomena has been combined with attempts to examine the physiologic mechanisms involved. The present investigation examines the re-
From the Department of Physiology, St. Louis University School of Medicine, St. Louis, MO. Acknowledgments: These experiments were supported by U.S.Air Force Grant 72-2362 and U S . Department of Health, Education, and Welfare Contract CDC-99-OSH-183. Address reprint requests to Dr. Lind at the Department of Physiology, St. Louis University School of Medicine, 1402 South Grand Blvd.. St. Louis, MO 63104. Received for publication July 7, 1978; revised manuscript accepted for publication January 15, 1979. 0148-639WO20410257 $oO.oO/O 1979 Houghton Mifflin Professional Publishers @
EMG and isometric Fatip-
lationship between isometric strength and endurance and the amplitude of the surface EMG at a variety of tensions, as well as the mechanisms that control the observed changes. METHODS
Five volunteers, ages 21 to 50 years, served as subjects. All subjects underwent a physical examination and a stress test (ECG) and were considered to be healthy. They were fully informed of the experimental procedures arid signed an informed-consent statement before taking part in the experiments.
Subjects.
Strength and Endurance. Isometric strength and endurance were measured on a handgrip dynamonieter.2 The maximal voluntary contraction (MVC) was assessed as the stronger of two brief (3 sec) maximal eff0rt.s; 3 min were allowed between each contraction. The MVC never varied more than 1% from day to day. Each day’s target tcnsions for the submaximal contractions were set in terms of the MVC for that day. Next, the subjects exerted a submaximal tension either for 3 sec or continuously until they could no longer maintain the required tcnsion (endurance time). In two other cxperiments, they maintained a continued maximal effort after a fatiguing contraction at a submaximal
MUSCLE & NERVE
Jul/Aug 1979
257
tension. The experiment ended when the tension had fallen by 15% of MVC during those continued maximal efforts. Training. All
subjects were trained in handgrip contractions three days per week for four to six weeks. On each training day, after establishing their MVC, the subjects performed five successive contractions to fatigue at a tension of 40% MVC; 3 min were allowed between each contraction. This procedure was repeated until the variation in endurance for three consecutive days was less than 5%. The subjects were then trained, in a similar manner, to sustain isometric contractions to fatigue at tensions varying from 25% to 70% MVC. Each subject’s arm was scrubbed and washed with acetone. The surface EMC, of the forearm was recorded from adhesive silver/silver chloride disc electrodes with an active surface area of 2 cm2.Two bipolar electrodes were placed approximately 8 cm apart over the medial and the lateral forearm surfaces. The DC resistance between the electrodes across the skin was always below 5,000 fl and the capacitance ranged from 0.04 to 0.05 p F . Care was taken to ensure that electrodes were placed in the same spot each day. The amplifier for the EMG had a differential input impedance of lo8 0. The raw EMG was recorded o n an FM tape recorder and analyzed later. A digital computer was used t o analyze the EMG over 1.5-sec periods for (1) peak amplitude, (2) average half-wave amplitude, and (3) root mean square (rms) amplitude. All these measurements were distributed in a similar manner; the results given here refer to the average half-wave amplitude, which in the text below is referred to simply as “amplitude.” For contractions in the first experiment, the EMG was sampled over the middle of the 3-sec interval for 1.5 sec. During any given submaximal tension sustained to fatigue, the data were sampled over 1.5-sec intervals at the onset and at 20%, 40%, 60%, 80%, and 100% of the duration of the contractions. I n those experiments in which a maximal effort (i.e., with rapidly falling tension) was exerted after fatigue was developed at submaximal tensions, t.he EMG was sampled at set periods as described under Results. Surface Electromyography.
I n the first two experiments, we investigated the relationship of the amplitude of the EMG to (1) brief isometric contractions at tensions varying from 5% to 100% ILIVC,
and to (2) contractions held to fatigue at submaximal tensions from 25% to 70% MVC. The remaining t ~ 7 oexperiments were intended to examine some of the mechanisms concerned with the changes in the integrated EMG found in the first two experiments. In all experiments, the subjects sat quietly with their arms bare to the shoulder and their elbows held at an angle of 90”; the environmental temperature was kept constant at 25” 2 1°C. A minimum of 24 hours was allowed between experiments on any subject. RESULTS
After measuring the MVC, 3-sec contractions were performed at tensions between 5% and 100% MVC; 3 min were allowed between contractions. Each subject performed this experiment twice. Although the amplitude of the EMG was greater on the lateral than on the medial surface of the forearm, the pattern of response of the amplitude to the brief tensions (and throughout the sustained submaximal contractions in later experiments) was the same on both surfaces of the arm; therefore, the data from both surfaces of the forearm were pooled. Each point in figure 1 represents the average EMG amplitude recorded from five subjects. T h e peak amplitude and the rms responded in the same fashion. As has been reported by other^,^+^^ we found a linear relationship (r = 0.84, p < 0.01) between the EMG amplitude and the tension exerted in the brief isometric contractions. The data relating strength and EMG amplitude in any one subject were as linear in any experiment as those of the mean results shown in figure 1, but there was a large intersubject variation in the absolute amplitude; for the group, the coefficient of variation was 2870. The coefficient of variation was also high (averaging 14%) for the results from any one subject, in spite of the care taken in preparing the skin arid placing the electrodes. However, this large variation was markedly reduced by normalizing the amplitude of the EMG in terms of the value found in response to the M V C a t the start of each experiment. This maneuver resulted in a reduction of the coefficient of variation for the group to only t4.6%,as shown in figure 1 . All the data from other experiments reported below have been treated in the same way.
Brief Isometric Contractions.
Experimental Procedures.
258
EMG and Isometric Fatigue
Figure 2 illustrates the changes in the EMG amplitude in response to fatiguing isometric contractions at ten-
Submaximal Tensions Held to Fatigue.
MUSCLE & NERVE
JuVAug 1979
Figure 1 Linear relationship between the tension of a brief (3 sec) hand-grip contraction up to the maximal voluntary contraction ( M V C ) and the resultant amplitude of the surface E M G , which is normalized to the maximal value found in response to the M V C Each point represents the mean f SD for two experiments on each of five subjects
% MVC
-
-
1
1 EMG and Isometric Fatigue
1
I
1
I
-
% MVC
Figure 2. Amplitude of the surface E M G during contract/ons sustained to fatigue at various submaximal tensions. At the onset of the contraction, there was a linear relationship between tension and EMG amplitude. The vertical arrows represent the direction and dimension of the linear increase of the EMG amplitude during the sustained contractions. The verbcal bars indicate the SD.The dotted arrows represent the redixtion in EMG amplitude when. after reaching fatigue at 40% and 70% MVC, a maximal effort was maintained until the tension had fallen to 25% and 5 5 % MVC, respectively (see also fig. 3). Each point represents the mean & SD for tlvo experiments on each of five subjects.
MUSCLE & NERVE
Jul/Aug 1979
259
Sustained Maximal Isometric Effort Following Fatigue Generated a t Submaximal Tensions. I n these
experiments, the subjects first held contractions to fatigue at 40% arid 70% MVC. After they had reached fatigue at either tension, the subject.s exerted a maximal effort until the tension had fallen by another 15% MVC (i.e., to 25% and 55% MVC, respectively), as shown in figure 3. During thc regular fatiguing contractions at constant tension, the F.MG amplitude increased linearly as described above (fig. 2). After the 70% MVC, the tension during the sustained effort fell almost linearly to 55% hlVC in an average time of 18 sec; whereas after the 40% MVC, the tension fell exponentially and took 27 sec to reach 255% MVC. As the tension k l l , EMG amplitude fell in parallel (fig. 3). In t.hese experiments, by the time the tensions had fallen to 25% o r 555% MVC, the EMG amplitude had decreased to levels similar to those lound
5
401
= 2 0
f
-
40% MVC
100
0
L I
I
0
1
I
20
40
1
1
60 80
Duration of Cmtraction
I
100
I
1
1
1
5
10
15
O/O
Loss of Tension
F/gure 3 . Changes in isometric tension (A)and the associated amplitude of the surface EMG ( 0 ) when five subjects first sustained handgrip contractions at 40% and 70% MVC to fatigue. Thereafter, the subjects continued to exert a maximal effort, during w h c h the tenson fell; the experiment ended when the tensions had fallen to 25% and 55% MVC, respectively. The abscissa is normalized in two ways. First, during the sustained contractions, it IS normalized to the endurance time; thereafter it is normalized for the fall In tension dtiring the rnaxjmal effort. All experiments were repeated, and the points repr6sent mean values 5 SD.
260
EMG and Isometric Fatigue
at the end of a contraction sustained to fatigue at the respective tensions, as represented by the dotted arrows in figure 2. Changes in Strength and Its Associated EMG during Fatiguing Submaximal Contractions. At set intervals
during fatiguing contractions at submaximal tensions of 25%, 40%, 5556, and 70'10 MVC, subjects exerted brief maximal efforts. The brief maximal efforts exerted during the contractions had little or no effect on the endurance times. Figure 4 shows the results. At each tension, thcre was a linear reduction of the strength left to the muscles throughout the fatiguing contraction. In all cases, the EMG amplitude increased linearly tliroughout the sustained submaximal contractions in the manner described earlier (fig. 2). But, in response to the successive brief maximal efforts during the sustained submaximal contractions, the amplitude varied from one extremc of a h e a r fall throughout the contraction at 25% MVC to the other extreme of no reduction at all from the maximal value during the sustained contraction at 70% MVC. Figure 5 shows the recovery of handgrip strength afler fatigue induced by sustained submaximal contractions, and the associated amplitude of the EMG. Following a contraction at 70% MVC, the strength of these subjects returned quickly to its original MVC; the recovery was nearly complete in 3 min and "as complete in 7 min. Recovery of strength after contractions of 25%, 40%, and 55% MVC was initially somewhat slower, but it too was complete in about 7 min. Absol~terecovery of strength tended t o he faster following fatigue at sustained submaximal tensions of increasing strength. However. since the absolute loss of strength at the point of fatigue was greater as t h e siibmaximal tension decreased, the rate of initial recovery was, in fact, at its greatest following the contraction at 25%, MVC and became slower as the tension of the preceding fatiguing contraction increased. 'l'he recovery of the amplitude of the EMG followed the general pattern described for the recovery 01' strength. Rut t.hc EMG amplitude had rcturned to its rnaximal value within 3 min after fatigue at the three lower tensions; after fatigue at 70% MVC, the amplitude values were always maximal in the recovery period. DISCUSSION
.lhe results of our first experiment were consistent with previous reports that during brief isorrietric
MUSCLE & NERVE
JuVAug 1979
n
0
\
I
0
50
1
100
I
I
I
1
50 100 50 100 Duration of C o n t r a c t i o n
I
1
50
100
t Figure 4 Ampktude of surface EMG for both sustained submaximal contractions and interm/ttent bnef maximal efforts (upper), strength left to the muscle during sustained subrnax/mal contractions (lower) Subjects sustained handgrip contractions to fatigue at 25%, 40%, 55%, and 70% MVC, corresponding data are shown in order from left to right Five subjects took part in each experiment Values are mean f SD
contractions there is a linear relationship between the tension exerted and the amplitude (peak, average, o r rms) of the surface EMG provided the muscle length does not c h a ~ i g e . " ~ , ~ ~Reports , ' ~ , ~ *of' a nonlinear relationship between the tension and the amplitude4~12~1g come mainly from study of the biceps muscle; it must be recognized that different muscles may not always behave in the same fashion, and that it is difficult to design a truly isometric dynamometer for some muscle groups. We also found, as have other^,^,^." that the amplitude of the surface EMG increases when submaximal tensions are sustained; we systematically extended that finding to show that when submaximal contractions are held to fatigue at tensions ranging from 25% to 70% MVC, the EMG amplitude increased consistently by about 30%' of its niaxinial amplitude during a brief MVC. At n o time did the amplitude in sustained contractions exceed that
EMG and Isometric Fatigue
found in response to a brief MVC. In sustained contractions, increase in the EMG amplitude has been considered to be due to an increase in either the number of active motor units, their firing frequency, or b ~ t h . " . ' ~ ~ ~ ~ ' Undoubtedly, the experiment of greatest importance in this investigation was that in which brief maximal efforts were made periodically during sustained, submaximal tensions between 25% and 70% MVC; the results show that the strength of the muscles declined linearly during all those contractions. The EMG amplitude associated with those hrief maximal efforts showed very different responses depending on thc submaximal tension exerted; it is convenient to consider only the changes during 25%) and 70% MVC, since they seem to describe either extreme of a continuum of responses. During the contraction at 70%1 MVC, while the maximal strength declined linearly: the
MUSCLE & NERVE
JuliAug 1979
261
Recovery
Time ( m i n )
Figure 5 Recovery of strength (e) and amphtude of the surface EMG (0)following fabgue in the experiment Illustrated in figure 4 The tensions at which faQgue was achieved are gwen /n each panel Brief maxjmal efforfs were exerted at varying bmes, u p to 20 rnin after fatigue 80th the strength and the amplitude of the surface EMG were recorded Experimental cond/t/ons and the number of subjects are the same as for figure 4
amplitude of the EMG in response to the brief maximal efforts did not fall. T h e customary interpretation placed on experimental data of these kinds is that where there is a fall in tension with no reduction of electrical activity, fatigue resides in the contractile apparatus of the muscle; a parallel fall in both tension and electrical activity is taken to show that the fatigue is in the mechanisms of electrical transmission.’”16*’”On that basis, the results fbund in the sustained 70% MVC must be interpreted to mean that there was no failure of electrical transmission either along the nerve, across the neuromuscular junction or along the m u d e rncmbrane. Thereby, the loss of muscular strength must be due solely to failure in the contractile events in the muscle. In contrast, at the lowest tension examined (25% MVC), the EMG amplitude, in response to the brief maximal efforts, fell linearly but only at about half the rate of the decline in strength. If these findings are construed consistently with the arguments applied to the results from the contraction at 70% MVC, the interpretation must be that fatigue at 25% MVC is due, in
262
EMG and Isometric Fatigue
roughly equal parts, to failure of the contractile events in the muscle and to the loss of electrical activity, whatever its origin. But we find this interpretation hard to accept. If there is to be any failure of electrical transmission, it should be expected at high tensions such as 70% MVC when (1) all or nearly all motor units are active and at high freq~encies,‘,’~ and (2) blood vessels in the forearm arc occluded l o ~ a l l yleaving , ~ ~ ~ the motor units and the unmyelinated portions of the motor nerve to operate in anaerobic conditions. in contrast, it is likely that at the start of a contraction at 25% MVC only about half of the motor units are functional, and their firing frequency is probably quite OW.^^'^ There is certainly some nutritional blood flow to the muscles, albeit insufficient for their metabolic needs.g In view of our reluctance to accept at face value our electromyographic evidence purporting to show electrical failure at 25% MVC but not at 70% MVC, it is worth considering the experimental conditions and the data in more detail. We did not measure the action potential induced by applying an artificial stimulus to motor
MUSCLE & NERVE
Jul/Aug 1979
nerves. This is a relatively complex contraction, in which the active muscles are served by two major nerve trunks, and it would be difficult to interpret the results accurately. But a case
2:257-264
1979
AMPLITUDE OF THE SURFACE ELECTROMYOGRAM DURING FATIGUING ISOMETRIC ALEXANDER R. UND, DPhil, DSc, and JERROLD S. PETROFSKY, PhD
A clear-cut linear relationship has been established between the force of brief isometric contractions and the average rectified surface electromyogram (EMG) as long as the muscles are kept at a constaiiLlength.5~1u*’4,18 Also, continued strcnuous isometric activity has beer s h o ~ .to ! ~result in an increase in the amplitude of the integrate-: surface EMG.””’’ This increase has been attributed to recruitment of additional motor units or to higher discharge frequencies in recruited units as the muscle f a t i g ~ e sSo . ~ far, no systematic-study of these phenomena has been combined with attempts to examine the physiologic mechanisms involved. The present investigation examines the re-
From the Department of Physiology, St. Louis University School of Medicine, St. Louis, MO. Acknowledgments: These experiments were supported by U.S.Air Force Grant 72-2362 and U S . Department of Health, Education, and Welfare Contract CDC-99-OSH-183. Address reprint requests to Dr. Lind at the Department of Physiology, St. Louis University School of Medicine, 1402 South Grand Blvd.. St. Louis, MO 63104. Received for publication July 7, 1978; revised manuscript accepted for publication January 15, 1979. 0148-639WO20410257 $oO.oO/O 1979 Houghton Mifflin Professional Publishers @
EMG and isometric Fatip-
lationship between isometric strength and endurance and the amplitude of the surface EMG at a variety of tensions, as well as the mechanisms that control the observed changes. METHODS
Five volunteers, ages 21 to 50 years, served as subjects. All subjects underwent a physical examination and a stress test (ECG) and were considered to be healthy. They were fully informed of the experimental procedures arid signed an informed-consent statement before taking part in the experiments.
Subjects.
Strength and Endurance. Isometric strength and endurance were measured on a handgrip dynamonieter.2 The maximal voluntary contraction (MVC) was assessed as the stronger of two brief (3 sec) maximal eff0rt.s; 3 min were allowed between each contraction. The MVC never varied more than 1% from day to day. Each day’s target tcnsions for the submaximal contractions were set in terms of the MVC for that day. Next, the subjects exerted a submaximal tension either for 3 sec or continuously until they could no longer maintain the required tcnsion (endurance time). In two other cxperiments, they maintained a continued maximal effort after a fatiguing contraction at a submaximal
MUSCLE & NERVE
Jul/Aug 1979
257
tension. The experiment ended when the tension had fallen by 15% of MVC during those continued maximal efforts. Training. All
subjects were trained in handgrip contractions three days per week for four to six weeks. On each training day, after establishing their MVC, the subjects performed five successive contractions to fatigue at a tension of 40% MVC; 3 min were allowed between each contraction. This procedure was repeated until the variation in endurance for three consecutive days was less than 5%. The subjects were then trained, in a similar manner, to sustain isometric contractions to fatigue at tensions varying from 25% to 70% MVC. Each subject’s arm was scrubbed and washed with acetone. The surface EMC, of the forearm was recorded from adhesive silver/silver chloride disc electrodes with an active surface area of 2 cm2.Two bipolar electrodes were placed approximately 8 cm apart over the medial and the lateral forearm surfaces. The DC resistance between the electrodes across the skin was always below 5,000 fl and the capacitance ranged from 0.04 to 0.05 p F . Care was taken to ensure that electrodes were placed in the same spot each day. The amplifier for the EMG had a differential input impedance of lo8 0. The raw EMG was recorded o n an FM tape recorder and analyzed later. A digital computer was used t o analyze the EMG over 1.5-sec periods for (1) peak amplitude, (2) average half-wave amplitude, and (3) root mean square (rms) amplitude. All these measurements were distributed in a similar manner; the results given here refer to the average half-wave amplitude, which in the text below is referred to simply as “amplitude.” For contractions in the first experiment, the EMG was sampled over the middle of the 3-sec interval for 1.5 sec. During any given submaximal tension sustained to fatigue, the data were sampled over 1.5-sec intervals at the onset and at 20%, 40%, 60%, 80%, and 100% of the duration of the contractions. I n those experiments in which a maximal effort (i.e., with rapidly falling tension) was exerted after fatigue was developed at submaximal tensions, t.he EMG was sampled at set periods as described under Results. Surface Electromyography.
I n the first two experiments, we investigated the relationship of the amplitude of the EMG to (1) brief isometric contractions at tensions varying from 5% to 100% ILIVC,
and to (2) contractions held to fatigue at submaximal tensions from 25% to 70% MVC. The remaining t ~ 7 oexperiments were intended to examine some of the mechanisms concerned with the changes in the integrated EMG found in the first two experiments. In all experiments, the subjects sat quietly with their arms bare to the shoulder and their elbows held at an angle of 90”; the environmental temperature was kept constant at 25” 2 1°C. A minimum of 24 hours was allowed between experiments on any subject. RESULTS
After measuring the MVC, 3-sec contractions were performed at tensions between 5% and 100% MVC; 3 min were allowed between contractions. Each subject performed this experiment twice. Although the amplitude of the EMG was greater on the lateral than on the medial surface of the forearm, the pattern of response of the amplitude to the brief tensions (and throughout the sustained submaximal contractions in later experiments) was the same on both surfaces of the arm; therefore, the data from both surfaces of the forearm were pooled. Each point in figure 1 represents the average EMG amplitude recorded from five subjects. T h e peak amplitude and the rms responded in the same fashion. As has been reported by other^,^+^^ we found a linear relationship (r = 0.84, p < 0.01) between the EMG amplitude and the tension exerted in the brief isometric contractions. The data relating strength and EMG amplitude in any one subject were as linear in any experiment as those of the mean results shown in figure 1, but there was a large intersubject variation in the absolute amplitude; for the group, the coefficient of variation was 2870. The coefficient of variation was also high (averaging 14%) for the results from any one subject, in spite of the care taken in preparing the skin arid placing the electrodes. However, this large variation was markedly reduced by normalizing the amplitude of the EMG in terms of the value found in response to the M V C a t the start of each experiment. This maneuver resulted in a reduction of the coefficient of variation for the group to only t4.6%,as shown in figure 1 . All the data from other experiments reported below have been treated in the same way.
Brief Isometric Contractions.
Experimental Procedures.
258
EMG and Isometric Fatigue
Figure 2 illustrates the changes in the EMG amplitude in response to fatiguing isometric contractions at ten-
Submaximal Tensions Held to Fatigue.
MUSCLE & NERVE
JuVAug 1979
Figure 1 Linear relationship between the tension of a brief (3 sec) hand-grip contraction up to the maximal voluntary contraction ( M V C ) and the resultant amplitude of the surface E M G , which is normalized to the maximal value found in response to the M V C Each point represents the mean f SD for two experiments on each of five subjects
% MVC
-
-
1
1 EMG and Isometric Fatigue
1
I
1
I
-
% MVC
Figure 2. Amplitude of the surface E M G during contract/ons sustained to fatigue at various submaximal tensions. At the onset of the contraction, there was a linear relationship between tension and EMG amplitude. The vertical arrows represent the direction and dimension of the linear increase of the EMG amplitude during the sustained contractions. The verbcal bars indicate the SD.The dotted arrows represent the redixtion in EMG amplitude when. after reaching fatigue at 40% and 70% MVC, a maximal effort was maintained until the tension had fallen to 25% and 5 5 % MVC, respectively (see also fig. 3). Each point represents the mean & SD for tlvo experiments on each of five subjects.
MUSCLE & NERVE
Jul/Aug 1979
259
Sustained Maximal Isometric Effort Following Fatigue Generated a t Submaximal Tensions. I n these
experiments, the subjects first held contractions to fatigue at 40% arid 70% MVC. After they had reached fatigue at either tension, the subject.s exerted a maximal effort until the tension had fallen by another 15% MVC (i.e., to 25% and 55% MVC, respectively), as shown in figure 3. During thc regular fatiguing contractions at constant tension, the F.MG amplitude increased linearly as described above (fig. 2). After the 70% MVC, the tension during the sustained effort fell almost linearly to 55% hlVC in an average time of 18 sec; whereas after the 40% MVC, the tension fell exponentially and took 27 sec to reach 255% MVC. As the tension k l l , EMG amplitude fell in parallel (fig. 3). In t.hese experiments, by the time the tensions had fallen to 25% o r 555% MVC, the EMG amplitude had decreased to levels similar to those lound
5
401
= 2 0
f
-
40% MVC
100
0
L I
I
0
1
I
20
40
1
1
60 80
Duration of Cmtraction
I
100
I
1
1
1
5
10
15
O/O
Loss of Tension
F/gure 3 . Changes in isometric tension (A)and the associated amplitude of the surface EMG ( 0 ) when five subjects first sustained handgrip contractions at 40% and 70% MVC to fatigue. Thereafter, the subjects continued to exert a maximal effort, during w h c h the tenson fell; the experiment ended when the tensions had fallen to 25% and 55% MVC, respectively. The abscissa is normalized in two ways. First, during the sustained contractions, it IS normalized to the endurance time; thereafter it is normalized for the fall In tension dtiring the rnaxjmal effort. All experiments were repeated, and the points repr6sent mean values 5 SD.
260
EMG and Isometric Fatigue
at the end of a contraction sustained to fatigue at the respective tensions, as represented by the dotted arrows in figure 2. Changes in Strength and Its Associated EMG during Fatiguing Submaximal Contractions. At set intervals
during fatiguing contractions at submaximal tensions of 25%, 40%, 5556, and 70'10 MVC, subjects exerted brief maximal efforts. The brief maximal efforts exerted during the contractions had little or no effect on the endurance times. Figure 4 shows the results. At each tension, thcre was a linear reduction of the strength left to the muscles throughout the fatiguing contraction. In all cases, the EMG amplitude increased linearly tliroughout the sustained submaximal contractions in the manner described earlier (fig. 2). But, in response to the successive brief maximal efforts during the sustained submaximal contractions, the amplitude varied from one extremc of a h e a r fall throughout the contraction at 25% MVC to the other extreme of no reduction at all from the maximal value during the sustained contraction at 70% MVC. Figure 5 shows the recovery of handgrip strength afler fatigue induced by sustained submaximal contractions, and the associated amplitude of the EMG. Following a contraction at 70% MVC, the strength of these subjects returned quickly to its original MVC; the recovery was nearly complete in 3 min and "as complete in 7 min. Recovery of strength after contractions of 25%, 40%, and 55% MVC was initially somewhat slower, but it too was complete in about 7 min. Absol~terecovery of strength tended t o he faster following fatigue at sustained submaximal tensions of increasing strength. However. since the absolute loss of strength at the point of fatigue was greater as t h e siibmaximal tension decreased, the rate of initial recovery was, in fact, at its greatest following the contraction at 25%, MVC and became slower as the tension of the preceding fatiguing contraction increased. 'l'he recovery of the amplitude of the EMG followed the general pattern described for the recovery 01' strength. Rut t.hc EMG amplitude had rcturned to its rnaximal value within 3 min after fatigue at the three lower tensions; after fatigue at 70% MVC, the amplitude values were always maximal in the recovery period. DISCUSSION
.lhe results of our first experiment were consistent with previous reports that during brief isorrietric
MUSCLE & NERVE
JuVAug 1979
n
0
\
I
0
50
1
100
I
I
I
1
50 100 50 100 Duration of C o n t r a c t i o n
I
1
50
100
t Figure 4 Ampktude of surface EMG for both sustained submaximal contractions and interm/ttent bnef maximal efforts (upper), strength left to the muscle during sustained subrnax/mal contractions (lower) Subjects sustained handgrip contractions to fatigue at 25%, 40%, 55%, and 70% MVC, corresponding data are shown in order from left to right Five subjects took part in each experiment Values are mean f SD
contractions there is a linear relationship between the tension exerted and the amplitude (peak, average, o r rms) of the surface EMG provided the muscle length does not c h a ~ i g e . " ~ , ~ ~Reports , ' ~ , ~ *of' a nonlinear relationship between the tension and the amplitude4~12~1g come mainly from study of the biceps muscle; it must be recognized that different muscles may not always behave in the same fashion, and that it is difficult to design a truly isometric dynamometer for some muscle groups. We also found, as have other^,^,^." that the amplitude of the surface EMG increases when submaximal tensions are sustained; we systematically extended that finding to show that when submaximal contractions are held to fatigue at tensions ranging from 25% to 70% MVC, the EMG amplitude increased consistently by about 30%' of its niaxinial amplitude during a brief MVC. At n o time did the amplitude in sustained contractions exceed that
EMG and Isometric Fatigue
found in response to a brief MVC. In sustained contractions, increase in the EMG amplitude has been considered to be due to an increase in either the number of active motor units, their firing frequency, or b ~ t h . " . ' ~ ~ ~ ~ ' Undoubtedly, the experiment of greatest importance in this investigation was that in which brief maximal efforts were made periodically during sustained, submaximal tensions between 25% and 70% MVC; the results show that the strength of the muscles declined linearly during all those contractions. The EMG amplitude associated with those hrief maximal efforts showed very different responses depending on thc submaximal tension exerted; it is convenient to consider only the changes during 25%) and 70% MVC, since they seem to describe either extreme of a continuum of responses. During the contraction at 70%1 MVC, while the maximal strength declined linearly: the
MUSCLE & NERVE
JuliAug 1979
261
Recovery
Time ( m i n )
Figure 5 Recovery of strength (e) and amphtude of the surface EMG (0)following fabgue in the experiment Illustrated in figure 4 The tensions at which faQgue was achieved are gwen /n each panel Brief maxjmal efforfs were exerted at varying bmes, u p to 20 rnin after fatigue 80th the strength and the amplitude of the surface EMG were recorded Experimental cond/t/ons and the number of subjects are the same as for figure 4
amplitude of the EMG in response to the brief maximal efforts did not fall. T h e customary interpretation placed on experimental data of these kinds is that where there is a fall in tension with no reduction of electrical activity, fatigue resides in the contractile apparatus of the muscle; a parallel fall in both tension and electrical activity is taken to show that the fatigue is in the mechanisms of electrical transmission.’”16*’”On that basis, the results fbund in the sustained 70% MVC must be interpreted to mean that there was no failure of electrical transmission either along the nerve, across the neuromuscular junction or along the m u d e rncmbrane. Thereby, the loss of muscular strength must be due solely to failure in the contractile events in the muscle. In contrast, at the lowest tension examined (25% MVC), the EMG amplitude, in response to the brief maximal efforts, fell linearly but only at about half the rate of the decline in strength. If these findings are construed consistently with the arguments applied to the results from the contraction at 70% MVC, the interpretation must be that fatigue at 25% MVC is due, in
262
EMG and Isometric Fatigue
roughly equal parts, to failure of the contractile events in the muscle and to the loss of electrical activity, whatever its origin. But we find this interpretation hard to accept. If there is to be any failure of electrical transmission, it should be expected at high tensions such as 70% MVC when (1) all or nearly all motor units are active and at high freq~encies,‘,’~ and (2) blood vessels in the forearm arc occluded l o ~ a l l yleaving , ~ ~ ~ the motor units and the unmyelinated portions of the motor nerve to operate in anaerobic conditions. in contrast, it is likely that at the start of a contraction at 25% MVC only about half of the motor units are functional, and their firing frequency is probably quite OW.^^'^ There is certainly some nutritional blood flow to the muscles, albeit insufficient for their metabolic needs.g In view of our reluctance to accept at face value our electromyographic evidence purporting to show electrical failure at 25% MVC but not at 70% MVC, it is worth considering the experimental conditions and the data in more detail. We did not measure the action potential induced by applying an artificial stimulus to motor
MUSCLE & NERVE
Jul/Aug 1979
nerves. This is a relatively complex contraction, in which the active muscles are served by two major nerve trunks, and it would be difficult to interpret the results accurately. But a case
