Download The Relationship of Motor-Unit Activation to Isokinetic Muscular

The Relationship of Motor-Unit Activation to Isokinetic
Muscular Contraction at Different Contractile Velocities
WILLIAM S. BARNES, PhD
This investigation was designed to examine 1) the relationship between motorunit activation (as recorded by integrated EMG) and speed of contraction and 2)
the relationship between mechanical work, power output, peak torque, average
torque, and both velocity of movement and integrated electromyographic recordings in the elbow flexor muscles. A series of isokinetic contractions of the
elbow flexor muscles was performed by six normal subjects over a range of
contractile velocities. Integrated electromyographic discharge and mechanical
torque were recorded simultaneously. The results of an analysis of variance,
corrected for repeated measures, indicated that both torque and motor-unit
electrical activity decreased as contractile velocity increased. The relationship
between torque and integrated electromyographic activity was linear and highly
significant (r = .95 and r = .93). Implications of a neural interpretation of the in
vivo torque-velocity relationship in muscle are discussed.
Key Words: Contractile proteins, Electromyography, Exertion, Muscle contraction.
The interdependence of muscle tension, velocity of
muscular contraction, and integrated electromyographic activity in contracting muscle has been only
superficially investigated because of the unavailability of suitable equipment.
An isokinetic constant-loading device now permits
exercise at a controlled rate of muscular contraction;
increased muscular output produces increased resistance rather than acceleration. In such a system, the
resistance developed is proportional to the amount of
force exerted. Thus, maximal efforts can be developed
with maximal load being applied at all joint angles
within the range of motion.
Undoubtedly much of the enthusiasm concerning
isokinetic exercise stems from the findings of Moffroid and Whipple, who showed that improved force
production through training depends upon the velocity of the training contractions.1 Isokinetic exercise
was found to increase muscular torque-producing
capability at speeds equal to and slower than the
training speed but not at faster speeds. This finding
has been supported by others.2 The specific nature of
the training effect with regard to speed has given rise
to theories suggesting that different muscle fibers may
Dr. Barnes is Assistant Professor, Department of Physical Education, University of Northern Colorado, Greeley, CO 80631 (USA).
This article was submitted June 23, 1978, and accepted February
13, 1980.
1152
be responsible for torque generation at high speed
than are responsible at slower speeds.
At least two types of skeletal muscle fiber exist in
human muscle.3 The two fiber types have been functionally classified as fast twitch (Type II) and slow
twitch (Type I) on the basis of enzymatic profile,
twitch contraction time, and fatigability. It has been
argued that with slow-speed, high-resistance exercise,
fast-twitch muscle fibers are only minimally involved
and consequently are not subject to the effect of the
training stimulus.4 Concurrently, it is suggested that
the slow-twitch fibers are maximally activated and
respond by increasing their functional capacity.4 The
reverse is proposed with fast-speed, high-resistance
training. Some support for this concept is provided
by Thorstensson and co-workers, who have shown
that high-resistance, low-repetition strength training
does result in specific fast-twitch fiber adaptation
with no change in slow-twitch fibers.5
Inasmuch as muscular contraction is controlled by
the CNS, selective activation of fast-twitch or slowtwitch fibers would have to depend upon unique
neurological recruitment patterns, specific to the
speed of the intended contraction. If these unique
recruitment patterns do exist, then EMG recordings
of motor-unit activation should change for one type
of fiber to the other as the velocity of contraction
changes. Komi found that when maximum contractions of the elbow flexor muscles were performed at
PHYSICAL THERAPY
different constant velocities, the maximum integrated
EMG recordings of both biceps brachii and brachioradialis muscles stayed relatively constant at each
velocity of contraction.6 Komi concluded that activation of the contractile component of a muscle in
maximal contraction was the same regardless of its
contraction velocity. Nelson and associates reported
contradictory results for the contraction velocities of
the anterior tibialis and soleus muscles.7 When contraction velocities were increased from 24°/sec to
216°/sec, the integrated EMG decreased 81.4 percent
for the anterior tibialis muscle and 86.0 percent for
the soleus muscle.
This investigation was designed 1) to examine the
relationship between motor-unit activation (as recorded by integrated EMG) and speed of contraction
and 2) to investigate the relationship between mechanical work, power output, peak torque, average
torque, and both velocity of movement and integrated
EMG recordings in the elbow flexor muscles.
METHOD
Six normal male subjects were selected for the
investigation. All were familiar with the testing technique employed. The exercise device was the Cybex
II* isokinetic dynamometer, described in detail elsewhere.8 Briefly, it consists of a motor that turns an
internal transmission mechanism at a preselected,
nonaccelerable velocity. Contraction velocities between 0 and 300°/sec may be preselected. All force
applied to the input shaft generates motion of both
the limb and shaft. Any tendency for the limb to
accelerate beyond the predetermined speed is absorbed by the internal mechanism, thus maintaining
an accurate constant velocity. All force generated in
excess of that necessary to match the preselected
speed setting is sensed by an internal load cell and
transmitted to a recording device.
In this experiment, a multichannel recording device
(Beckman 612R Dynagraphf) was used to display
permanent torque-production curves. Angular displacement was measured with a potentiometer (Cybex Electrogoniometer*) attached to the lever arm,
thus providing accurate determination of joint position at any moment during the contraction. Output
of the potentiometer was displayed on the dynagraph.
Simultaneous electromyographic activity was recorded and integrated using an integrating high-gain
amplifier (Beckman type 9852af). The resulting analog signal was displayed on the paper-recording
* Lumex Inc, Cybex Division, 100 Spence St, Bay shore, NY
11706.
† Beckman Instruments, Inc, Electronic Instruments Division,
Dept 131, 3900 N River Rd, Schiller Park, IL 60176.
Volume 60 / Number 9, September 1980
device for further analysis. Voltage gain for the amplifier was set at 500 µv/cm, with lower and upper
cut-off frequencies of 5.3 and 1,000 Hz, respectively.
Bipolar surface electrodes (Beckman silver-silver
chloride type†)in a bipolar lead system were applied
to the right arm over the belly of the biceps brachii
muscle. The electrodes were positioned one inch on
either side of the midpoint between the antecubital
space and the axillary fold. The grounding electrode
was applied to the volar aspect of the right wrist.
The subject was seated beside the dynamometer
with the axis of rotation of the right elbow in line
with the axis of the machine. The humerus was
aligned in a vertical position with the forearm placed
alongside the lever arm of the machine. The forearm
was supinated and the wrist affixed to the lever arm
by a padded wrist yoke. In addition to binding the
wrist to the lever arm of the apparatus, the subject
was instructed to grasp a handle projecting from the
far end of the dynamometer lever arm. A shoulder
harness increased trunk stabilility and prevented
movement of the shoulder. Subject position and the
arrangement of the dynamometer was similar to that
employed by Rodgers and Berger.9 Starting position
for the test was 0 degrees of flexion (arm fully extended). The termination of the test was at 120 degrees
of elbow flexion.
The test velocities selected were 60, 120, 180, 240,
and 300°/sec. At each speed the subject was requested
to perform four maximal contractions of the elbow
flexor muscles. All reported measurements represent
the mean of the four contractions. The data accumulated provided values for: peak torque, defined as
the highest torque value recorded through the entire
range of motion; average torque, defined as the average torque produced within one contraction; mechanical work, defined as the total integrated torque curve;
average power, defined as the mechanical work performed in each contraction divided by the contraction
time; peak EMG, defined as the highest integrated
voltage recorded through the entire range of motion;
and mean integrated EMG, defined as the total integrated EMG divided by the contraction time.
RESULTS
The means and standard deviations for peak and
average torque, peak and average integrated EMG,
average power, and mechanical work at each test
velocity are given in the Table. Peak values were
normalized by expressing all values as a percentage
of the highest recorded value obtained, regardless of
contraction speed. Most of the time the highest torque
values were observed at the slowest test speed (60°/
sec), and therefore, these values usually represent 100
1153
TABLE
Results of Study of Isokinetic Elbow Flexion (N = 6)
Contraction Velocity (°/sec)
Variables
Peak IEMG
(% max)
Average IEMG
(% max)
Total IEMG
(% max)
Peak Torque
(% max)
Average Torque
(% max)
Total Work
(% max)
Average Power
(% max)
60°
120°
180°
240°
300°
98.90
2.20
100.00
0.00
100.00
0.00
96.97
6.05
99.97
0.05
100.00
0.00
100.00
0.00
91.85
10.23
92.67
4.90
50.97
3.05
90.07
11.06
88.95
10.95
43.85
6.03
89.00
11.00
74.60
18.38
86.77
9.23
37.50
7.94
79.85
15.15
78.52
11.93
27.10
5.76
78.50
11.90
71.45
11.93
71.97
17.31
25.22
5.27
68.55
16.20
72.90
21.96
18.37
5.21
72.90
22.00
59.80
17.48
66.65
15.97
19.80
3.75
57.12
11.29
57.90
19.15
11.45
2.55
57.90
19.20
percent of peak torque. Comparison of torque production during the series of isokinetic contraction
velocities revealed that the elbow flexor muscles produced the highest peak torque values, as well as the
highest average torque values, at the lowest test velocity (60°/sec). Higher contraction speeds resulted
in lower torque values. A peak torque decrement of
42.88 percent and an average torque decrement of
42.10 percent were found between the fastest and
slowest test speeds. Torque-velocity relationships are
presented in Figure 1. Linear regression analysis indicates a highly significant inverse linear relationship
between both the peak torque (r = -.97) and average
torque (r = —.97) and the contraction velocity.
The integrated EMG output of the elbow flexor
muscles demonstrated a direct inverse relationship to
the increasing isokinetic test velocities. Peak integrated EMG decreased 40.20 percent; average integrated EMG decreased 33.35 percent (Fig. 2). This
finding suggests an inverse linear relationship between both the peak integrated EMG (r = -.99) and
average integrated EMG (r = -.98) and the contraction velocity. The relationship between peak torque
and peak integrated EMG is presented in Figure 3.
The data suggest that torque production and motorunit activation are related in a linear fashion (r =
.95). The relationship between average torque produced during the contraction and average integrated
EMG is illustrated in Figure 4. These results indicate
a highly significant linear relationship (r = .93).
The ratio of measurable mechanical work (total
integrated torque) done to muscle electrical activity
Fig. 1. Torque-velocity relationship for the elbow flexor
muscles (N - 6).
Fig. 2. Relationship between integrated electrical activity
(IEMG) in milliseconds and velocity of contraction (N =
6).
1154
PHYSICAL THERAPY
Fig. 5. Ratio of total work and total integrated EMG (%
max) related to velocity (N = 6).
Fig. 3. Relationship between peak integrated electrical
activity and peak torque (N = 6).
DISCUSSION
If torque produced at high contractile velocities is
a consequence of the neurological recruitment of fast
motor units and if slow motor units are recruited at
slower speeds, then it seems reasonable that the electrical activity manifested at different contractile velocities would reflect these differences in recruitment
patterns. Early investigations by Henneman have suggested a stable, orderly recruitment of motor units in
which the smaller, slower contracting units are recruited initially at low force thresholds.10 As the in-
Fig. 4. Relationship between average integrated electrical activity and average torque (N = 6).
(total IEMG) appears to decline as contraction speed
increases (Fig. 5).
The ratio of average power per contraction to the
average integrated EMG per contraction remains relatively constant (Fig. 6). This finding would appear
to indicate that power output of the muscle is associated with similar amounts of electrical activity regardless of the contraction speed.
Volume 60 / Number 9, September 1980
Fig. 6. Ratio of average power (% max) and average
integrated EMG (% max) related to velocity (N = 6).
1155
tensity of contraction increases, larger motor units are
recruited, until at maximum all available motor units
are activated. The theory has become known as the
"size principle" of motor-unit recruitment. If the size
principle were true for isokinetic muscular contractions, then the quantity of electrical activity in the
muscle might be expected to remain the same regardless of the dynamometer's preselected contractile
speed. Because each isokinetic contraction is maximal, all available motor units should be activated
regardless of the contractile velocity maintained. The
present findings indicate a clearly defined reduction
in muscular electrical activity as contractile speeds
increase (Fig. 2).
Most physiologists agree that there are two neural
mechanisms responsible for controlling muscular contractions: rate coding and recruitment.11 Rate coding
represents the changes in firing frequency of the
active motor units, whereas recruitment indicates the
total number and type of motor unit activated. Investigations by Milner-Brown and associates have suggested that quantitative recruitment accounts for increased tension production at levels below about 30
percent of maximum voluntary contraction.12 Above
this tension level, increases were attributed to increased firing rate. Inasmuch as isokinetic contractions are by nature maximal contractions at different
contractile velocities, it would seem probable that all
available motor units were recruited and that each of
these units had achieved its maximum firing frequency. What, then, accounts for the differences in
measured electrical activity at different contractile
velocities? One explanation could be the existence of
"qualitative recruitment," which is recruitment based
upon specific demand, whereby motor units with
different functional characteristics could be selectively recruited depending upon the exact nature of
the intended contraction. Several investigations have
demonstrated that voluntary recruitment patterns
may change, depending upon the type of contraction
performed.13,14 Grimby and Hannerz found that, on
voluntary initiation of a contraction, the recruitment
order of the motor units may differ, depending upon
the velocity of initiation.14 One type of unit becomes
active during slow, sustained contractions, and an
entirely different unit becomes active during rapid
contractions. This form of specific recruitment, dependent upon contractile velocity, would appear to
explain the differences in motor-unit activation demonstrated in the present study. The differences in
recruitment order cited in the literature, however,
resulted from voluntary controlled velocity changes
initiated by the subjects themselves. Contractile velocity during isokinetic loading is determined externally by the machine setting. In fact, the subject is
1156
always instructed to perform a maximum velocity
contraction regardless of the machine speed selection.
Therefore, it would seem reasonable that an initial
recruitment pattern designed to achieve maximum
velocity would always occur regardless of how the
experimentor chose to manipulate the machine's
speed. The subject's preprogrammed recruitment instructions for very rapid movements would be called
upon to initiate the movement. If this hypothesis is
true, then specificity of training based upon the idea
of qualitative recruitment would be difficult to accept.
An alternative explanation for the differences in muscle electrical discharge reported here might be that
they result from either facilitory or inhibitory effects
of peripheral feedback acting upon the motor-neuron
pool. The proprioceptive feedback resulting from isokinetic contractions at different velocities may substantially influence the quantitative or qualitative
activation of motor units participating in such contractions.
Another possible explanation for the apparently
greater electrical activation observed during the lower
velocity contractions might be the presence of significant agonist-antagonist cocontractions. Antagonist
activity, were it to exist, might be picked up at the
recording electrodes through volume conduction.
This explanation of the present findings seems unlikely. Nelson and associates subtracted the integrated
EMG output of the antagonistic action of the soleus
muscle from the integrated EMG of the tibialis anterior muscle during dorsiflexion and found an 81.5
percent decrease in agonist electromyographic discharge over a series of isokinetic velocities ranging
from 24°/sec to 216°/sec.7 Furthermore, the extent
to which cocontraction exists during ballistic contractions has been questioned and at the present remains
unresolved.15
The torque-velocity relationship constructed from
the present data does not closely follow the classic
hyperbolic force-velocity relationship found by Fenn
and Marsh16 and Hill.17 The classic curve reflects a
very specific inverse relationship between work capacity and speed and was derived from all velocities
of shortening up to and including maximal. Mechanical limitations of the test instrument allowed only a
narrow range of velocities to be investigated in the
present study. Nevertheless, over the range of velocities tested, there appears to be a linear relationship
between torque and velocity that does not exist in
vitro. Marked differences between the in vitro forcevelocity relationship and the in vivo torque-velocity
relationship in human muscle has also been reported
by Perrine and Edgerton.18
A highly significant positive correlation was found
between the amount of torque developed at different
PHYSICAL THERAPY
contractile speeds and the concomitant electrical activity generated by the contracting muscle mass. The
basis for such a relationship is unclear. If the decreased ability to produce torque at higher contraction velocities found in the present study is due to the
intrinsic character of the muscle fiber, then why
should the level of motor-unit activation, measured
electromyographically, so closely mimic the external
work production of the muscle? Bigland and Lippold
have shown muscular tension and the integrated
EMG to be linearly related at constant velocities.19 It
seems reasonable to suggest that the decrease in
torque production at higher contractile velocities reported here is largely due to a corresponding decrease
in muscle-fiber activation. If such is the case, the
mechanochemical factors responsible for the forcevelocity relationship in excised muscle do not appear
to influence muscle in vivo. Perhaps such factors are
automatically compensated for by supraspinal control
mechanisms and in the intact individual do not attain
the same functional significance as in the isolated
muscle preparation. The data suggest a decline in the
ratio of total integrated torque to total IEMG. This
decline may be due to the fact that at higher contraction speeds the muscle is less efficient and can produce
less force per unit activation. However, a more likely
explanation for the decrease in work done per unit of
muscle electrical activity is artifact resulting from the
testing apparatus. The dynamometer only registers
external torque when the machine's lever arm can be
made to exceed the preselected machine speed. If the
subject is unable to accelerate his limb to a speed
greater than the machine speed, the machine's internal load cell records zero torque being produced. This
recording obviously does not mean that the contracting muscle is generating zero force, but rather that all
of the resultant force is being transformed into limb
movement. As the machine's speed setting is progressively increased, more of the muscle's force production goes into generating limb speed and less is responsible for producing externally measured torque.
At high contraction speeds, a greater percentage of
the integrated EMG associated with the contraction
may be involved with limb acceleration. Consequently, a lesser percentage of the total electrical
activity would actually be involved with the external
Volume 60 / Number 9, September 1980
torque recorded. If work is defined as the integral of
the torque curve produced with each contraction,
then it would appear that greater motor-unit activation, that is, more recorded muscle electrical activity,
is necessary to produce decreasing amounts of work.
This interpretation may in fact not be accurate.
The isokinetic loading device allows direct control
over contraction velocity. If it were shown that contractions performed at different speeds resulted from
the recruitment of certain motor units specifically
designed for those speeds, this device would seem to
have great potential for specific muscular rehabilitation. Muscular deficiencies occurring at specific contractile speeds might be identified and corrected by
having the patient exercise isokinetically at the speeds
at which the deficiencies occurred.
In addition to its rehabilitative potential, the isokinetic device appears to have great value in the
investigation of basic principles of muscular contraction. The present investigation provides evidence suggesting that the in vitro force-velocity relationship of
muscle is significantly different from the isokinetically determined in vivo torque-velocity relationship.
Although it is generally accepted that the in vitro
force-velocity relationship is the result of intrinsic
characteristics of the muscle tissue, the in vivo torquevelocity relationship appears to be largely determined
by the neural pattern of motor-unit activation associated with the contraction.
CONCLUSIONS
The results of this investigation have demonstrated
significant differences in the total integrated electrical
activity of muscle contracting at different speeds.
These findings might suggest that the differences are
due to different neurological recruitment patterns.
However, closer scrutiny of the isokinetic after-loading device suggests that alternative explanations for
these findings may exist. Although the present findings tend to support the concept of qualitative recruitment, proof of this phenomenon awaits further investigation. For this reason, physical and occupational
therapists should exercise caution in viewing the isokinetic loading device as a tool for improving muscle
function at specific contractile velocities.
1157
REFERENCES
1. Moffroid MT, Whipple R: Specificity of speed of exercise.
PhysTher 50:1629-1699, 1970
2. Pipes TV, Wilmore JH: Isokinetic vs isotonic strength training
in adult men. Med Sci Sports 7:262-274, 1975
3. Burke RE, Edgerton VR: Motor unit properties and selective
involvement in movement. Exerc Sports Sci Rev 3:31-81,
1975
4. Perrine J: Muscle power output and performance. Read at
Centinela Hospital-National Athletic Health Institute Symposium: Maximizing Performance in Track and Field Sports.
Los Angeles, February 1978
5. Thorstensson A, Hulten B, von Dobelin W, et al: Effect of
strength training on enzyme activities and fiber characteristics in human skeletal muscle. Acta Physiol Scand 96:392398, 1976
6. Komi PV: Relationship between muscle tension, EMG, and
velocity of contraction under concentric and eccentric work.
In Desmedt J (ed): New Developments in Electromyography
and Clinical Neurophysiology. Basel, Switzerland, S. Karger,
1973, pp 596-606
7. Nelson AJ, Moffroid MT, Whipple R: The relationship of
integrated electromyographic discharge to isokinetic contractions. In Desmedt J (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Switzerland, S. Karger, 1973, pp 584-595
8. Moffroid MT, Whipple R, Hofkosh J, et al: A study of isokinetic exercise. Phys Ther 49:735-746, 1969
9. Rodgers KL, Berger RA: Motor-unit involvement and tension
during maximum, voluntary concentric, eccentric, and iso-
1158
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
metric contractions of the elbow flexors. Med Sci Sports 6:
253-259, 1974
Henneman E: Organization of the spinal cord. In Mountcastle
VB (ed): Medical Physiology. St. Louis, C. V. Mosby Co,
1968, pp 1717-1732
Stein RB: Peripheral control of movement. Physiol Rev 54:
215-243,1974
Milner-Brown HS, Stein RB, Yemm R: Changes in firing rate
of human motor units during voluntary isometric contractions.
J Physiol 230:371-390, 1973
Grimby L, Hannerz J: Recruitment order of motor units in
voluntary contraction: Changes induced by proprioceptive
afferent activity. J Neurol Neurosurg Psychiatry 31:565573,1968
Grimby L, Hannerz J: Differences in recruitment order of
motor units in phasic and tonic flexion reflex in spinal man.
J Neurol Neurosurg Psychiatry 33:562-570, 1970
Basmajian JV: Muscles Alive. Baltimore, Williams & Wilkins
Co, 1974
Fenn WO, Marsh BS: Muscular force at different speeds of
shortening. J Physiol 85:277-297, 1935
Hill AV: The heat of shortening and the dynamic constants of
muscle. Proc R Soc Lond [Biol] 126:136, 1938
Perrine JJ, Edgerton RV: Muscle force- and power-velocity
relationships under isokinetic loading. Med Sci Sports 10(3):
159-166, 1978
Bigland B, Lippold OCJ: The relation between force, velocity
and integrated electric activity in human muscles. J Physiol
123:214-224, 1954
PHYSICAL THERAPY