Motor unit recruitment is the process by which different motor units are activated to produce a given level and type of muscle contraction. At minimal levels of muscle contraction (innervation), muscle force is graded by changes in firing rate (rate coding) of individual motoneurons (MNs). At higher levels of innervation, recruitment is accomplished by the addition of different motor units firing at or above physiologic tremor rate. During slowly graded and ballistic increases in force, motor units are recruited in rank order of their size. In addition to MN soma diameter, other factors contribute to the selectivity of MN activation. For la afferent MN activation in the cat, synaptic density and efficacy as well as specific membrane resistance are also rank ordered for slow, fatigue resistant, and fast fatigue motor units with slow motor units recruited first. The central drive for motor unit activation is distributed to all the MNs of the pool serving a given muscle. Sizestructure organization of the MN pool determines the order of recruitment and how MNs interact with each other. Disorders of the motor unit affect recruitment. A method for the clinical electromyographic assessment of recruitment is suggested. Assessment is made at three levels of innervation: minimal contraction for onset and recruitment firing rates; moderate contraction required to maintain the limb against gravity for the maximum number of motor units, their firing rates, and motor unit spikes/s; maximal voluntary contraction (MVC) for detection of high threshold enlarged motor units characteristic of reinnervation and completeness of the interference pattern (IP). Loss of muscle fibers results in early and excessive recruitment at minimal and moderate levels of innervation. Loss of motor units can result in both an increased rate and range of single motor unit firing at all levels of innervation. With reinnervation and enlargement of motor units, firing rates increase significantly and the interference pattern during MVC is incomplete. Key words: recruitment motor unit recruitment innervation contraction MUSCLE & NERVE 14:489-502 1991

AAEM MINIMONOGRAPH #3: MOTOR UNIT RECRUITMENT JACK H. PETAJAN, MD, PhD

Motor unit recruitment is defined as the process by which different motor units are activated to produce a given level and type of muscular contraction. A study of motor unit recruitment is of considerable importance in clinical electromyography. A reduction in the force contribution of motor units

From the Department of Neurology, University of Utah School of Medicine, Salt Lake City, Utah. This publication is a revision of AAEE Minimonograph #3: Motor Unit Recruitment, originally published in 1977. Address reprint requests to American Association of Electrodiagnostic Medicine (formerly the American Association of Ellectromyography and Electrodiagnosis), 21 Second Street S . W , Suite 306, Rochester, MN 55902. Accepted for publication June 30, 1990 CCC 0148-639W911060489%014 $04.00 0 1991 Jack H. Petajan, MD, PhD Published by John Wiley & Sons, Inc.

AAEM Minimonograph #3; Motor Unit Recruitment

as in myopathy, or a reduced number of motor units as in neurogenic atrophy, is expected to influence the recruitment process. In the following discussion of recruitment, mechanisms will be reviewed, the physiology and anatomy of activation will be discussed, an approach to assessment by the clinical electromyographer will be outlined and abnormalities seen in some neuromuscular disorders will be described. MECHANISMS OF RECRUITMENT

The development of a sustained force of muscle contraction depends upon the appropriately times periodic force contributions of individual motor units. The ability to produce a steady low-level force, brief high levels of force, and all gradations in between require the activation of the appropriate number of different motor units firing at their proper rates and the contractile properties of skel-

MUSCLE ti NERVE

June 1991

489

eta1 muscle fibers, all orchestrated to produce the desired result. It is also essential that proprioception be adequate to detect gradations in muscle force sufficient for adaptive motor control. Proprioception alone is sufficient in most people to detect the activation of only a few motor units in small muscles, such as the first dorsal interosseous.*' The force contribution of each motor unit in comparison to the force of maximum voluntary contraction (MVC) is highly variable. It has been demonstrated that for slowly developing muscle contraction, motor units with small force contributions are activated first, then larger ones, and so on, as total force output increases (Fig. 1).32-34250,77p78798.99 In the cat, this results in a slight relative decrease in force contribution as the level of innervation increase~.~~ This system for grading muscle force output makes sense. It permits the neuromuscular control system to deliver small delicate forces required for fine motor control, an even gradation of force at higher levels of innervation, and high levels of sustained maximal contraction. The firing rate of motor units, the rate of development of muscle force in each fiber within the motor unit, and the extent to which muscle contraction is isometric or isotonic are also important.

/

0 O .

011 t

I

20

200

I 2000

Threshold force rg)

FIGURE 1. Twitch tension determined by spike trigger averaging for each recruited motor unit is plotted against total muscle tension of first dorsal interosseous muscle. Scales are both logarithmic and slope is nearly unity indicating a linear relation. Data are from a single subject. Reproduced from Milner-Brown et a176with permission.

490

AAEM Minimonograph #3: Motor Unit Recruitment

In order to sustain a nontremulous contraction, motor units should fire at a rate in relation to each other such that fusion of muscle contraction occurs. The increase in physiologic tremor present during minimal muscle contraction and inability to sustain a steady muscle contraction when motor units are lost, as in denervation, indicates the importance of matching firing rate with contractile properties of the muscle fibers. In the normal motoneuron (MN) pool, firing rate is nicely matched with muscle fiber one-half relaxation time; MNs innervating slow muscles (soleus) fire at slow rates (have low-fusion frequencies), in comparison with MNs innervating fast muscles that fire at high rates and have high-fusion frequencies.I6 As we shall see, still finer accommodations are made by the neuromuscular system in order to achieve delicate control. PHYSIOLOGY AND ANATOMY OF RECRUlTMENT

At minimal effort while recording motor unit action potentials (MUAPs) with needle electrode, it can be demonstrated that the order of recruitment remains essentially constant as innervation is increased and decreased, then increased again over the same range of force.51,52280285 Also, it has been demonstrated in the spinal cat that, in response to Ia afferent stimulation, MNs are recruited in the order of their size.51 This observation was made possible by the fact that the amplitude of nerve action potential spikes of recruited MNs recorded extracellularly are proportional to axon diameter. There is a high positive correlation between M N soma diameter, axon diameter, and conduction v e l o ~ i t y . ' ~ ~ ' ~ ~ ' ~ ~ ' ~ ~ ' ~ The technique of spike trigger averaging (STA) has been helpful in studies of recruitment. STA uses the MUAP of interest to trigger an averager of muscle force produced by that motor unit. STA has been used to quantify the twitch tensions produced by motor units recruited at increasing levels of voluntary innervation in man (Fig. 1). An orderly increase in twitch tension as a function of axon conduction velocity has been demon~ t r a t e d . ' ~ , 'These ~ and other observations have led to the formulation of the "size principle," which states that motor units are recruited in order of size from small to large, most easily demonstrated for slowly developing muscle c o n t r a ~ t i o n . ~ ~ In addition to the order of recruitment, other properties of high- and low-threshold motor units are of interest. In the spinal cat, firing rates within the primary range for low-threshold motor units are significantly lower and more stable than rates

MUSCLE & NERVE

June 1991

for high-threshold units. ‘ 5 2 6 2 The period of afterhyperpolarization (AHP) is longer for low-threshold MNs, which limits maximum firing These units are the “S,” or slow fatigue type.5 Motor units activated at higher levels of innervation fire faster and more irregularly. These are more likely to be the “FF” (fast fatigue), or “FR’ (fatigue resistant) type^.'^'^' The “S” type motor units are most likely the ones followed through their primary range in man during minimal muscle contraction such as that required to maintain the limb against gravity. Under these circumstances, MNs fire at approximately the rate of physiologic tremor (10 to 13 Hz). Rates of firing greater than 25 Hz are above normal under these conditions. 1,7985 Considerably more innervation is required to drive these low-threshold motor units to fire at higher rates, since recruitment of additional motor units is the primary mechanism utilized for an increase in force output (see below). Changes in firing rate provide the basis for fine gradations in the force of muscle contraction, especially at low levels, and is especially important in small muscles. Recruitment of new motor units provides a more coarse 5- to 10-fold gradation in force. l 3 With special recording procedures allowing single MUAPs to be distinguished from within a full IP, increasing firing rate can be detected in low-threshold motor units as force increases to higher levels.29 Also, in the special case of denervation, such low-threshold units can be observed to fire at high Other features of low- versus high-threshold motor units include less slowing of firin rate with sustained firing, along with less fatigue!’ T h e degree of slowing has been shown to be a function of the initial rate.@ With sustained firing, the period of AHP may lengthen. This may explain slowing of rate with fatigue and the relatively smaller effect upon firing rate of reduced AHP upon lowthreshold MNs.43-45799One can demonstrate these effects quite easily in large reinnervated motor units in man. These units require high levels of facilitation to activate, fire at high frequency, do not persist in firing, and slow significantly in rate with sustained How might M N soma diameter influence threshold for activation? There are three theoretic

consideration^.^^ 1. Larger MNs may require higher voltages for activation than smaller ones. 2. The synaptic density or effective synaptic activation may be less for large than for small MNs.

AAEM Minimonograph #3: Motor Unit Recruitment

3. T h e ratio of dendrites to soma o r other differences in dendritic structure o r organization might differentiate small from large MNs. This final consideration would relate back to the first one and possibily explain differences in threshold. Direct measures of MN-specific membrane resistance (RJ and conductance using microelectrodes reveal higher resistance and reduced conductance in small MNs in compared with large ones.1’322,26 A regression line with a slope of 0.83 is obtained when in ut conductance is plotted versus soma dia m e t e r 6 Conductance increases faster, however, with soma diameter than one would predict if R , were the same for all cells.63 Thus, a given amount of current applied to a small M N will produce a greater degree of depolarization in comparison with a large MN. Therefore, assuming symmetric and proportional dendritic and somatic enlargement of the MN, increasing size alone might account for the threshold for activation increasing with soma diameter. With regard to synaptic density and efficacy, the problem becomes more complex. Electrical stimulation of single Ia afferents results in larger excitatory postsynaptic potentials (EPSPs) for small MNs (S type) than large MNs (FF or FR type).22It might be inferred from this observation that selective MN activation occurring with increased levels of synaptic activation results primarily from size differences between MNs. In the case of descending inputs to MNs, interneurons activate most MNs, and functional specialization exists both in synaptic organization, as well as in the distribution of cells within the spinal cord. In primates and man, a monosynaptic relationship exists between cortical MNs and the alpha MNs for distal muscles of the upper extremities (intrinsic hand muscles), in which there is a high degree of selectivity of MN activation8’ and a wide range of control defined as the difference between onset and recruitment frequencies (see next secti~n).~~ RECRUITMENT AND ITS RELATION TO CONTRACTILE PROPERTIES OF MUSCLE FIBERS

The electromyographer focuses attention on the discharge properties of motor units, ie, firing rate (rate coding) and recruitment of different motor units under various conditions of innervation. It is easy to lose sight of the purpose for motor unit activation which is to produce a desired force of muscle contraction developed at a specific rate for the performance of a voluntary act. As Henneman

MUSCLE & NERVE

June 1991

491

has stated: “It is the mechanical events in muscle that should be the centre of attention. The motoneurone pool must be organized to achieve precise timing of these events in single motor units and to bring about optimal combination of them for all types of beha~iour.”~’ Ideally, the force of muscle contraction and/or movement should be recorded during the analysis of recruitment and firing rate. From the instant muscle fibers begin to fire repetitively during sustained muscle contraction, it is most probable that their contractile properties change as the result of fatigue. Firing rate is adjusted to accommodate to such changes in contractile properties. Muscle fibers with low contraction- relaxation rates have lower firing rates than fibers with fast contraction- relaxation rates. If one considers a single motor unit and the minimum firing rate necessary to sustain a nearly constant force, then fusion of contractions will occur at a lower rate for “slow” than for “fast” muscle. Consider next the influence of fatigue as firing is sustained. T h e reuptake of Ca++ released from the sarcoplasmic reticulum (SR) is not 100% efficient. As firing rate increases above those levels required to maintain posture or at levels approaching MVC, myoplasmic Ca+ increases, relaxation rate decreases, and finally contracture develops. Although relatively little is known about the kinetics of Ca++ reuptake into the SR in skeletal muscle, experiments on cardiac muscle indicate that uptake is reduced and available high energy phosphate decreases as contraction continues under conditions of energy deprivation or i ~ c h e m i a . ~ As a consequence, the rate of contraction-relaxation slows and twitch tension decreases. Similar changes have been described in skeletal muscle ‘during sustained contraction associated with high rates of motor unit Skeletal muscle is more difficult to study than cardiac muscle because of higher rates of contraction- relaxation. T h e ability of the SR to sequester Ca++ will define the upper limit of firing for a given motor unit. Experiments designed to examine the influence of sustained maximum voluntary contraction upon firing rate have all demonstrated that firing rate declines as fatigue T h e neuromuscular control system “knows” that maintenance of firing rate at prefatigue levels will result in a greater fall in tension. In elegant experiments examining firing rate during slowly increasing and decreasing tension, De Luc2’ demonstrated that motor units with low firing rates and slower rates of contraction-relax+

492

AAEM Minimonograph #3: Motor Unit Recruitment

ation were recruited before those with higher firing rates and faster rates of contraction-relaxation. During sustained contraction (constant force output), motor unit firing rate declined and fluctuations in firing rate between motor units were highly correlated. This correlation in firing rate fluctuations resulted in small fluctuations in force also positively correlated with the fluctuations in firing rate but displaced in time. From these observations, it was concluded that the drive activating MNs was distributed to all MNs of the pool serving a given muscle with M N size-structure (synaptic density, etc.) determining the order of recruitment. T h e conclusion was further supported by the observation that when force declined, derecruitment occurred first in slow or small MNs that decreased their firing rates beforr the decline in force occurred. Large fast motor units were the last to derecruit. Thus, the same size- structure factors governing recruitment (activation) were operating during derecruitment (inhibition). ‘This phenomenon has also been demonstrated in the cat.24 Another observation demonstrating the adjustment of firing rate as a function of contractile properties was that derecruitment rates for a given motor unit were slower than those for recruitment at the same force level. This can be explained by the slower rates of contraction- relaxation as force decreases compared with force increase. How does the M N know to adjust its firing rate to maximize force output? Feedback to MNs is provided by Ia and I1 muscle spindle afferents, Ib Golgi tendon organ afferents, and Renshaw cells. For both spindle537323*72 and Golgi tendon organ afferent8 homonymous feedback to all MNs of the pool serving the muscle has been demonstrated. Feedback and synaptic connectivity are greatest upon the M N innervating muscle fibers nearest the spindle10311,69 or tendon organ6 that feedback to the whole homonymous M N pool. Activation of a single motor unit can alter the discharge from single muscle spindles or tendon organs. In addition, such feedback returns to influence the output of adjacent motor units. Interaction between motor units has been demonstrated with a second recruited motor unit inhibiting or delaying the firing of a first recruited unit when the two fire within a critical time window.” This window is approximately half the interspike interval equivalent to the recruitment interval (45 to 60 ms). T h e more closely the two motor units fire in time, the more likely one is to

MUSCLE & NERVE

June 1991

see this effect, which can be demonstrated during ordinary needle electrode examination. The presence of doublets or marked enlargement of motor units as seen in neurogenic atrophy enhances the effect. These observations suggest that adjacent spindle unloading or tendon organ activation is responsible. Another mechanism for MN interaction is by means of the Renshaw cell. It has been shown that Renshaw cells producing recurrent inhibition can be activated by a single M N , and that the are more strongly excited by large MNs.65,8l,91,93 Thus, motor units recruited at higher levels of innervation are effective in inhibiting previously recruited motor units. It has been demonstrated that a transient slowing of firing rate in the earliest recruited motor unit occurs following recruitment of a second higher threshold motor unit as force output slowly increases.’* This can be seen to smooth the transition of force associated with recruitment of the larger motor unit. Broman and colleagues’2 have proposed the following scenario: 1. The need to increase muscle force results in increased excitation of the MN pool and a motor unit is recruited. 2. Discharge of the MN causes a disfacilitation (inhibition) of the MN pool via Ia, 11, and Ib afferents but small MNs are inhibited more than large ones. 3. In order to maintain force, central facilitation is increased. 4. The net difference between common MN pool facilitation and disfacilitation of smaller MNs results in an increase in firing rate of later recruited motor units and a decrease in firing rate of earlier recruited MNs. As stated above, the slowing of rate in the earlier recruited motor unit allows for a smoother and more delicate increase in force. In the case of fatigue developed during sustainled MVC, marked slowing of firing rate occurs. Following 1 minute of MVC, it is not unusual to discover that MUAPs cannot be recorded at all for several minutes in biceps brachii muscle while the elbow remains flexed at 45 degrees.42 Fatigue develops more rapidly in high-threshold (fast fatigue) motor units. Both slowing of firing rate and greater synchronization of motor unit firing develops. In distance swimmers, such synchronization occurs only at MVC, but at 60% of MVC, or less, in untrained individual^.^' These athletes have been found to have an increased proportion of fatigueresistant motor units and larger “firing rate leads”

AAEM Minimonograph #3: Motor Unit Recruitment

during rapid reversal of muscle force indicating the presence of motor units with slow contraction-relaxation rates2’ Therefore, both slowing of firing rate and synchronization of motor unit firing seem to be correlated with the presence of fast fatigue motor units, resulting in a net disfacilitation of the M N pool via both Renshaw cell activation and peripheral muscle receptors. An increasing disproportion can be seen to develop between Renshaw cell activation by large MNs and their peripheral force contribution. One might hypothesize that altered muscle spindle afferent activity might play an important role in producing this effect because tendon organ stimulation would lessen as muscle tension declines. During isometric contraction, the contraction-relaxation rate of each motor unit acting against the series elastic component becomes slowed. This should reduce the relative change in Ia input (dynamic spindle input) to the M N pool resulting in disfacilitation (inhibition). Further, contracture developing in fatigued muscle fibers should reduce group I1 afferent input. More investigation of this interesting phenomenon demonstrating the “wisdom”71of the neuromuscular control system must be undertaken before the mechanism of slowed firing rate seen during fatigue and synchronization of firing are completely understood. From the physiologic viewpoint, one must also consider the role of muscle contracture in setting the length of muscle fibers prior to their activation by MNs and changes in muscle membrane conductance occurring with fatigue. T h e factors influencing recruitment and differentiating large from small motor units are shown in Table 1 and Figure 2.

Table 1. Factors influencing recruitment and differentiating large from small motor units. Motor unit type

Size, axon speed Relative twitch speed Relative tetanic force Relative fatigue Specific membrane resistance Specific membrane conductance Synaptic density (efficacy) Histochemical type ATPase *FR: fatigue resistance. fFF: fast fatigue. Key: low, ++ intermediate,

+

S

FR*

t

++

t

t+ t+

+

+ ++t t tt+ I

FFt +t+ t+t t++

ti-

t+t

t+ tt 2A

t++

++

+ +

28

+++ hfgh.

MUSCLE & NERVE

June 1991

493

0

al v)

2. al c .

L 10 Myofibrrllor ATPose Glycolylic Enzymes

-Ea

’c .- 6 ” 4 2

0 0

Recruit

Increasing E f f o r t FIGURE 3. The primary range of motor unit firing rate from onset to recruitment is shown for two motor units. The difference between onset and recruitment rates is greater for distal than proximal muscles. A second motor unit is recruited when the firing rate of the first approaches that of physiologic tremor.e5

FIGURE 2. A three-dimensional diagram that summarizes known physiologic and histochemical data obtained from MNs of the cat medial gastrocnemius muscle. Reproduced from Burke15 with permission.

Other investigators have failed to find a bimodal distribution of interspike intervals. This failure seems to result from maintaining a level of facilitation well above threshold where firing is more stable and the likelihood of a lapse in firing is substantially reduced. Using audiovisual feedback of the MUAP with the subject instructed to resist appropriate force applied to the limb, and recording MUAPs with a needle electrode, it is possible to selectively record

ASSESSMENT OF RECRUITMENT IN HUMAN SUBJECTS

In normal subjects, a comparison of motor unit (MU) firing rates and recruitment for slow ramp and fast twitch muscle contractions reveals interesting and significant differences. During the development of slow muscle contraction, single motor units recorded by needle electrode can be seen to turn on and off and to fire at relatively stable rates as the level of facilitation exceeds threshold. The lowest level of continuous firing has been called the onset or lower limiting rate, (Figs. 3 and 4).85(It is important to differentiate between mean rate, which signifies the presence of a variable periodic function, and frequency, which is invariable.) At this level of activation, the level of facilitation may rise above or drop below M N threshold. As a consequence, pauses or lapses in firing may occur,82so that interspike interval histograms of first dorsal interosseous motor units in subjects using audiovisual feedback of the MUAP to establish control have a bimodal distribution. The lower variability of rate in the “on” condition is clearly separated from longer lapses in interspike interval that occur when facilitation falls below threshold.

494

AAEM Minimonograph #3. Motor Unit Recruitment

Onset

+., Recruitment IOO

msec

m o t o r unit loss

muscle f i b e r loss

111111

1ILLL,1Iu

FIGURE 4. Essential features of motor unit recruitment are shown in the diagram. Onset level of firing is near threshold for activation and lapses in firing may occur. Recruitment rate is that rate present just before recruitment of a second motor unit during slowly increasing muscle force (innervation). Loss of motor units by disease results in the appearance of individual easily discriminated single MUAPs firing at high rates. Loss of muscle fibers results in the recruitment of many different motor units at minimal effort.

MUSCLE & NERVE

June 1991

from a single motor unit. Small adjustments of the electrode position may be necessary to achieve maximum amplitude and minimum rise time of the action potential. At times, two or more different MUAPs may be recorded as a consequence of the chance recording of motor units with nearly identical thresholds and motor unit territories which overlap. In general, this will occur less than 10% of the time in normal subjects. At minimal effort, a monopolar electrode records from a larger volume so that a larger number of MUAPs can be recorded than with a concentric electrode. Using a monopolar electrode, the activation of three or more motor units, with minimal effort or barely perceptible muscle contraction, has been called early recruitment (Fig. 4). If the level of muscle contraction is slowly increased, the motor unit that has initiated contraction will fire more rapidly. Eventually, a second motor unit will be recruited that is within the recording territory of the electrode tip. The firing rate achieved by the first motor unit when the second is recruited is termed the recruitment rates5 (Figs. 3 and 4). A normal recruitment rate depends on having a normal distribution of muscle fibers representing different motor units within the muscle. The presence of normal onset and recruitment firing rates requires the presence of the small MNs that normally initiate muscle contraction. The recruitment firing rate may be very near the upper limit of the primary range for the low-threshold motor units that initiate muscle contraction. In studies of mean firing rate of individual motor units firing as additional motor units were recruited, it was observed that firing rate stabilized in initially recruited units as other units were added to the recruitment pattern. A slight increase in firing rate was observed in the initially recruited units, but the primary mechanism for increasing force output was the addition of more motor units, ie, spatial recruitment.82 This observation implies that recruitment of different motor units takes precedence over an increase in mean rate of firing until recruitment is nearly complete, defined as a level of motor unit activation where it is no longer possible to discriminate individual MUAPs. At this level and beyond, MNs may be driven to fire in their secondary range to rates greater than 50 Hz. At this point, a word should be said about the classification of motor units into “tonic” and “phasic” categories. The original concept of tonic (pos-

AAEM Minirnonograph #3: Motor Unit Recruitment

tural) and phasic (adaptive) motor control arose from investigations of spinal reflexes and their role in postural maintenance. This concept was extended and applied to studies of the variance in motor unit firing rates by a number of investigat o r ~It. was ~ ~ observed that MNs innervating slow muscles, such as soleus, fired at a slow and quite even rate, much like a signal generator. This behavior could be elicited by microelectrode stimulation of the M N in the spinal cat o r during voluntary contraction in man. On the contrary, MNs innervating fast muscles, such as extensor hallucis longus, fired at high rates and more i r r e g ~ l a r l y . ~ ~ Characterization of M N size and threshold for activation revealed small size and low threshold for slow MNs and the opposite for fast MNs.20*58,68 Although intrinsic differences in variability of motor unit firing rate do exist, with fast motor neurons more variable, an important source of variability is the increased probability of stimulus intensity falling below threshold in the case of large Y N s . Stable firing of MNs can be achieved if stimulus intensity is kept above thre~hold.~’ The intermittent firing of large nlotor neurons, seen in neurogenic atrophy with reinnervation, most likely results from the very high levels of facilitation required for their sustained a~tivation.’~ During rapid or ballistic activation, motor units fire repetitively at rates significantly higher than during slow ramp contraction. The order of recruitment from small to large motor units is retained, but the tension threshold for activation is reduced.34 The motor units that initiate contraction usually fire repetitively at rates greater than 50 H z . Rather ~ ~ than reverse the order of recruitment to generate high levels of force very rapidly, recruitment is “compressed.” This is accomplished by a great increase in firing rate of individual units and an increase in the tempo of recruitment of different motor units. Thus, by utilizing the same distribution of synapses and, possibly, the same mechanisms of facilitation, rapid force generation can be achieved. In studies of first dorsal interosseous acting first as agonist in abduction and then as synergist in finger flexion, motor units were found to change their recruitment order depending upon the type of m ~ v e m e n t . ~ Collateral ’ innervation of first dorsal interosseous MNs by finger flexor MNs may provide the “double” innervation necessary for this alteration in recruitment order. Cutaneous inputs into the M N pool (sural nerve or flexor reflex afferent stimulation) via polysynaptic pathways can change the threshold for activation of so-

MUSCLE & NERVE

June 1991

495

leus and gastrocnemius MNs as well as the recruitment order of motor units within each muscle.41~56,97 Stimulation of Ia afferents in tibia1 nerve has been demonstrated to affect the excitability of MNs in soleus muscle, an effect most prominent at low firing rates, without apparent alteration in recruitment order.66 Different inputs to MNs subserve different behavioral objectives, ie, maintenance of posture, avoidance of injury, adaptation to environmental stimuli, and goal-directed behavior, all with both static and dynamic components. A more representative classification of the motor system would view the M N pool serving a specific muscle as the effector for many behavioral responses, all of which can be graded quantitatively and qualitatively with respect to mechanisms of M N activation and their rates of activation. ELECTROYYOORAPHJCASSESSMENT OF RECRUITMENT

In view of the many factors that can influence M N activity, it is essential that the conditions under which motor units are activated be carefully defined when the recruitment process is being examined. Recording conditions should be specified as follows: 1. The MUAPs recorded must be optimally focused, have minimum rise time, and maximum amplitude. 2. The type of electrode must be specified. 3. The maximum number of MUAPs should be recorded by appropriate positioning of the electrode. 4. T h e condition of innervation must be indicated. This will include the position of the patient, the degree of joint flexion, and the action of the muscle; ie, is the muscle acting as an agonist or synergist? Finally, the method for determining the force of muscle contraction should be specified. It is common practice to examine MUAPs at minimal effort and at varying’ levels to MVC, sometimes specifying the force level as a percentage of the force generated at MVC. 5. Factors that can influence recruitment and introduce spurious results must be considered and ruled out. These include pain (inhibits motor unit activation), cocontraction of antagonist muscles (yields false assessment of the force of muscle contraction), movement of the electrode during muscle contraction (MUAP “focus” is lost), and poor subject cooperation. The electromyographer can then examine several

496

AAEM Minimonograph #3: Motor Unit Recruitment

characteristics of motor unit activation during the needle electrode examination: The number of different motor units activated at a given level of innervation-spatial recruitment. The rate of firing of each individual motor unit, rate coding or temporal recruitment for the single motor unit, and the mean rate offirzng for all motor units activated or mean rate coding for all motor units activated (total motor unit spike& divided by the number of different motor units). Total spikesh-a measure of both spatial and temporal recruitment, a visual and auditory estimate of the total number of MUAPs activated per second is usually referred to as the recruitment or inte$erence pattern (IP). It must be noted that this estimate of spikesls does not take into account whether or not the pattern is generated by an increased rate of firing of different motor units, an increase in the number of different motor units activated, or both. If spi+es/s independent of whether such spikes represent single motor units are counted, then complex MUAPs with multiple spikes will be counted in the interference pattern. ABNORMAL RECRUITMENT IN RELATION TO FORCE OF MUSCLE CONTRACTION

At just-perceptible levels of muscle contraction, just sufficient to produce movement at a joint or just detectable by palpation, the needle electrode will record from 1 to 3 single motor units. The ability to activate such a small number of motor units is utilized in jitter studies. At this lowest level of effort it is possible to assess the morphology of the MUAPs and to determine onset and recruitment firing rates (Fig. 2). During the clinical examination of recruitment, it is customary to request the patient to gradually increase the level of muscle contraction against resistance supplied by the examiner until MVC is achieved. The examiner evaluates the number of MUAPs present in the IP in proportion to the level of effort and the force of muscle contraction. This requires only an evaluation of the total number of MUAPs in the IP. A technique for estimating the number is to count the number of spikesls that exceed a iven amplitude. This can be done automatica~y.~ ,73

!

The term early recruitment is used to describe a marked increase in numbers of MUAPs present upon initiating muscle contraction; too many MUAPs are present in the IP in proportion to the level of muscle contraction (Fig. 5). This condition

MUSCLE & NERVE

June 1991

mild atrophy

severe atrophy

myopat hy

‘msec. 100 ‘ 1mv. FIGURE 5. With the elbow flexed at 45” against gravity, the needle electrode records from biceps brachii muscle a normal number of different motor units in a patient with a just-detectable weakness; a single MUAP firing at a fast rate (>20Hz) in a patient just able to support the limb against gravity; and an abundance of small MUAPs in a patient with myopathy.

is common in most myopathic disorders, in which the force contribution of each motor unit is decreased.8.14,27,48,64,67 This state may also be present in end-stage neurogenic atrophy when reinnervated motor units become smaller as muscle fibers are lost as a result of neuronal dyingback or overuse.9 In this situation, early recruitment will be detected while the IP may, otherwise, be incomplete at higher levels of effort.842ss*96 When MNs are lost, either as a consequence of neurogenic atrophy or myopathy, a less-than-norma1 number of MUAPs are present in the IP at moderate to high levels of muscle contraction. Subjective analysis of this abnormality can be difficult. A normal subject may fail to produce a complete IP, owing to poor cooperation, musculoskeletal pain, or pain at the site of needle insertion. The examiner can inquire about the presence of pain or observe patient response. If the muscle seems to be weak and the IP is incomplete, how can the examiner decide whether this is normal or abnormal? It may be possible to elicit a greater level of muscle contraction by simply changing the condition for muscle activation. For example, instead of the patient moving his limb against resistance, he is requested to keep the limb in one position while the examiner applies force. Rather than attend to the force of muscle contraction, the patient can be told to make as much noise as possible when he is provided audio-

AAEM Minimonograph #3: Motor Unit Recruitment

feedback of the MUAPs through the loudspeaker. Such maneuvers may increase the level of effort and result in a complete IP. If the IP remains incomplete at what is assumed to be full effort, then an interpretation of this finding can be made. If strength and other features of the EMG are normal, then the incomplete IP may result from reduced effort. Attention to certain features of the IP can be helpful in interpreting it. An incomplete IP containing identifiable single motor units firing at rates in the range of 20 to 60 Hz is abnormal. Motor units firing at 20 Hz are at a rate approximately 3 SD above the mean rate at 30% maximum isometric contraction (MIC) for biceps brachii and tibialis anterior muscles at which level recruitment is complete, ie, the rate of firing of individual MUAPs cannot be discriminated visually.36 In addition, it has been demonstrated that recruitment order is altered in neurogenic atrophy, with motor units generating high-twitch tensions sometimes recruited first.53 The neuromuscular control system attempts to compensate for the loss of motor units by increasing motor unit firing rate- homeostasis being a normal number of spikesls. This can be seen rarely when early recruitment is present. Most often early recruitment is associated with a full IP at moderate or low levels of muscle contraction. Finally, frequency components of the IP itself may be abnormal. The term “white noise” is used to refer to the normal complete IP. It is, in fact, not “white noise,” a term that is used to describe sounds containing roughly equal components of all frequencies in the audible range. When listening to the complete IP, one cannot hear discrete MUAPs, but can rather hear the “static” sound of numerous superimposed MUAPs. If the IP contains numerous short duration, low amplitude, polyphasic MUAPs, one will hear a “Sh-Sh” sound. On the other hand, if long duration, high-amplitude MUAPs containing major spikes with a long rise time are present, one will hear a rumble, somewhat like boulders being moved down a rapidly moving stream. Investigators have taken advantage of this observation by analyzing the frequency components of the IP in myopathy and neurogenic atrophy. An increased proportion of high- and low-frequency components, respectively, is found in the IP in myopathy and neurogenic atrophy.79 Automated techniques have permitted more quantitative and accurate description of the

MUSCLE & NERVE

June 1991

497

1p.36,37,79 It is still essential to provide well “focused” MUAPs for analysis and to consider the results of such analyses in the light of clinical findings. Without automatic analysis, interpretation at moderate to maximal levels of muscle contraction is highly subjective. It is not unusual to have experienced electromyographers give different interpretations when both are observing the same information on the electromyograph. DIMINISHING SUBJECTIVE ASPECTS OF IP INTERPRETATION

Lower limiting (onset) and recruitment firing rates should be determined for 10 or more motor units during minimal muscle contraction. Minimal muscle contraction should be elicited by requesting the patient to move the limb slowly and gently against gravity. If this is not satisfactory, then the patient is requested to resist gentle manual pressure applied to the appropriate place on the limb. It is preferred that standard limb positions be used just as in manual muscle testing. For example, when assessing biceps brachii, the elbow should be fully extended. Large muscles such as gastrocnemius may be examined by means of isotonic contraction or use of a foot board. It is essential to have a consistent method for examining each muscle, both for assessment of onset and recruitment firing rates, and higher levels of muscle contraction. At minimal levels of muscle contraction, early recruitment can be detected. In disorders characterized by structural or functional loss of muscle fibers, a complete IP may be present. Judgment about exccessive recruitment is based upon one’s experience with the electromyogram at minimal levels of muscle contraction. At this level of innervation, it should be possible to discriminate MUAPs with different waveforms, and to determine their respective firing rates. If this cannot be done, then it is likely that recruitment is excessive, or that the patient is unable to produce low levels of muscle contraction. Single motor units (SMUs) firing at abnormal rates can also be detected at minimal effort. Using a sweep speed of 50 to 100 ms/division, firing rates of SMUs within the incomplete IP can be ascertained. In the case of decreased numbers of motor units, abnormally high rates are found (Fig. 5). In amyotrophic lateral sclerosis (ALS), increased varability during sustained firing is Also, onset and recruitment rates are more variable and higher than normal.86 In neur-

Step 1-

498

AAEM Minimonograph #3: Motor Unit Recruitment

opathy, these rates are high but variability is not.86 A statistically significant number of different motor units should be examined before one concludes that abnormality is present. The finding of increased firing rate in a few motor units may be weighed excessively when other evidence of neurogenic atrophy is present. The extent and degree of abnormality will determine the need for a specific sample size. Ideally, muscle force should be measured concomitantly with the assessment of recruitment and MU firing rates. In a recent study of first dorsal interosseous muscle by Reiners et al,go such measurements in patients with chronic neurogenic atrophy revealed an increased change in firing rate/ unit force increment, increased frequency modulation for 10% increments of residual force, and impairment of rate modulation when maximum motor nerve conduction velocity was below normal. These results quantify what has been the subjective impression of abnormally high range of control (recruitment minus onset firing rates) in patients with neurogenic atrophy. Convenient techniques for measuring muscle force in relation to MU recruitment and firing rate would enhance the sensitivity of these measures in the detection of an abnormal force contribution of diseased motor units. Step 2. The number of different motor units discriminated by amplitude and waveform, their respective firing rates, and the total motor unit spikes/s may be determined with the limb held at 45” against gravity. This level of innervation corresponds approximately to “threshold” levels of activation studied during automated analysis of recruitment during which MUAP characteristics of amplitude, duration, waveform, and firing rate are also determined.”*‘47This is a useful level of innervation because the muscle is “adjusted” to the size of the extremity so that the proportion of M N pool needed to accomplish this task can be expected to be relatively constant. The needle is moved until the maximum number of different motor unit spikes/s are recorded. Cocontraction of the antagonist muscle, painful needle placement, and trick movements may result in factitious findings of increased or decreased recruitment. In our laboratory, normal subjects were found to recruit 4.2 (SD = 1.7) different motor units in biceps brachii muscle using a monopolar needle electrode with individual motor units firing at a mean rate of 10 cycles/s (range = 7 to 11 Hz).

MUSCLE & NERVE

June 1991

Forty (SD = 20) MUAP spike& were recorded with amplitudes >0.2 mV. In a second independent study, 48.3 (SD = 18.7) MUAP spikes/s were recorded in 21 young adult subjects4* Triceps, deltoid, quadriceps, biceps femoris, and tibialis anterior muscles can be evaluated in the same manner by the subject supporting the limb against gravity. In general, it is necessary to provide resistance to the gastrocnemius and soleus muscles in order to activate them. With the limb maintained in a 45” antigravity posture while recording from the primary agonist, conditions associated with muscle fiber loss, such as progressive muscular dystrophy, cause a significant increase in numbers of different motor units and MUAP spikes/s. Values for MUAP spikes/s greater than 100 are common. The antigravity posture is simply another method for corroborating the existence of early recruitment present in myopathy (Fig. 5). Except for the presence of short-duration, low-amplitude, polyphasic, and simple biphasic MUAPs characteristic of muscle fiber loss, recruitment characteristics in severe myopathy may mimic those of neurogenic atrophy (see subsequent text). In patients with neurogenic atrophy and minimal weakness, recruitment may be normal at minimal to moderate levels of muscle contraction since levels of facilitation are not yet high enough to activate high-threshold enlarged motor units. Under these circumstances, it is essential to gradually increase the level of innervation to maximum, observing along the way the recruitment of these enlarged motor units that very often fire at high rates. It is hard to sustain firing, in these motor units, since it is difficult to maintain their level of facilitation above threshold. As in the application of manual muscle testing, it is important to standardize the method of testing. The 45” antigravity posture may be used in this way. In the normal subject, as innervation is slowly increased to reach MVC, it is normally impossible to discriminate individual MUAPs or their respective firing rates. In the case of inadequate numbers of MUAPs available for recruitment, especially in the special case of reinnervation associated with type grouping, motor units will be activated at high levels of facilitation and will fire at high rate^.'^,^^,*^ These motor units may be activated only at high levels of effort and for short periods of time. These abnormal units may appear anywhere along the course of increas-

Step 3.

AAEM Minirnonograph #3: Motor Unit Recruitment

ing muscle contraction, but are best detected at MVC. By listening carefully to the sound of the IP and discriminating the acoustic “signature” of each MUAP, one may be able to detect that the mean motor unit firing rate is increasing faster than the number of different MUAPs recorded. T h e ability to hear and verify, by visual inspection, that one or more motor units are firing above 20 Hz during increasing muscle contraction, is indicative of a decreased number of motor units. This is the essential recruitment abnormality when type grouping is pre~ent.’~’’~ In myopathic disorders, the IP will consist of more high-frequency components. The spikes that are seen will be of shorter duration and lower amplitude than normal. T h e peak-to-peak voltage of the IP will also be lower than normal.g0 Lower than normal frequency components may also be present in the IP in the case of motor unit loss, especially when reinnervation has occurred, in which case high-amplitude, long-duration spikes may be visible in the IP, as mentioned previously. Changes may be noted in the IP as contraction is maintained beyond 30 to 60 s. Slowing of motor unit firing rate and increased synchronization of firing between motor units occurs in association with muscle fatigue.71 Increased amplitude of physiologic tremor occurs, and tremor rate slows over time.70 This phenomenon is greatly enhanced in patients with neuromuscular disorders. PROBLEMS FOR FUTURE RESEARCH The mechanisms by which spatial recruitment and rate coding of the M N are adjusted in response to changes in muscle fiber contraction- relaxation properties and the forces generated by motor units require further investigation. Such inquiry will explore inputs to MNs that have both rapid and slow time constants. Current investigations usually involve examination of M N response to single stimuli, a condition that rarely occurs in nature. Tonic stimulation of a variety of inputs alone and together may induce changes in MN excitability over time that are as yet unknown. A great variety of inputs affect the MN. Previous studies have focused primarily upon analysis of muscle stretch/tension receptors. Future studies may focus on the role of tactile receptors, the interaction of inputs, and the role of attention in motor performance. For example, it has been demonstrated in subjects creating a mental image of movement, that liminal cortical stimulation is

MUSCLE & NERVE

June 1991

499

effective in activating the MNs focused by the “internal ~ o m m a n d . ” ~Such ’ studies can greatly increase our understanding of how MNs are activated to produce adaptive behavior. Motoneuron activation is sometimes automatic and/or reflexive and, at other times, “volitional.” Complex patterns of MN activation can be generated by “long loop” connections that include the thalamus, cerebral cortex, basal ganglia, and cerebellum. The neurophysiologist has utilized relatively simple stimulus- response paradigms for the analysis of mechanisms that might subserve such complex behaviors as grasping or walking. Clearly, it is not possible to understand the complexities of M N activation without analysing more complex behaviors that examine how cortical and other regions interact “endogenously” between areas that “store” motor information and those that act upon it. Computers now make such studies possible. Questions can now be asked of the motor system that were previously unanswer~

able owing simply to the volume of data required for analysis. The MUAP provides a window into the central nervous system (CNS) that permits examination of how MNs converse with each other. The M N also converses with the muscle and plays an important role in modifying its contractile properties. Automated computer analysis of M N firing patterns and recruitment may lead to further understanding of the neurotrophic influence on muscle. In the future, studies will also focus on such subjects as proprioceptive memory, the role of attention and arousal in M N activation, and the study of how multiple inputs to MNs affect their behavior. The routine analysis of motor unit recruitment will, for the most part, be automated and computer assisted. Such explorations of motor unit recruitment should provide electromyographers, neurophysiologists and others with many exciting opportunities for investigation for many years to come.

~~~~~

REFERENCES 1. Adrian ED, Bronk DW: The discharge of impulses in motor nerve fibers 11. The frequency of discharges in reflex and voluntary contractions. J Physiol (Lond) 1929; 67: 119151. 2. Allen DG, Lee JA, Westeblad H: Intracellular calcium and tension during fatigue in isolated muscle fibers from Xenopus Leavus.J Physiol 1989; 415:433-458. 3. Barry WH, Peeters GA, Rasmussen CAF, Cunningham MJ: Role of changes in [Ca”] in energy deprivation contracture. Circulation Res 1987; 61:726-734. 4. Bigland-Ritchie B: EMG and fatigue, in Porter R, Whelan J (eds): Ciba Foundation Symposium 82: Human Muscle Fatigue: Physiologacal Mechanisms. London, Pitman Medical, 1980, p p 130- 156. 5. Binder MD, Kroin JS, Moore GP, Stauffer EK, Stuart DG: Correlation analysis of muscle spindle response to single motor unit contractions. J Physiol (Lond) 1976; 257:325336. 6. Binder MD, Kroin JS, Moore GP, Stuart DG: T h e response of Golgi tendon organs to single motor unit contractions. J Physiol (Lond) 1977; 271:337-349. 7. Binder MD, Stuart DG: Response of Ia and spindle group I1 afferents to single motor unit contraction. J Neurophysiol 1988; 43~621-629. 8. Black JT, Bhatt GP, DeJesus PV, Schotland DL, Rowland LP: Diagnostic accuracy of clinical data, quantitative electromyography and histochemistry in neuromuscular disease: A study of 105 cases.J Neurol Sci 1974; 21:59-70. 9. Borg K, Borg J , Edstrom L, Grimby L: Effects of excessive use of remaining muscle fibers in prior polio and LV lesion. Muscle Nerve 1988; 11:1219-1230. 10. Botterman BR, Hamm TM, Reinberg RM, Stuart DG: Localization of monosynaptic Ia excitatory post-synaptic potentials in the motor nucleus of the cat biceps femoris muscle.] Physiol (Lond) 1983; 338:355-377. 11. Brink EE, Jinnai K, Wilson VJ: Pattern of segmental monosynaptic input to cat dorsal neck motoneurons. J Neurophysiol 198 1; 46:496-505. 12. Broman H, DeLuca CJ, Mambrito B: Motor unit recruit-

500

AAEM Minimonograph #3: Motor Unit Recruitment

ment and firing rates interactions in the control of human muscles. Brain Res 1985; 337-319. 13. Brooks VB: Neural Basis of Motor Control. New York, Oxford University Press, 1986, p 63. 14. Buchthal F, Pinelli P: Muscle action potentials in polymyositis. Neurology (Minneap) 1953; 3:424-436. 15. Burke RE: Firing patterns of gastrocnemius motor units in the decerebrate cat.] Physiol (Lond) 1968; 196:631-654. 16. Burke RE: Motor units: Anatomy, physiology and functional organization, in Brooke VB (ed): Handbook of Physiology, Section I : The Nervous System, Vol 2: Motor Systems. Washington, DC, American Physiol Soc, chap 10, 1981, p p 345-422. 17. Burke RE, Dunn RP, Fleshman JW, Glenn LL, Levtov A, O’Donovan MJ, Pinter MD: An HRP study of the relation between cell size and motor unit type in the cat ankle extensor motoneurons. J Comp Neurol 1982; 209: 17-28. 18. Burke RE, Levine DN, Sakman M, Tsairis P: Motor units in cat soleus muscle: Physiological, histochemical and morphological characteristics. J Physiol (Lond) 1978; 238:503519. 19. Burke RE, Levine DN, Tsairis P, Lajac RE: Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J Physiol (Lond) 1973; 234:723-748. 20. Burke RE, Strick PL, Kandor K, Kim C, Walmsley B: Anatomy of medial gastrocnemius and soleus motor nuclei in cat spinal cord. J Neurophysiol 1977; 40:667-680. 21. Burke RE, Tsairis P: T h e correlation of physiological properties with histochemical characteristics in single motor units. Ann NY Acad Sci 1978; 228: 145- 159. 22. Burke RE, Walmsley B, Hodgson JA: Structural-functional relations in monosynaptic action on spinal motoneurons, in Asanuma H, Wilson VJ (eds): Integration in the Nervow System. Tokyo, Igaku-Shoin, 1970, pp 27-45. 23. Cameron WE, Binder MD, Botterman BR, Reinkind RM, Stuart DG: Motor unit- muscle spindle interaction in active muscles of the decerebrate cat. Neurosci Lett 1980; 19:55-60. 24. Clamman HP, Gillies JD, Henneman E: Effects of inhibi-

MUSCLE & NERVE

June 1991

tory input on critical firing level and rank order of motoneurons. J Neurophysiol 1974; 37: 1350- 1360. 25. Clamman HP, Henneman E: Electrical measurement of axon diameter and its use in relating motoneuron size to critical firing level. J Neurophysiol 1976; 39:844-855. 26. Cullheim S: Relations between cell body size, axon diameter and axon conduction velocity of cat sciatic S-motoneurons stained with horseradish peroxidase. Neurosci Lett 1978; 8117-20. 27. Daube JR: Electrodiagnosis of Muscle Disorders, in Engel AG, Banker BQ (eds): Myology, Basic and Clinical. New York, McGraw-Hill Book Co, 1986, pp 1081-1121. 28. Dengler R, Stein RB, Thomas CK: Axonal conduction velocity and force of single human motor units. Muscle Nerve 1988; 11: 136- 145. 29. DeLuca CJ, Lefever RS, McCue MP, Xenakis AP: Control scheme governing concurrently active motor units during voluntary muscle contraction. J Physiol 1982; 329: 129142. 30. Desmedt JE: T h e size principle of motoneuron recruitment in ballistic ramp voluntary contraction in man, in Desmedt JE (ed): Motor Unit Types, Recruitment and Plasticity in Health and Disease, Progress in Clinical Neurophysiology. Basel, Karger 1981, vol 9, p p 97-136. 31. Desmedt JE: T h e principles of motoneuron recruitment and the calibration of muscle force and speed in man, in Desmedt JE (ed): Motor Control Mechanisms in Health and Disease New York, Raven Press, 1983, p p 227-251. 32. Desmedt JE, Godaux E: Fast motor units are not preferentially activated in rapid voluntary contractions in man. Nature (Lond) 1977; 267:717-719. 33. Desmedt JE, Godaux E: Ballistic contractions in fast or slow human muscles: Discharge patterns of single motor units.] Physiol (Lond) 1978; 285:185- 196. 34. Desmedt JE, Godaux E: Recruitment patterns of single motor units in the human masseter muscle during brisk jaw clenching. Arch Oral B i d 1979; 24: 17 1 - 178. 35. Desmedt JE, Godaux E: Ballistic contractions in man: Characteristic recruitment patterns of single motor units of the tibialis anterior muscle. J Physiol (Lond) 1977; 264:673693. 36. Dorfman LJ, Howard JE, McGill KC: Influence of contractile force on properties of motor unit action potentials: ADEMG analysis. J Neurol Sci 1988; 86:125- 136. 37. Dorfman LJ, McGill KC: AAEE minimonograph #29: Automatic quantitative electromyography. Rochester, MN, Am Assoc Electromyogr and Electrodiagn, 1988. 38. Eccles RM, Lundberg A: Integrative patterns of Ia synaptic actions in motoneurones of hip and knee muscles. J Physiol (Lond) 1958; 144:271-298. 39. Fisher MH, Millner-Brown HS, Garcia HS: Automatic analysis of motor unit potentials with varying loads. Electromyogr Clin Neurophysiol 1985; 25(7-8):553-556. 40. Gandevia SC, Rothwel JC: Knowledge of motor commands and the recruitment of motoneurones. Brain 1988; 110:1117-1130. 41. Garnett R, Stephens JA: Changes in the recruitment threshold of motor units produced by cutaneous stimulation. J Physiol (Lond) 1987; 311:403-474. 42. Gooch JL, Petajan JH: Changes in motor unit recruitment after maximum voluntary contraction, abstract AAEE. Mmcle Nerve 1988; 11:9:970. 43. Granit R, Renkin B: Net polarization and discharge rate of motoneurones as measured by recurrent inhibition. Physiol (Lond) 1961; 158:461-475. 44. Granit R, Kornell D, Shortess GK: Quantitative aspects of repetitive firing of mammalian motoneurones, caused by injected currents.J Physiol (Lond) 1963; 168:911-931. 45. Granit R, Kernell D, Shortess GK: T h e behavior of mammalian rnotoneurones during long lasting orthodromic, antidromic and transmembrane stimulation. J Physiol (Lond) 1964; 169:743-754.

AAEM Minimonograph #3: Motor Unit Recruitment

46. Granit R, Phillips CG, Skoglund S, Stag G: Differentiation of tonic and phasic ventral horn cells by stretch, pinna and crossed extensor reflex. J Neurophysiol 1987; 20:470-481. 47. Hannerz J : Discharge properties of motor units in relation to recruitment order in voluntary contraction. Acta Physiol Scand 1974; 9 1:374-384. 48. Hausmanova-Petrusewicz I, Jedrzejewska H: Correlation between electromyographic findings and muscle biopsy in cases of neuromuscular disease. J Neurol Sci 1971; 13:85106. 49. Henneman E: Organization of the motoneuron pool: T h e size principle, in Mountcastle VB (ed): Medical Physiology, ed 14. St Louis, CV Mosby, 1974, vol 1, p 732. 50. Henneman E: The size principle of motoneuron recruitment, in Desmedt JE (ed): Motor Unit Types, Recruitment, and Plasticity in Health and Disease. Progress in Clinical Neurophysiology Basel, Karger 1981, vol 9, p p 26-60. 51. Henneman E, Mendell LM: Functional organization of the motoneuron pool and its inputs; in Brooks VB (ed): Handbook of Physiology, Sec 1: The Neruom System, uol2.: Motor Syst e m , Washington DC, Am Physiol SOC,1981, chap 11, p p 423-507. 52. Henneman E, Shahani BT, Young RR: Voluntary control of human motor units, in Shahani (Bombay) (ed): Symposium on Motor Control. Amsterdam, Elsevier, 1976, p p 7378. 53. Herdmann J, Reinerz, Freund J: Motor unit recruitment order in neuropathic disease. Eletromyogr Clin Neurophysiol 1988; 28~53-60. 54. Iaizzo PA, Seewald M, Oakes SG, Lehmann-Horn F: The use of Fura-2 to estimate myoplasmic Cai+ in human skeletal muscle. Cell Calcium 1989; 10:151- 158. 55. Jakobsson F, Borg K, Edstrom L, Grimby L: Use of motor units in relation to muscle fiber type and size in man. M m cle Nerve 1988; 11:1219-1230. 56. Kanda K, Burke RE, Walmsley B: Differential control of fast and slow twitch motor units in the decerebrate cat. Exp Brain Res 1977; 29:57-74. 57. Kernell D: Input resistance, electrical excitability and size of ventral horn cells in cat spinal cord. Science 1966; 152:1627- 1640. 58. Kernell D: Functional Properties of spinal motoneurons and gradations of muscle force, in Desmedt JE (ed): Motor Control Mechnnisms in Health and Disease. 1983, pp 213-226. 59. Kernell D: Rhythmic properties of motoneurons innervating muscle fibers of different speed in gastrocnemius medialis of the cat. Brain Res 1979; 160:159- 162. 60. Kernell D: T h e limits of firing frequency in cat lumbosacral motoneurone possessing different time course after hyperpolarization. Acta Physiol Scand 1965; 65:87- 100. 61. Kernell D: Functional properties of spinal motoneurons and gradation of muscle force, in Desmeddt JE (ed): Motor Control Mechanisms in Health and Disease. New York, Raven Press, 1983, p p 213-226. 62. Kernell D: Rhythmic properties of motoneurones innervating muscle fibers of different speeds in muscle gastrocnemius medialis of the cat. Brain Res 1979; 160:159- 162. 63. Kernell D, Zwagstra B: Input conductance, axonal conduction velocity and cell size among hind limb motoneurones of the cat. Brain Res 1981; 204:311-326. 64. Kimura J: Electrodiagnosis in Diseases of Nerue and Muscle: Principles and Practice. Philadelphia, FA Davis, 1983. 65. Kudina LP, Pantseva RE: Recruitment inhibition of firing motoneurones in man. Electroencephalog-r Clin Neurophysiol 1988; 69: 179- 185. 66. Kudina LP: Excitability of firing motoneurons tested by Ia afferent volleys in human triceps surae. Electroencephalogr Clin Neurophysiol 1988 ; 69 :576 - 580. 67. Kugelberg E: The electromyogram in muscular dystrophies: Differentiation between dystrophic and chronic lower motor neurone lesions. J Neurol Neurosurg Psychiatry 1949; 121129- 136.

MUSCLE & NERVE

June 1991

501

68. Kuno M: Excitability following antidromic activation in spinal motor neurons supplying red muscles. J Physiol (Lond) 1959; 149:374-383. 69. Lucas SM, Binder MD: Topographic factors in distribution of homonymous group la afferent input to cat medial gastrocnemius motoneurones. J Neurophysiol 1984; 3 1:66-79. 70. Marsden CD: T h e mechanisms of physiological tremor, in Desmedt JE (ed): Progress in Clinical Neurophysiology: Physiologacal Tremor, Patholopcal Tremor and Clonus. Basel, Karger, 1978, p p 1-16. 71. Marsden CD, Meadows JC, Merton PA: “Muscular wisdom” that minimizes fatigue during prolonged effort in man: Peak rates of motoneuron discharge and slowing of discharge during fatigue, in Desmedt J E (ed): Advances in Neurology: Motor Control Mechanisms in Health and Disease. New York, Raven Press, 1983, vol 9, pp 169-212. 72. McKeon B, Burke B: Muscle spindle discharges in response to contraction of single motor units. J Neurophysiol 1983; 49:291-302. 73. Middleton L, Lovelace RE: Use of monopolar electrode in automatic EMG analysis with ANOPS computer. Electromyogr Clan Neurophysiol 1985; 25(7 - 8):539- 548. 74. Miller RG, Sherrat M: Firing rates of human motor units in partially denervated muscle. Neurology 1978; 28: 12411248. 75. Milner-Brown HS, Stein RB, Yemm R: T h e contractile properties of human motor units during voluntary muscle isometric contraction. J Physiol (Lond) 1973; 228:285-306. 76. Milner-Brown HS, Stein RB, Yemm R: T h e orderly recruitment of motor units during voluntary isometric contractions. J Physiol (Lond) 1973; 230:359-370. 77. Milner-Brown HS, Stein RB, Yemm R: Changes in firing rate of human motor units during linearly changing voluntary contractions. J Physiol (Lond) 1973; 230:371-390. 78. Monster AW, Chan H: Isometric force production by motor units of extensor digitorum communis in man. J Neurophysiol 1977; 40: 1432- 1443. 79. Nandedkar SD, Sanders DB, Stalberg EV: Automatic analysis of the electromyographic interference pattern. Part 11. Findings in control subjects and in some neuromuscular diseases. Muscle Nerve 1986; 9(6):491-500. 80. Person RS, Kudina LP: Discharge frequency and discharge pattern of human motor units during voluntary contraction of muscle. Electroencephalogr Clin Neurophysiol 1972; 32:47 1-483. 81. Petajan JH: Motor unit control in movement disorders, in Desmedt JE: Motor Control Mechanisms in Health and Disease, New York, Raven Press, 1983, pp 897-906. 82. Petajan JH: Motor unit frequency control in normal man, in Desmedt JE (ed): Motor Unit Types, Recruitment and Plasticity in Health and Disease. Progress in Clinics Neurophysiology Basel, Karger 1981, vol 9, p p 184-200. 83. Petajan JH: Clinical electromyographic studies of diseases of the motor unit. Electroencephalogr Clin Neurophysiol 1974; 361395-401.

502

AAEM Minimonograph #3: Motor Unit Recruitment

84. Petajan JH: Antigravity posture for analysis of motor unit recruitment: T h e “45” degree test.” Muscle Nerve 1990; 13:355- 359. 85. Petajan JH, Phillip BA: Frequency control of motor unit action potentials. Electroencephalogr Clin Neurophysiol 1969; 27166-72. 86. Petajan J H , Thurman DJ: EMG and histochemical findings in neurogenic atrophy with electrode localization. J Neurol Neurosurg Psychiatry 1981; 44: 1050- 1053. 87. Phillips CG, Porter R: Corticospinal Neurones: Their Role in Movement. New York, Academic Press, 1977. 88. Pinelli P, Buchthal F: Muscle action potentials in myopathies with special regard to progressive muscular dystrophy. Neurology (Minneap) 1953; 3:347-359. 89. Pompeiano 0 , Wand P, Sontag KH: The relative sensitivity of Renshaw cells to orthodromic group Ia volleys caused by static stretch and vibration of extensor muscles. Arch Ztal Biol 1973; 113:238-279. 90. Reiners K, Herdmann J , Freund H: Altered mechanisms of muscular force generation in lower motor neuron disease. Muscle Nerve 1989; 12:647-659. 91. Rosenfalk H: Evaluation of the electromyogram by mean voltage recording, in Medical Electronics: Proc 2nd Znt. Conference on Medical Electronics, Paris, 1959, pp 9- 12. 92. Ross HG, Cleveland S, Haasi J: T h e contribution of single motoneurons to Renshaw cell activity. Neurosci Lett 1975; 1 :105- 108. 93. Ryall RW, Piercey M, Popolosa C, Goldfarb J: Excitation of Renshaw cells in relation to orthodromic and antidromic excitation of motoneurons. J Neurophysiol 1972; 35: 137148. 94. Walmsley B, Hodgson JA, Burke RE: Forces produced by medial gastrocnemius and soleus muscles during locomotion in freely moving cats. J Neurophysiol 1976; 39:844855. 95. Warmolts JR, Engel WK: Open-biopsy electromyography. Correlation of motor unit behavior with histochemical muscle fiber type in human limb muscles. Arch Neurol 1972; 2 7 ~ 125 5 18. 96. Warmolts JR, Mendell JR: Open-biopsy electromyography: Direct correlation of a pattern of excessively recruited, pathologically small motor unit potentials with histologic evidence of neuropathy. Arch Neurol 1979; 36:406-410. 97. Wyman RJ, Waldron J, Wachtel GM: Lack of fixed order of recruitment in cat motoneuron pools. Exp Brain Res 1974; 20:101-114. 98. Yemm R: The orderly recruitment of motor units of the masseter and temporalis muscles during voluntary isometric contraction in man.] Physiol (Lond) 1977; 265: 163- 174. 99. Zucker RA: Theoretical implications of the size principle of motoneurone recruitment. J Theor Biol 1973; 37:587596.

MUSCLE & NERVE

June 1991

AAEM minimonograph #3: motor unit recruitment.

Motor unit recruitment is the process by which different motor units are activated to produce a given level and type of muscle contraction. At minimal...
1MB Sizes 0 Downloads 0 Views