Experimental procedures are described that were designed to assess central motor disorders quantitatively. Initially delineated is a study of triggered ballistic movement performed in a reaction-time situation. The reciprocaltriphasic EMG pattern recorded in an antagonistic muscle pair was clearly abnormal in Parkinsonian patients; increased duration, diminished synchronization of motor units, and a tendency for coactivation of agonist and antagonist muscles and for action tremor were observed. Transport time of elbow movement was prolonged, particularly in a choice reaction-time situation. Rapid, passive displacements of the forearm combined with excitability measurements of hindlimb motoneurons in monkeys revealed the existence of a transcortical loop that may contribute to increased muscle tone in Parkinsonian patients. MUSCLE & NERVE
1:407-412
1978
ELECTROMYOGRAPHIC ASSESSMENT OF CENTRAL MOTOR DISORDERS MARIO WIESENDANGER, MD, and DIETER G. ROEGG, PhD
Electromyography is now a well-established tool in the recognition and evaluation of peripheral neuromuscular disorders. In assessing central deficits of movement and posture, however, it has been much less useful. This is true partly because voluntary, as well as reflex, movements and posture are complex and difficult to reproduce consistently. Moreover, a different experimental approach is required for each of the many different facets of central control-for example, muscle tone and stability of posture; timing of movement initiation; speed, force, or precision of movement performance; and strategy and economy of movement. Since management of central motor disorders has been enhanced by the advent of new therapeutic procedures, such as stereotaxic surgery, dopa medication, and dorsal-column and cerebellar stimulation, it is important to establish objective criteria for assessing therapeutic effects quantitatively. In the present paper, we wish to summarize two aspects of motor control: (1) the pattern of innervation and temporal structure of EMG activity in simple ballistic movements, and (2) reflex activity evoked by an external perturbation occurring during the performance of a motor task.
BALLISTIC MOVEMENTS I N NORMAL SUBJECTS AND IN PARKINSONIAN PATIENTS
From the lnstitut de Physiologie, Universit6 de Fribourg, Switzerland
The Pattern of Innervation. Figure 1 illustrates a typical
Acknowledgments Foundation
This study was supported by the Swiss Science
Address reprint requests to Dr Wiesendanger at the lnstitut de Physiologle, PBrolles, Universit6 de Fribourg, CH-1700 Fribourg, Switzerland 0148-639X/0105/0407 $00 OO/O a 1978 Houghton Mifflin Professional Publishers
Central Motor Disorders
In a series of experiments, we studied the slowing of voluntary movements of parkinsonian patients by measuring response latencies of visually triggered elbow movements. In each subject, we compared the response latencies of movements elicited by a single light signal (simple reaction time) with the response latencies of movements occurring after brightness discrimination of two light signals (complex or choice reaction time). The method has been fully d e ~ c r i b e d . ~ . ' ~ , ' ~ The parameters measured are best illustrated in a recording of the raw data (fig. 1). The visual "go" signal occurred at the beginning of the rectangular pulses of the third and fourth trace (also marked by small bars in the EMG recordings). Response latencies determined in the electromyogram are termed EMG latencies. The first rectangular pulse is terminated when the hand leaves the starting platform. This signal thus measures the movement latency. The second rectangular pulse terminates when the hand reaches the target key; it thus indicates the contact latency. The difference between contact latency and movement latency is the transport time. In the complex (choice) reaction, the fifth trace indicates whether or not the choice was correct.
trial of a nonnal subject. The triphasic pattern fits well with the classic description of Wacholder:lo the activity of the agonist (biceps) consists of a burst, often interrupted by a brief silent period. Activity of the antagonist (triceps) is usually weaker, occurs about 20 msec later, and reaches its maximum during the silent period of the
MUSCLE & NERVE
SepVOct 1978
407
Figure 1. Performance of a normal subject in a simple (A) and a choice (6) reaction-time situation. The visual trigger signal occurred at the onset of the rectangular pulses. a = biceps EMG, b = triceps EMG, c = movement onset latency, d = contact latency, e = correct response (choice reaction). Note the typical trrphasic EMG pattern in the antagonistic muscle pair.
Figure 2. Performance of a parkinsonian patient with moderate akinesia in a simple (A) and a choice (6)reaction-time situation. Same representation as in figure 1. Note prolonged EMG activity with cocontraction and action tremor.
-
Response Latencies. Figure 3 summarizes the results obtained in normal subjects (left) and in parkinsonian patients (right) of varying age classes. It is evident from this graph that reaction time, as measured by EMG, was not significantly different between normal subjects and patients for both simple and complex reactions. The most pronounced difference was noted for the contact latency in the complex reaction time situation (open triangles in fig. 3C). Thus, slowing of movement is more pronounced in parkinsonian patients if movement depends each time on a new decision to move to the right or left key, positioned under the two light signals. On the average, movement latencies were only slightly retarded. Since EMG response latencies of parkinsonians were in the normal range, the delayed onset of movement was
Central Motor Disorders
mms
e
I t
L
C
I
d
looms
biceps. Movement is terminated by a mechanical stop (the target key); EMG activity usually outlasts the movement by 100-200 msec (sometimes in repetitive small bursts). In parkinsonian patients the most common finding, even in early stages of the disease, was the less synchronized activity of both agonists and antagonists, which is in obvious contrast to the reciprocal, triphasic pattern of normal subjects (fig. 2). A burst of activity occurs and corresponds to action tremor occurring synchronously in the agonist and the antagonist. EMG activity outlasts the movement (halted by the mechanical stop) by several hundred milliseconds.
408
-
e
I
an outcome of the temporally dispersed EMG pattern, which resulted in a slow buildup of force. In summary, EMG analysis of a rapid volitional movement, performed in a simple and in a complex reaction-time situation, revealed the following abnormalities in parkinsonian patients:
1. Prolonged-i.e., temporally dispersed-EMG activity was observed in both agonist and antagonist muscles, and was often characterized by coactivation and action tremor. Movement initiation was delayed in some patients for lack of a rapid force buildup, rather than because of a delay in onset of the first voluntary EMG response. 2. The slowing was more pronounced for movement that required a decision process (brightness discrimination) than for those movements that were simple reactions to a single light signal. It is interesting to note that even in the complex reaction-time situation, EMG onset in parkinsonian patients was found to be in the normal range, whereas the contact latencies were about twice as long on the average as compared to normal subjects. These patients may have activated their muscles before they decided where to go; thus, “hesitation” may explain the prolonged transport time of these movements. However, it must be emphasized that even the simple ballistic movements of parkinsonian patients were abnormal in their innerva-
MUSCLE & NERVE
SepUOct 1978
50
L5
60
55
I sec
65
70
L5
B
'i
1.2
1°i
7
1
I
50
55
60
65
70
50
55
60
65
1
4 I
-
-
45 sec
r
-
50
T
55
-
-
.-
-
60
65
I
70
L5
J
C
2.0
70
I
14
12 10
1
I
08 06 04
02
Years
0 2-
Years
Figure 3 Means and standard deviations of response latencies in normal subjects (left) and Parkinsonian subjects (right) The ages of the subjects are marked on the abscissa (A) EMG latencies for the biceps muscle (5) Movement latencies (C) Contact latencies The mean values of simple reaction times are plotted as filled circles, those of complex reaction times as open triangles (adapted from Schneiderg)
Central Motor Disorders
MUSCLE & NERVE
Sept/Oct 1978
409
Figure 4 . Perturbation experiment in a monkey. (A) The monkey is trained to hold the manipulandum in a narrow target zone (indicated to the monkey by a visual cue).. The cylinder of the manipulandum has to be squeezed with a prescribed amount of force (also signaled to the monkey by a visual cue). Perturbations (deflections of the manipulandum) are injected randomly by means of a torque motor stretching either the biceps or the triceps muscles. (B)Averaged signals of 15 trials with perturbations (load pulses of 30 Newton-cm), with ensuing deflection of the arm toward extension. Averaged and rectified biceps activity. Note the occurrence of two peaks in the biceps activity, M , and M2,which are evoked by stretch. The M 2 peak has an onset latency of about 35 msec and is considered to represent the response mediated via a transcortical loop.
B
I
4.5.
I I
MOTOR REACTIONS TO PASSIVE LIMB DISPLACEMENTS
Description of the Problem. Abnormal muscle tone is
a symptom that frequently accompanies central motor disorders. The therapeutic benefits of thalamotomy are reasonably well established, but newer surgical ap-
Central Motor Disorders
M2
I
tion pattern. This fact argues against the hypothesis of Kornhuber3 that the basal ganglia control slow movements while the cerebellum controls fast ballistic movements. A direct clinical application of this experimental paradigm was found to be useful for follow-up studies of parkinsonian patients undergoing stereotaxic surgeryl or medication with dopa and other drugs,2and for analysis of hereditary mirror movements.'
410
M1
-
50msec
proaches, such as dentatotomy or cerebellar and dorsalcolumn stimulation, need careful assessment. A basis for clinical testing of muscle tone may be provided by the animal experiments to be described; more detailed information may be found elsewhere. In brief, monkeys were trained to perform a postural task-i.e., squeezing a manipulandum with the hand using a prescribed amount of pressure, and at the same time keeping the handle positioned in a narrow zone. The experimental variable was derived from the application of random, sudden load pulses. Animal subjects were rewarded for rapid correction of the external perturbations. T h e object of these experiments was to study the mechanisms underlying internal controls capable of minimizing the effects of external perturbations. Muscle
MUSCLE & NERVE
SeptIOct 1978
tone is clinically evaluated by stretching a muscle group; this maneuver may be viewed as a perturbation, and the stiffness of the muscle is equivalent to the clinical term muscle tow, estimated by rapid stretching of the muscles. The question we may ask, therefore, is what neural mechanisms contribute to muscle stiffness. ElectromyographicResponses to Load Perturbations in Monkeys. Figure 4A illustrates the experimental par-
adigm, and figure 4B delineates the result of 15 perturbations of a torque pulse of 30 Newton-cm and of 70msec duration (onset marked by arrow) on the position signal, as well as the EMG activity of the biceps muscle. EMG activity was recorded using surface electrodes. The position signal (measured by a potentiometer) and the EMG activity was recorded using surface electrodes. The pling rate of 3 kHz, and were then rectified and averaged on-line by a minicomputer. Two peaks of stretchevoked activity emerged in the EMG. These responses were termed M, and M2 by Lee and T a t t ~ nLatency .~ of the early response was 15 msec and was thus compatible with the latency of a segmental stretch reflex; the second burst occurred at a latency of 35 msec. Several lines of evidence suggest that the late response, MB,is mediated via a long transcortical loop. l2 It is now well established that signals from muscle spindles are transmitted to the cerebral ~ o r t e x . ~ , ~ , ' ~ The late response, M2, has attracted considerable interest since Lee and Tatton4 showed that it may be greatly exaggerated in parkinsonian patients, and that it may thus contribute to the increased muscle stiffness of parkinsonian rigidity. It would be important to extend these observations to a larger number of patients, and to study various types of increased muscle tone. EXCITABILITY MEASUREMENTS OF SOLEUS MOTOR NEURONS USING THE HOFFMANN REFLEX I N MONKEYS
T o demonstrate more directly the participation of the motor cortex in the proposed long-loop reflex, we tested the effect of' reversible cooling of the motor cortex on the excitability of H reflexes in monkeys.13 Monkeys were trained to sit in a primate chair with one leg fixed in a fiberglass cast containing the stimulating nerve electrodes and the recording EMG electrodes. The monkeys were required to exert a light, steady pressure with the foot. When the rectified and integrated EMG activity of the soleus muscle was in a prescribed voltage window, the animal was reinforced. With this procedure, stable H reflexes evoked by stimulation of the popliteal nerve could be recorded with an H:M ratio larger than that which has been hitherto described for baboons.' A stimulus strength that evoked a maximal H wave without an M wave was used as the test response. A train of weak (i.e., just subthreshold for the H response) pulses also applied to the popliteal nerve, was used as a conditioning stimulus. With these parameters, secondary effects produced by contraction were avoided, and
Central Motor Disorders
the recovery curve of the test-reflex amplitude reflected the effects of group I volleys only. The conditioning stimulus facilitated or depressed the amplitude of the monosynaptic test reflex, depending on the interval between conditioning and test stimuli. Averaged excitability curves, obtained 10 times in each of the four monkeys tested, showed an early facilitation and a late facilitation that were superimposed on a long-lasting (approximately 1-sec) depression of the test response. The early and late facilitation are comparable in latency and duration to the MI and M, responses, respectively, obtained by sudden stretch. The longer latency of the late facilitation, as compared to the M2 response recorded from the upper limb, conforms with the assumption of a transcortical loop. This was further strengthened by the results of reversibly blocking the contralateral hindlimb area of the motor cortex; in all four monkeys, the late facilitation was reversibly abolished during cooling without any other changes in the time course of the excitability curve. In subsequent investigations (Chofflon, Riiegg, and Wiesendanger, to be published), it was shown that the pyramidal tract is the efferent limb of the transcortical loop. It is noteworthy that the late facilitation of the H-reflex recovery curve has been reported to be abnormally high in patients with increased muscle tone." I n previous studies on humans, however, a strong conditioning stimulus was usually given that complicated interpretation of late facilitation (e.g., by activation of higher threshold afferent fibers, or by mechanical changes bringing into play secondary actions of receptors such as Golgi tendon organs and muscle spindles). Since our experiments with the use of weak conditioning stimuli seem to indicate that the M2 response and late facilitation of the H-reflex recovery curve are an expression of the same transcortical mechanism, and since M2 responses and late facilitation have been reported to be abnormally high in parkinsonian patients, we suggest that these experimental procedures be used in such patients with the aim of quantifying muscle tone to assess the extent of improvement induced by therapy.
REFERENCES 1. Butz P, Kaufmann W, Wiesendanger M: Analyse einer raschen Willkurbewegung bei Parkinsonpatienten vor u n d nach stereotaktischem Eingriff am Thalamus. Z Neurol
198:105-119, 1970. 2 . Kaufmann W, Butz P, Wiesendanger M: Effekt einer komhinierten Behandlung von Parkinsonpatienten mit I-Dopa und einem Decarhoxylaseheminer (Ro 4-4602). Quantitative Analyse der Bradykinesie mittels Reaktionszeitmessungen. Dtsch 2 Nervenheilk 197:85-100, 1970. 3. Kornhuber HH: Motor functions of cerebellum and basal ganglia: the cerehello-cortical saccadic (ballistic) clock, the cerehellonuclear hold regulator, and the basal ganglia ramp (voluntary speed smooth movement) generator. Kybemetzk 8:157-162, 1971. 4. Lee RG, Tatton WG: Motor responses to sudden limb displacements in primates with specific CNS lesions and in
MUSCLE & NERVE
SepVOct 1978
411
human patients with motor system disorders. Can J Neurol Scz 2:285-293, 1975.
5. Lucier GE, Ruegg DG, Wiesendanger M: Responses of neurones in motor cortex and in area 3a to controlled stretches of forelimb muscles in Cebus m0nkeys.J Physiol (Lond) 251:833-853, 1975. 6. Phillips CG, Powell TPS, Wiesendanger M: Projection from low-threshold muscle afferents of hand and forearm to area 3a of baboon’s cortex.] Physiol (Lond) 217:419-446, 1971. 7. Regli F, Filippa G, Wiesendanger M: Hereditary mirror movements. Arch Neurol 16:620-623, 1967. 8. Roll JP, Bonnet M, Hugon M: Comparative study of proprioceptive reflexes in man and baboon (Papio pupio). In Goldsmith EI, Moor-Janowski J (Editors): Medical Primatology, Vol 2. Basel, Karger, 1972, pp 305-314. 9. Schneider P: Quantitative Analyse und Mechanismen der Bradykinesie bei Parkinsonpatienten. Dtsch Z Nemenheilk
412
Central Motor Disorders
194:80-102, 1968. 10. Wacholder K: Willkiirliche Haltung und Bewegung. Ergeb Physiol 26:568-775, 1928. 1 1 . Wiesendanger M: Pathophysiology of Mwcle Tune. Neurology Senes, Vol 9. Berlin, Springer Verlag, 1972, p 46. 12. Wiesendanger M: Comments on the problem of transcortical reflexes.] Physiol (Paris), in press. 13. Wiesendanger M, Riiegg DG, Lachat JM: The role of proprioceptive afferents in motor cortical output. Electruencephalogr Clin Neurophysiol (suppl), in press. 14. Wiesendanger M, Riiegg DG, Lucier GE: Why transcortical reflexes? Can J Neurol Scz 2:295-301, 1975. 15. Wiesendanger M, Schneider P, Villoz JP: Elektromyographische Analyse der raschen Willkiirbewegung.Schweiz Arch Neurol Neurochir Psychiatr 100:88-99, 1967. 16. Wiesendanger M, Schneider P, Villoz JP: Electromyographic analysis of a rapid volitional movement. Am J Phys Med 48:17-24, 1969.
MUSCLE 8 NERVE
Sept/Oct 1978
1:407-412
1978
ELECTROMYOGRAPHIC ASSESSMENT OF CENTRAL MOTOR DISORDERS MARIO WIESENDANGER, MD, and DIETER G. ROEGG, PhD
Electromyography is now a well-established tool in the recognition and evaluation of peripheral neuromuscular disorders. In assessing central deficits of movement and posture, however, it has been much less useful. This is true partly because voluntary, as well as reflex, movements and posture are complex and difficult to reproduce consistently. Moreover, a different experimental approach is required for each of the many different facets of central control-for example, muscle tone and stability of posture; timing of movement initiation; speed, force, or precision of movement performance; and strategy and economy of movement. Since management of central motor disorders has been enhanced by the advent of new therapeutic procedures, such as stereotaxic surgery, dopa medication, and dorsal-column and cerebellar stimulation, it is important to establish objective criteria for assessing therapeutic effects quantitatively. In the present paper, we wish to summarize two aspects of motor control: (1) the pattern of innervation and temporal structure of EMG activity in simple ballistic movements, and (2) reflex activity evoked by an external perturbation occurring during the performance of a motor task.
BALLISTIC MOVEMENTS I N NORMAL SUBJECTS AND IN PARKINSONIAN PATIENTS
From the lnstitut de Physiologie, Universit6 de Fribourg, Switzerland
The Pattern of Innervation. Figure 1 illustrates a typical
Acknowledgments Foundation
This study was supported by the Swiss Science
Address reprint requests to Dr Wiesendanger at the lnstitut de Physiologle, PBrolles, Universit6 de Fribourg, CH-1700 Fribourg, Switzerland 0148-639X/0105/0407 $00 OO/O a 1978 Houghton Mifflin Professional Publishers
Central Motor Disorders
In a series of experiments, we studied the slowing of voluntary movements of parkinsonian patients by measuring response latencies of visually triggered elbow movements. In each subject, we compared the response latencies of movements elicited by a single light signal (simple reaction time) with the response latencies of movements occurring after brightness discrimination of two light signals (complex or choice reaction time). The method has been fully d e ~ c r i b e d . ~ . ' ~ , ' ~ The parameters measured are best illustrated in a recording of the raw data (fig. 1). The visual "go" signal occurred at the beginning of the rectangular pulses of the third and fourth trace (also marked by small bars in the EMG recordings). Response latencies determined in the electromyogram are termed EMG latencies. The first rectangular pulse is terminated when the hand leaves the starting platform. This signal thus measures the movement latency. The second rectangular pulse terminates when the hand reaches the target key; it thus indicates the contact latency. The difference between contact latency and movement latency is the transport time. In the complex (choice) reaction, the fifth trace indicates whether or not the choice was correct.
trial of a nonnal subject. The triphasic pattern fits well with the classic description of Wacholder:lo the activity of the agonist (biceps) consists of a burst, often interrupted by a brief silent period. Activity of the antagonist (triceps) is usually weaker, occurs about 20 msec later, and reaches its maximum during the silent period of the
MUSCLE & NERVE
SepVOct 1978
407
Figure 1. Performance of a normal subject in a simple (A) and a choice (6) reaction-time situation. The visual trigger signal occurred at the onset of the rectangular pulses. a = biceps EMG, b = triceps EMG, c = movement onset latency, d = contact latency, e = correct response (choice reaction). Note the typical trrphasic EMG pattern in the antagonistic muscle pair.
Figure 2. Performance of a parkinsonian patient with moderate akinesia in a simple (A) and a choice (6)reaction-time situation. Same representation as in figure 1. Note prolonged EMG activity with cocontraction and action tremor.
-
Response Latencies. Figure 3 summarizes the results obtained in normal subjects (left) and in parkinsonian patients (right) of varying age classes. It is evident from this graph that reaction time, as measured by EMG, was not significantly different between normal subjects and patients for both simple and complex reactions. The most pronounced difference was noted for the contact latency in the complex reaction time situation (open triangles in fig. 3C). Thus, slowing of movement is more pronounced in parkinsonian patients if movement depends each time on a new decision to move to the right or left key, positioned under the two light signals. On the average, movement latencies were only slightly retarded. Since EMG response latencies of parkinsonians were in the normal range, the delayed onset of movement was
Central Motor Disorders
mms
e
I t
L
C
I
d
looms
biceps. Movement is terminated by a mechanical stop (the target key); EMG activity usually outlasts the movement by 100-200 msec (sometimes in repetitive small bursts). In parkinsonian patients the most common finding, even in early stages of the disease, was the less synchronized activity of both agonists and antagonists, which is in obvious contrast to the reciprocal, triphasic pattern of normal subjects (fig. 2). A burst of activity occurs and corresponds to action tremor occurring synchronously in the agonist and the antagonist. EMG activity outlasts the movement (halted by the mechanical stop) by several hundred milliseconds.
408
-
e
I
an outcome of the temporally dispersed EMG pattern, which resulted in a slow buildup of force. In summary, EMG analysis of a rapid volitional movement, performed in a simple and in a complex reaction-time situation, revealed the following abnormalities in parkinsonian patients:
1. Prolonged-i.e., temporally dispersed-EMG activity was observed in both agonist and antagonist muscles, and was often characterized by coactivation and action tremor. Movement initiation was delayed in some patients for lack of a rapid force buildup, rather than because of a delay in onset of the first voluntary EMG response. 2. The slowing was more pronounced for movement that required a decision process (brightness discrimination) than for those movements that were simple reactions to a single light signal. It is interesting to note that even in the complex reaction-time situation, EMG onset in parkinsonian patients was found to be in the normal range, whereas the contact latencies were about twice as long on the average as compared to normal subjects. These patients may have activated their muscles before they decided where to go; thus, “hesitation” may explain the prolonged transport time of these movements. However, it must be emphasized that even the simple ballistic movements of parkinsonian patients were abnormal in their innerva-
MUSCLE & NERVE
SepUOct 1978
50
L5
60
55
I sec
65
70
L5
B
'i
1.2
1°i
7
1
I
50
55
60
65
70
50
55
60
65
1
4 I
-
-
45 sec
r
-
50
T
55
-
-
.-
-
60
65
I
70
L5
J
C
2.0
70
I
14
12 10
1
I
08 06 04
02
Years
0 2-
Years
Figure 3 Means and standard deviations of response latencies in normal subjects (left) and Parkinsonian subjects (right) The ages of the subjects are marked on the abscissa (A) EMG latencies for the biceps muscle (5) Movement latencies (C) Contact latencies The mean values of simple reaction times are plotted as filled circles, those of complex reaction times as open triangles (adapted from Schneiderg)
Central Motor Disorders
MUSCLE & NERVE
Sept/Oct 1978
409
Figure 4 . Perturbation experiment in a monkey. (A) The monkey is trained to hold the manipulandum in a narrow target zone (indicated to the monkey by a visual cue).. The cylinder of the manipulandum has to be squeezed with a prescribed amount of force (also signaled to the monkey by a visual cue). Perturbations (deflections of the manipulandum) are injected randomly by means of a torque motor stretching either the biceps or the triceps muscles. (B)Averaged signals of 15 trials with perturbations (load pulses of 30 Newton-cm), with ensuing deflection of the arm toward extension. Averaged and rectified biceps activity. Note the occurrence of two peaks in the biceps activity, M , and M2,which are evoked by stretch. The M 2 peak has an onset latency of about 35 msec and is considered to represent the response mediated via a transcortical loop.
B
I
4.5.
I I
MOTOR REACTIONS TO PASSIVE LIMB DISPLACEMENTS
Description of the Problem. Abnormal muscle tone is
a symptom that frequently accompanies central motor disorders. The therapeutic benefits of thalamotomy are reasonably well established, but newer surgical ap-
Central Motor Disorders
M2
I
tion pattern. This fact argues against the hypothesis of Kornhuber3 that the basal ganglia control slow movements while the cerebellum controls fast ballistic movements. A direct clinical application of this experimental paradigm was found to be useful for follow-up studies of parkinsonian patients undergoing stereotaxic surgeryl or medication with dopa and other drugs,2and for analysis of hereditary mirror movements.'
410
M1
-
50msec
proaches, such as dentatotomy or cerebellar and dorsalcolumn stimulation, need careful assessment. A basis for clinical testing of muscle tone may be provided by the animal experiments to be described; more detailed information may be found elsewhere. In brief, monkeys were trained to perform a postural task-i.e., squeezing a manipulandum with the hand using a prescribed amount of pressure, and at the same time keeping the handle positioned in a narrow zone. The experimental variable was derived from the application of random, sudden load pulses. Animal subjects were rewarded for rapid correction of the external perturbations. T h e object of these experiments was to study the mechanisms underlying internal controls capable of minimizing the effects of external perturbations. Muscle
MUSCLE & NERVE
SeptIOct 1978
tone is clinically evaluated by stretching a muscle group; this maneuver may be viewed as a perturbation, and the stiffness of the muscle is equivalent to the clinical term muscle tow, estimated by rapid stretching of the muscles. The question we may ask, therefore, is what neural mechanisms contribute to muscle stiffness. ElectromyographicResponses to Load Perturbations in Monkeys. Figure 4A illustrates the experimental par-
adigm, and figure 4B delineates the result of 15 perturbations of a torque pulse of 30 Newton-cm and of 70msec duration (onset marked by arrow) on the position signal, as well as the EMG activity of the biceps muscle. EMG activity was recorded using surface electrodes. The position signal (measured by a potentiometer) and the EMG activity was recorded using surface electrodes. The pling rate of 3 kHz, and were then rectified and averaged on-line by a minicomputer. Two peaks of stretchevoked activity emerged in the EMG. These responses were termed M, and M2 by Lee and T a t t ~ nLatency .~ of the early response was 15 msec and was thus compatible with the latency of a segmental stretch reflex; the second burst occurred at a latency of 35 msec. Several lines of evidence suggest that the late response, MB,is mediated via a long transcortical loop. l2 It is now well established that signals from muscle spindles are transmitted to the cerebral ~ o r t e x . ~ , ~ , ' ~ The late response, M2, has attracted considerable interest since Lee and Tatton4 showed that it may be greatly exaggerated in parkinsonian patients, and that it may thus contribute to the increased muscle stiffness of parkinsonian rigidity. It would be important to extend these observations to a larger number of patients, and to study various types of increased muscle tone. EXCITABILITY MEASUREMENTS OF SOLEUS MOTOR NEURONS USING THE HOFFMANN REFLEX I N MONKEYS
T o demonstrate more directly the participation of the motor cortex in the proposed long-loop reflex, we tested the effect of' reversible cooling of the motor cortex on the excitability of H reflexes in monkeys.13 Monkeys were trained to sit in a primate chair with one leg fixed in a fiberglass cast containing the stimulating nerve electrodes and the recording EMG electrodes. The monkeys were required to exert a light, steady pressure with the foot. When the rectified and integrated EMG activity of the soleus muscle was in a prescribed voltage window, the animal was reinforced. With this procedure, stable H reflexes evoked by stimulation of the popliteal nerve could be recorded with an H:M ratio larger than that which has been hitherto described for baboons.' A stimulus strength that evoked a maximal H wave without an M wave was used as the test response. A train of weak (i.e., just subthreshold for the H response) pulses also applied to the popliteal nerve, was used as a conditioning stimulus. With these parameters, secondary effects produced by contraction were avoided, and
Central Motor Disorders
the recovery curve of the test-reflex amplitude reflected the effects of group I volleys only. The conditioning stimulus facilitated or depressed the amplitude of the monosynaptic test reflex, depending on the interval between conditioning and test stimuli. Averaged excitability curves, obtained 10 times in each of the four monkeys tested, showed an early facilitation and a late facilitation that were superimposed on a long-lasting (approximately 1-sec) depression of the test response. The early and late facilitation are comparable in latency and duration to the MI and M, responses, respectively, obtained by sudden stretch. The longer latency of the late facilitation, as compared to the M2 response recorded from the upper limb, conforms with the assumption of a transcortical loop. This was further strengthened by the results of reversibly blocking the contralateral hindlimb area of the motor cortex; in all four monkeys, the late facilitation was reversibly abolished during cooling without any other changes in the time course of the excitability curve. In subsequent investigations (Chofflon, Riiegg, and Wiesendanger, to be published), it was shown that the pyramidal tract is the efferent limb of the transcortical loop. It is noteworthy that the late facilitation of the H-reflex recovery curve has been reported to be abnormally high in patients with increased muscle tone." I n previous studies on humans, however, a strong conditioning stimulus was usually given that complicated interpretation of late facilitation (e.g., by activation of higher threshold afferent fibers, or by mechanical changes bringing into play secondary actions of receptors such as Golgi tendon organs and muscle spindles). Since our experiments with the use of weak conditioning stimuli seem to indicate that the M2 response and late facilitation of the H-reflex recovery curve are an expression of the same transcortical mechanism, and since M2 responses and late facilitation have been reported to be abnormally high in parkinsonian patients, we suggest that these experimental procedures be used in such patients with the aim of quantifying muscle tone to assess the extent of improvement induced by therapy.
REFERENCES 1. Butz P, Kaufmann W, Wiesendanger M: Analyse einer raschen Willkurbewegung bei Parkinsonpatienten vor u n d nach stereotaktischem Eingriff am Thalamus. Z Neurol
198:105-119, 1970. 2 . Kaufmann W, Butz P, Wiesendanger M: Effekt einer komhinierten Behandlung von Parkinsonpatienten mit I-Dopa und einem Decarhoxylaseheminer (Ro 4-4602). Quantitative Analyse der Bradykinesie mittels Reaktionszeitmessungen. Dtsch 2 Nervenheilk 197:85-100, 1970. 3. Kornhuber HH: Motor functions of cerebellum and basal ganglia: the cerehello-cortical saccadic (ballistic) clock, the cerehellonuclear hold regulator, and the basal ganglia ramp (voluntary speed smooth movement) generator. Kybemetzk 8:157-162, 1971. 4. Lee RG, Tatton WG: Motor responses to sudden limb displacements in primates with specific CNS lesions and in
MUSCLE & NERVE
SepVOct 1978
411
human patients with motor system disorders. Can J Neurol Scz 2:285-293, 1975.
5. Lucier GE, Ruegg DG, Wiesendanger M: Responses of neurones in motor cortex and in area 3a to controlled stretches of forelimb muscles in Cebus m0nkeys.J Physiol (Lond) 251:833-853, 1975. 6. Phillips CG, Powell TPS, Wiesendanger M: Projection from low-threshold muscle afferents of hand and forearm to area 3a of baboon’s cortex.] Physiol (Lond) 217:419-446, 1971. 7. Regli F, Filippa G, Wiesendanger M: Hereditary mirror movements. Arch Neurol 16:620-623, 1967. 8. Roll JP, Bonnet M, Hugon M: Comparative study of proprioceptive reflexes in man and baboon (Papio pupio). In Goldsmith EI, Moor-Janowski J (Editors): Medical Primatology, Vol 2. Basel, Karger, 1972, pp 305-314. 9. Schneider P: Quantitative Analyse und Mechanismen der Bradykinesie bei Parkinsonpatienten. Dtsch Z Nemenheilk
412
Central Motor Disorders
194:80-102, 1968. 10. Wacholder K: Willkiirliche Haltung und Bewegung. Ergeb Physiol 26:568-775, 1928. 1 1 . Wiesendanger M: Pathophysiology of Mwcle Tune. Neurology Senes, Vol 9. Berlin, Springer Verlag, 1972, p 46. 12. Wiesendanger M: Comments on the problem of transcortical reflexes.] Physiol (Paris), in press. 13. Wiesendanger M, Riiegg DG, Lachat JM: The role of proprioceptive afferents in motor cortical output. Electruencephalogr Clin Neurophysiol (suppl), in press. 14. Wiesendanger M, Riiegg DG, Lucier GE: Why transcortical reflexes? Can J Neurol Scz 2:295-301, 1975. 15. Wiesendanger M, Schneider P, Villoz JP: Elektromyographische Analyse der raschen Willkiirbewegung.Schweiz Arch Neurol Neurochir Psychiatr 100:88-99, 1967. 16. Wiesendanger M, Schneider P, Villoz JP: Electromyographic analysis of a rapid volitional movement. Am J Phys Med 48:17-24, 1969.
MUSCLE 8 NERVE
Sept/Oct 1978
![[Electrooculographic-electromyographic examination of peripheral and central motor disorders of the eye].](/assets/img/hold.png)
