- Email: [email protected]
Late muscular responses to transcranial cortical stimulation in man
Electroencephalography and clinical Neurophysiology , 1990, 75:161-172
161
Elsevier Scientific Publishers Ireland, Ltd. EEG02145
Late muscular responses to transcranial cortical stimulation in man Helen Holmgren, Lars-Erik Larsson and Stephen Pedersen Dept. of Clinical Neurophysiology, University of Linki~ping, LinkiJping (Sweden) (Accepted for publication: 26 July 1989)
Summary
Transcranial cortical stimulation was performed in healthy human subjects. Electrical and magnetic stimuli were applied over the motor areas of the hand and the leg. Surface EMGs were recorded from the antagonistic extensor and flexor carpi radialis and the anterior tibial and triceps surae muscles. Except for the well known early (primary) responses several secondary responses have been recorded. (1) A response with a latency of approximately 100 msec and with higher threshold than the primary response. It was found in approximately half of the muscles recorded from except for the anterior tibial where it was comparatively rare. It could appear both contra- and ipsilaterally and increased in amplitude with preactivation of the target muscle. It is probably not of startle origin but its appearance may be influenced by an approximately simultaneous startle effect from the scalp stimulus. (2) In some subjects a response with a latency of 50-60 msec was recorded. Sometimes it was easily separated from the primary response and from S 100 but sometimes the differentiation was difficult. (3) It was found that the known inhibitory period which follows the primary response was contralateral and increased in duration with increasing stimulus strength. The increase in duration may be due to inhibitory feedback from receptors activated by the simultaneously increasing primary muscle twitch. (4) In relaxed muscle we also observed a contralateral response with a latency exceeding 150 msec. The latency increased with increasing stimulus strength. This increase seems to be related to discontinuance of the inhibitory period. Key words: Cortical stimulation; Man; Compound muscular responses
Ever since the classical experiment of Fritsch and Hitzig (1870), Ferrier (1873) and Leyton and Sherrington (1917), electrical stimulation of the exposed cortex has provided a wealth of information about cerebral function both in animals and in man (Penfield and Boldrey 1937; Philips and Porter 1977). Transcranial cortical electrical stimulation in man (TCCS) has also been introduced (Merton and Morton 1980a,b; Merton et al. 1982; Marsden et al. 1982, 1983; Levy et al. 1984; Hassan et al.
The financial assistance of The Vivian L. Smith Foundation, Houston, TX, is gratefully acknowledged.
Correspondence to: Dr. Helen Holmgren, Dept. of Clinical Neurophysiology, University Hospital, S-581 85 LinkiSping (Sweden).
1985; Rossini et al. 1985). In the papers hitherto published, the greatest interest has been directed towards an early muscular discharge which has been considered to represent a response to excitation of fast cortico-spinal fibres. Following this early response there is an inhibitory pause of long duration (Marsden et al. 1983). Normal values for latencies to different muscles after electrical stimulation, which take stimulus strength and body length into consideration, have been published (Hacke et al. 1987; Meyer et al. 1987). In addition, transcranial magnetic stimulation, which gives less discomfort, has been introduced (Hess et al. 1986, 1987; Barker et al. 1987). It has already been shown that TCCS may be a useful tool for clinical neurophysiological diagnosis (Young and Cracco 1985). Reports of prolonged latencies, which sometimes may be due to
0013-4649/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland, Ltd.
162 prolonged central conduction time, have been published for multiple sclerosis (Cowan et al. 1984; Mills and Murray 1985; Berardelli et al. 1988; Ingram et al. 1988), in motor neurone disease (Berardelli et al. 1987; Hugon et al. 1987; Ingram and Swash 1987) and in spinal cord disease (Snooks and Swash 1985; Dimitrijevi6 et al. 1988). It is obvious that a prolonged latency following TCCS is not confined to any specific condition or pathology (Thompson et al. 1987). In repeated experiments with TCCS under different conditions we have found at least 2 additional responses with comparatively long latencies. A preliminary report of our findings has been presented to the Swedish Society for Clinical Neurophysiology (Holmgren et al. 1988). Similar discharges were independently described in a paper by Calancie et al. (1987) for the extensor carpi radialis, and for triceps surae by Dimitrijevi6 et al. (1988). The aim of the present study was to describe and analyse the late E M G responses to electrical and magnetic TCCS in the forearm and lower leg at varying stimulus strengths.
Methods Forty experiments have been performed on ourselves and on healthy medical students (10 women and 19 men) who, after being fully informed, consented to participate. Approval for the experiments was obtained from the Ethical Board of Linkrping University. The subject was comfortably seated in a half reclining position with the lower legs supported and with the arms either on the elbow rests of the investigation chair or in the lap. The hands were pronated. C a r e was taken to make the position comfortable and the subject was told to relax except when otherwise instructed. The degree of relaxation in the target muscles was checked by the experimenters on a 4-channel oscilloscope. When preactivation of the target muscles in the hands was needed the subjects pinched small springloaded devices between thumb and forefingers. When used bilaterally they gave approximately the same resistance to the pinch. For
H. HOLMGREN ET AL. preactivation of the leg muscles the experimenter administered a slight degree of resistance by hand. In 20 experiments we used a Digitimer electrical stimulator (D180), which gives high voltage condensor shocks (max 750 V) with a short time constant (50-100/~sec). The stimulator strength is given in percent of maximum. Following preliminary trials with different electrode positions, including the ring electrode described by Rossini et al. (1985), we decided to administer the stimulus through Beckman electrodes (diameter 7 mm) fastened to the scalp with collodion. The anode was Cz, C3 or C4 (Jasper 1958). Originally we placed the cathode on Fz, F3 and F4, respectively. This gives an interelectrode distance of approximately 8 cm. In later experiments we moved the F electrode in the frontal direction, thereby increasing the interelectrode distance to 10-12 cm. According to measurements made in two cases, an increase of interelectrode distance lowered the stimulus threshold but, on the other hand, resulted in a slight loss of somatotopic specificity. The time interval between the stimuli varied but was around 5 min. With a stimulus duration of 100 /~sec the threshold for a peripheral motor response was generally found to be 20-40% of the maximum strength available. In 20 additional experiments we used a Cadwell magnetic stimulator (MES-10) with a coil measuring 9 cm. For strict contralateral stimulation of the hand area the upper circumference of the coil was placed approximately 5 cm lateral to the midline. For bilateral stimulation we usually placed the centre of the coil over the vertex. The stimulation gave a twitch of the scalp which, for the electrical stimulus, was found unpleasant by some subjects. There was considerable habituation after repeated stimuli. The magnetic stimulation was not considered to be unpleasant, neither were the peripheral motor responses. Recording in most cases was done from the two antagonistic muscle pairs, extensor and flexor carpi radialis and anterior tibial and triceps surae, respectively. Although several other muscles, both proximal and distal, were investigated, the results will not be presented in this context. Skin electrodes were taped over the thickest part of the muscle belly with an interelectrode distance of 2
TCCS: LATE M U S C U L A R RESPONSES
cm. Conventional AC amplifiers were used and were connected to a Nicolet 1170 averager and to an F M tape recorder for replay when necessary. The bandwidth of the recording system was from 1 to 1000 Hz ( - 6 dB at the ends).
163 TABLE I Latencies and standard deviations (range for S >150) for muscular responses in relaxed target muscles following electrical TCCS. In the fraction values the numerator gives the n u m b e r of muscles in which the response was recorded and the denominator the n u m b e r of muscles investigated.
Results Ext. carpi
Several different responses have been recorded, some of which have already been described by other authors. They are illustrated in Fig. 1. Latencies at submaximal electrical stimulation, suggested designations and frequencies of occurrence are presented in Table I. Except for latencies of the primary responses, which may be shorter for electrical stimulation, we have not found obvious differences between the results from electrical and magnetic stimulation. (1) Primary responses were obtained in all subjects; they were contralateral to the stimulus, increased in amplitude when the target muscle was preactivated and had shorter latency and longer duration with increasing stimulus strength. In the forearm they tended to be larger in the extensor compared to the flexor carpi; in the lower leg they
Prim (msec)
S 50 (msec)
S 100 (msec)
S > 150 (msec)
16.1 + 2.5
57.3 + 8.8 11/11 50.8___5.9 11/11
85.7 + 6.5 2/11 89.6 + 7.6 3/11 117.6+9.8 8/8 109.8-t-6.8 8/8
113-225 8/11 175-200 8/11 130-225 3/8 175 6/8
9/11 Flex. carpi Tib. ant. Tric. sur.
16.5 + 2.0 4/11 30.8+2.2 2/8 33.4+2.8 1/8
were clearly larger in the anterior tibial muscle compared to the triceps surae. Most of these characteristics have been described by other authors. They will not be further discussed. Several other responses were observed; they did not appear after each single stimulus but their appearances were so constant and the latencies so consistent within the given range, that their existence cannot be questioned.
$5o
S >
150 M. ext carpi Fad.
M. f l e x o r carpi rad.
0.1 mV I
50 ms Fig. 1. Single muscular responses to electrical TCCS in a relaxed subject. The stimulus was applied over the scalp at C3-F3 with C3 positive. The records were made from the extensor carpi (upper trace) and the flexor carpi radialis (lower trace) muscles.
H. H O L M G R E N ET AL.
164
I
A
M. t r t c e p s
S 100 could appear both contralaterally and ipsilaterally to the stimulus. In 7 subjects a bilateral investigation of the arm muscles was made using magnetic stimulation. In 6 of them an S 100 response could be observed, in 1 case, however, only after preactivation of the muscles. In these
surae
0. 5 mV
A ~_
Relaxed, M. flexor carpi rod. sin
M. ext. carpi rod. sin
M flexor carpi rad. dx
I lmV .50 m s
M. ext. carpi rad. dx Preactivation
Fig. 2. Appearance of S 100 in the anterior tibial muscle as a result of preactivation. Electrical stimulus at Cz-Fz with Cz positive. A: subject relaxed. B: slight preactivation. Note the different amplification in the two records. .
(2) S 50. A small response with a latency of 50-60 msec recorded in the forearm muscles in 4 subjects (Fig. 1). It was very close to the last part of the primary response but was clearly separated from it as well as from S 100 when both occurred. (3) S 100. A response with a latency of 80-110 msec recorded in slightly more than half of the investigated extensor and flexor carpi radialis and triceps surae muscles. In the relaxed anterior tibial muscle it was found only in a few subjects. Its threshold was usually slightly higher than that of the primary response. This response probably corresponds to the E2 response observed by Calancie et al. (1987) in the extensor carpi and by Dimitrijevi6 et al. (1988) in the triceps surae muscle. Although S 100 was definitely found in relaxed muscles, it clearly increased with preactivation and then was also often found in the anterior tibial muscle (Fig. 2). The average latency of S 100 was significantly longer in the legs than in the arms ( P < 0.0001, t test). The average difference was 22 msec, which is 5 msec more than the average difference between the primary responses recorded from the same muscles.
Relaxed
|
/ 10 mV |
50 ms Fig. 3. Single muscular responses to magnetic TCCS (90%). The upper circumference of the coil was applied over the left hand area. The records were made bilaterally from the flexor and extensor carpi radialis muscles as indicated. In A and C the subject was relaxed, in B he pinched a small springloaded device on both sides.
FCCS: LATE M U S C U L A R RESPONSES
165
.__
70 */,
9 0 */o
M. e x t . carpu r'ad. dx
-
n
~tlmm
... _.
-
. i
L
M. f l e x o r c a r p i tad. dx
M. e x t . carpi tad. sin T
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
~""
_'_L
-
" -
mL__
y
_Ln
-jL--'-_
- _-~--_
_--~.
~ r C
----
M, f l e x o r carpi rad. sin
..__
100 °Io
i
NNI I
J
I ----
T
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
t
. . . . . . . . . . .
j
I
J
I
n
• . . . . .
0 . 5 mV !
50 ms Fig. 4. Single muscular responses to magnetic TCCS at the stimulus strength indicated. The upper circumference of the coil was applied over the left hand area. The records were made bilaterally from the extensor carpi and flexor carpi radialis muscles. A secondary response appears first in the contralateral extensor muscle and then bilaterally with higher stimulus strength.
cases S 100 appeared both ipsi- and contralaterally, partly independent of each other. In 2 cases there was only an ipsilateral response. An illustration of the variability is given in Fig. 3. In Fig. 3A, B and C S 100 is clearly contralateral in the extensor carpi muscle. In Fig. 3C there is in addition a slight ipsilateral response. It is not seen in Fig. 3B in spite of the fact that the muscle is slightly preactivated. In 1 case we had difficulty in deciding whether the secondary response observed was a late S 50 or an early S 100 (Fig. 4). In a few individuals we sometimes found a
generalized muscle contraction at the beginning of the experimental session, even at very low stimulus strength. It disappeared on repeated stimulation, even if the stimulus strength was increased (Fig. 5). These responses had other characteristics than the S 100, which did not habituate to repeated stimulation and which demanded higher stimulation intensity to appear. We consider them to be startle reactions due to the scalp stimulus evoked in subjects who were a bit uncertain at the beginning of the experiment. Even if these possible startle reactions are sub-
H. HOLMGREN ET AL.
166 4
I
~
~
~
%
_
M. extensor carpi rad. I M. flexor carpi
I
rad.
M. abd. dig. V
25"/o
. . . . . .
14~,~ta/"-
!
20 */o
.
_
. . . . .
j
45 */, 0.1 mV I
100 ms Fig. 5. Possible startle reaction. Electrical TCCS. Records from the extensor and flexor carpi radialis muscles and from the abductor dig. V muscle. The figures show the results from the 1st, 2nd, 4th and 14th stimuli. Note that the response disappeared at repeated stimulation even though the stimulus strength was increased. Note also the different latency and configuration of the secondary response at 45% stimulation strength.
liminal in later parts of an experimental session, they might still have an influence on the responses recorded. (4) An inhibitory response. It was contralateral to the stimulus (Fig. 6) and so far seems to be related to the p r i m a r y response. We were, however, able to confirm that inhibitory periods could appear also at very low stimulation strength without: a discernible p r i m a r y response (Fig. 7; M a r s d e n et al. 1983; Calancie et al. 1987). In 3 subjects we investigated the relation between the duration of inhibition and stimulus strength. In all cases the length of the inhibitory period increased with increased intensity of the stimulus (Figs. 8 and 10B). In some cases the inhibition was divided into several parts, with interposed excitatory periods (Figs. 2 and 7).
(5) S > 150. A response with a latency usually longer than 150 msec and which increased with increasing stimulus strength (Fig. 9). It was definitely found in relaxed muscles contralateral to the stimulus. Fig. 10A shows examples of regressions between latencies and stimulus strength in 2 subjects. This response had a strong resemblance to the r e b o u n d effect f o u n d after inhibition of a previously contracted muscle. The latency increase with increasing stimulation for S > 150 is thus similar to the increasing duration of inhibition during the same conditions (Fig. 10A and B). Discussion It is clear that the stimuli given result in a complex sequence of excitability changes in the
TCCS: LATE M U S C U L A R RESPONSES
~
d
167
x rad. dx
J
~
__ 7 ext'carp' rad. s'n
i~ flexor carpi rod sin i . . . . . . . . . . . .
D
•Jl mV 50 ms Fig. 6. Single muscular responses to magnetic TCCS (100%). The upper circumference of the coil was applied over the left hand area.
The records were made bilaterally from the extensor and flexor carpi radialis muscles preactivated by bilateral pinching. From A to D the subject increased the degree of contraction. Note that the inhibitory response is strictly contralateral and that the duration tends to decrease with increasing contraction.
motoneurones for a long period after the primary response. We have not observed obvious differences in the late responses depending on whether electrical or magnetic stimulation has been used. We have found that the inhibitory process previously described is contralateral to the stimulus and that its duration is dependent on the stimulus strength. We have also found several excitatory episodes causing contra- and ipsilateral electromyographic responses. Even though these responses do not appear after each stimulus in a given subject and at a given stimulus strength, there can be no doubt of their existence. Possible
explanations of their somewhat capricious appearance may be a competition with the simultaneous inhibitory process and partly also variations in the subjects expectations before each single stimulus. A few words about terminology. The responses that we call primary, S 100 and S > 150 have been described by Calancie et al. (1987) as E 1, E 2 and E 3. This terminology was probably inspired by their method of recording the discharge frequency of single motor units. In the present paper we have also observed another possible response, S 50. Its latency will place it between E 1 and E:.
168
H. H O L M G R E N ET AL.
responses. Finally, even if overt startle effects due to the scalp stimulation seem comparatively easy to exclude (Fig. 5), there is still a possibility that
°Io
20 ~¢~--
. . . . . . .
Relaxed
.....
0.2 mV I lOO
ms
Fig. 7. Single muscular responses to electrical TCCS applied at Cz-Fz with Cz positive. Records from the preactivated anterior tibial muscle. In A the stimulus was supraliminal, in B and C there was no discernible primary response.
Instead of calling it Ea.5 or something similar we prefer to designate the different responses with approximate latencies. It is obviously difficult to draw conclusions regarding the origin of the late responses. The electrical field of the stimulus must cover a large area, which includes not only motor but also premotor and to some extent also parietal cortex, and the stimulus may influence spinal motoneurones not only via direct cortico-spinal connections but also via intermediary segmental spinal neurones, which by themselves may be sources of both facilitatory and inhibitory effects (Baldissera et al. 1981). It is also quite possible that there are influences mediated via the spinal projections from different nuclei in the brain-stem, which themselves receive cortical projections (Lundberg 1966). In addition, the distinct twitch which is connected to ihe primary response may give rise to secondary reflex
45
~%"~"~
Relaxed 50
ms
Fig. 8. Inhibitory responses as related to increasing stimulus strength. Electrical TCCS at C3-F3 with C3 positive. The percentage stimulus strength is given to the right. Records from the right extensor carpi radialis muscle. The upper and the lowermost records show responses when the subject was relaxed. In all other records the muscle was slightly preactivated.
169
TCCS: LATE M U S C U L A R RESPONSES
the scalp stimulus produces subliminal effects in the central tone producing systems. These alternatives cannot be thoroughly discussed without further experimental analysis. A few comments may, however, be justified on the basis of some of the characteristics described. The overt inhibitory process is very constant, it is clearly contralateral to the stimulus, its duration increases with increasing stimulus strength and there is some evidence that it is also deeper. It starts immediately after the primary response and therefore has a latency related to its duration. Since the primary response is also contralateral and also increases with increasing stimulus strength, the inhibitory process may be dependent on the existence of the primary twitch. The connection between the two could, for instance, be Golgi inhibition from the twitching muscle. The inhibition should in such a case increase with increasing stimulus strength, which gives an increasing twitch. The fact that it sometimes may be obtained without any discernible primary response (Fig. 7) is not necessarily contradictory, since it is difficult to exclude a minimal primary response hidden in the voluntary activity, which in the
Latency to S > 1 5 0
A
aLE o
220
E
v
1BO
140
100 20
25
30
35
M
Stimulus strength
ext. corp. rod.
M. ~lex. carp rod
30*/*
40 *lo
Contralaterul preactivotion
,
100 ms
1 o,1 mv I
Fig. 9. Latency increase of S > 1 5 0 (arrow) with increasing stimulus strength. Responses from right extensor carpi radialis (upper trace) and flexor carpi radialis (lower trace). Electrical TCCS at C3-F3 with C3 pos. Stimulus strength 20% in A and D; 30% in B and E; 40% in C and F. In A, B and C the subject was relaxed, in D, E and F he contracted muscles in his left forearm.
40
45
50
Stimulus (*/.)
L a t e n c y to enc} of inhibition
m a g n stim
O AM
• PB
190
o
170
E
0
e 0 ~
0
0
70
"
v
0
>~ 150
o, 130
o
0
•
50
0
"
dO"
Stimulus D~
electrical stim
4G
80
"
gO
"
I00
(%)
Fig. 10. Examples of regressions for the relation between stimulus strength and the latency of S > 1 5 0 (A) and the latency to the end of inhibition (B). In A the stimulus was electrical, in B magnetic. The target muscle was extensor carpi radialis in all cases.
present experiments has been the background necessary to demonstrate the inhibition. Neither can influences from minimal twitches in adjacent muscles, which were not recorded from, be completely excluded. It is known that inhibitory action may also occur between muscles not belonging to the same myotatic unit (Granit 1950). Calancie et al. (1987), who recorded from single motor units, noted a decrease in repetition frequency as an expression of the inhibitory period. With weak stimuli this decrease could be observed also without previous excitation as evidence of a contraction. The S > 150, which is also contralateral, seems to be a counterpart of the inhibitory response,
170
appearing at the end of inhibition, thus explaining why its latency increases with increasing stimulus strength. To avoid misunderstanding it should be stressed that even if a similar discharge is seen at the end of inhibition when the target muscle has been preactivated our present description builds upon observations in relaxed muscles. Calancie et al. (1987) call it a rebound effect, which may be a very appropriate designation. They suggest that it is due to the proprioceptive inflow from the contracting target muscle. This is an obvious possibility. The excitatory effect that seems to be released at the end of inhibition may, however, also be due to long-lasting discharges from tone-producing structures in the brain-stem, themselves activated by the cortical stimulation. In fact, it seems that the cortical stimulus gives rise to long-lasting excitatory and inhibitory processes competing for dominance in the engaged muscles and that this competition ends with a rebound phenomenon. The mechanisms behind S 100 may be complex. The average latency difference for S 100 between the arms and the legs was approximately 22 msec. This should be compared to the average latency difference of 17 msec for the primary responses in the same muscles. The difference of 5 msec is highly significant ( P < 0.0001, t test). Assuming that S 100 uses the same peripheral pathway as the primary response this would speak for a slower central conduction for fibres evoking S 100 than for those giving the primary response. The assumption of the same peripheral pathway for the two responses is to some extent supported by the illustrations provided by Calancie et al. (1987), showing that the same motor unit takes part in both the primary and the late response. Although the average latencies in arms and legs were calculated on different subjects, a hypothesis of a possibly slower central pathway is still not sufficiently founded. The latency of S 100 suggests a possible startle origin due to the somewhat unpleasant electrical scalp stimulus. However, the probable startle that we observed in some subjects at the beginning of an experiment had a somewhat longer latency and habituated rapidly, which is not typical for S 100. In addition, the response is readily obtainable also with magnetic stimulation, which gives much less
H. H O L M G R E N ET AL.
discomfort. The S 100 response sometimes appears bilaterally, which may speak for startle, but the appearance is asymmetrical in an unpredictable way. Sometimes the contralateral response is better developed, sometimes the ipsilateral. A probable explanation of ipsilateral dominance, when it occurs, is that a contralateral response is suppressed by the simultaneous contralateral inhibition. Such a mechanism may explain why a slight preactivation of the target muscle can result in the appearance of a previously unobserved S 100 discharge (Fig. 2). Nothing in the previous discussion explains why S 100 is sometimes contralateral or dominates contralaterally. Supposing that collaterals from the cortico-spinal tract activate the reticular formation one would expect a bilateral response. A similar activation of the rubro-spinal system would, on the other hand, further a contralateral response. Finally, backfiring receptors activated by the primary twitches both in agonists and antagonists might contribute to the response both in facilitatory and inhibitory ways. The latency of the response is sufficiently long to give all those afferents activated by the primary twitch (Ia, Ib and FRA) the possibility of a renewed influence at the segmental level. Can it be that the late responses described in this paper are some kind of epiphenomena, without real significance, or is it possible that they are related to feedback mechanisms, which facilitate or inhibit the effect of the cortico-spinal signal (Lundberg 1979)? The fact that these late responses are comparatively weak in relation to the primary responses may contradict this suggestion, but may have its explanation in the very unphysiological way in which the cortical stimuli are given.
References Baldissera, F., Hultborn, H. and lllert, M. Integration in spinal neuronal systems. In: H a n d b o o k of Physiology. The Nervous System. Motor Control. Part I. Williams and Wilkins, Baltimore, MD, 1981: 50-95. Barker, A.T., Freeston, I.L., Jalinous, R. and Jarrat, J.A. Magnetic stimulation of the h u m a n brain and peripheral nervous system: an introduction and the results of an initial clinical evaluation. Neurosurgery, 1987, 20: 100-109.
TCCS: LATE MUSCULAR RESPONSES Berardelli, A., Inghilleri, M., Formisano, R., Accornero, N. and Manfredi, M. Stimulation of motor tracts in motor neuron disease. J. Neurol. Neurosurg. Psychiat., 1987, 50: 732-737. Berardelli, A., Inghilleri, M., Cruccu, G., Fornarelli, M., Accomero, N. and Manfredi, M. Stimulation of motor tracts in multiple sclerosis. J. Neurol. Neurosurg. Psychiat., 1988, 51: 677-683. Calancie, B., Nordin, M., Wallin, U. and Hagbarth, K.E. Motor-unit responses in human wrist flexor and extensor muscles to transcranial cortical stimulation. J. Neurophysiol., 1987, 58: 1168-1185. Cowan, J.M.A., Rothwell, J.C., Dick, J.P.R., Day, B.B., Thompson, P.D. and Marsden, C.D. Abnormalities in the central motor pathways in multiple sclerosis. Lancet, 1984, ii: 304-307. Dimitrijevir, M.R., Eaton, W.J., Sherwood, A.M. and Van der Linden, C. Assessment of corticospinal tract integrity in human spinal cord injury. In: P.M. Rossini and C.D. Marsden (Eds.), Noninvasive Stimulation of Brain and Spinal Cord; Fundamentals and Clinical Applications. Neurology and Neurobiology. Alan R. Liss, New York, 1988: 243-253. Ferrier, D. Experimental researches in cerebral physiology and pathology. West Riding Lunatic Asylum Med. Rep., 1873, 3: 1-50. Fritsch, G. und Hitzig, E. Ober die elektrische Erregbarkeit des Grosshirns. Arch. Anat. Physiol. Wiss. Med., 1870, 37: 300-332. Granit, R. Reflex self-regulation of muscle contraction and autogenetic inhibition. J. Neurophysiol., 1950, 13: 351-372. I-Iacke, W., Buchner, H., Schnippering, H. und Karsten, C.H. Motorische Potentiale nach spinaler und transcranieller Stimulation: Normalwerte fiir die Ableitung ohne willkiJrliche Vorinnervation. Z. EEG-EMG, 1987, 18: 173-178. Hassan, N.F., Rossini, P.M., Cracco, R.Q. and Cracco J.B. Unexposed motor cortex activation by low voltage stimuli. In: C. Morocutti and P.A. Rizzo (Eds.), Evoked Potentials. Neurophysiological and Clinical Aspects. Elsevier, Amsterdam, 1985: 107-113. Hess, C.W., Mills, K.R. and Murray, N.M.F. Magnetic stimulation of the human brain: facilitation of motor responses by voluntary contraction of ipsilateral and contralateral muscles with additional observations on an amputee. Neurosci. Lett., 1986, 71: 235-240. Hess, C.W., Mills, K.R. and Murray, N.M.F. Responses in small hand muscles from magnetic stimulation of the human brain. J. Physiol. (Lond.), 1987, 388: 397-419. Holmgren, H., Larsson, L.E. and Pedersen, S. Muscular responses following transcranial cortical stimulation. Electroenceph, din. Neurophysiol., 1988, 70: 36P. Hugon, J., Lubeau, M., Tabaraud, F., Chazot, F., VaUat, J.M. and Dumas, M. Central motor conduction in motor neuron disease. Ann. Neurol., 1987, 22: 544-546. Ingram, D.A. and Swash, M. Cerebral motor conduction is abnormal in motor neuron disease. J. Neurol. Neurosurg. Psychiat., 1987, 50: 159-166.
171 Ingram, D.A., Thompson, A.J. and Swash, M. Central motor conduction in multiple sclerosis: evaluation of abnormalities revealed by transcutaneous magnetic stimulation of the brain. J. Neurol. Neurosurg. Psychiat., 1988, 51: 487-494. Jasper, H.H. Report of the committee on methods of clinical examination in electroencephalography. Electroenceph. clin. Neurophysiol., 1958, 10: 370-375, Levy, W.J., York, D.H., McCoffrey, M. and Tanzer, F. Motor evoked potentials from transcranial stimulation of the motor cortex in humans. Neurosurgery, 1984, 15: 287-302. Leyton, A.S,F. and Sherrington, C.S. Observations on the excitable cortex of the chimpanzee, orangutan and gorilla. Quart. J. Exp. Physiol., 1917, 11: 135-222. Lundberg, A, Integration in the Reflex Pathway. Almqvist and Wiksell, Stockholm, 1966: 275-305. Lundberg, A. Multisensory control of spinal reflex pathways. In: R. Granit and O. Pompeiano (Eds.), Reflex Control of Posture and Movement. Progress in Brain Research, Vol. 50. Elsevier, Amsterdam, 1979: 11-28. Marsden, C.D., Merton, P.A. and Morton, H.B. Percutaneous stimulation of spinal cord and brain: pyramidal tract conduction velocities in man. J. Physiol. (Lond.), 1982, 328: 6P. Marsden, C.D., Merton, P.A. and Morton, H.B. Direct electrical stimulation of the corticospinal pathways through the intact scalp in human subjects. In: J.E. Desmedt (Ed.), Motor Control Mechanisms in Health and Disease. Raven Press, New York, 1983. Merton, P.A. and Morton, I-I.B. Electrical stimulation of human motor and visual cortex through the scalp. J. Physiol. (Lond.), 1980a, 305: 9P-10P. Merton, P.A, and Morton, H.B. Stimulation of the cerebral cortex in the intact human subject. Nature, 1980b, 285: 227. Merton, P.A,, Morton, H.B., Hill, D.K. and Marsden, C.D. Scope of a technique for electrical stimulation of the human brain, spinal cord and muscle. Lancet, 1982, i: 597-600. Meyer, B.U., Benecke, R., G~hmann, M., Zipper, S. und Conrad, B. Mrglichkeiten und C;renzen der Bestimmung zentraler motorischer Leitungszeiten beim Menschen. Z. EEG-EMG, 1987, 18: 165-172. Mills, K.R. and Murray, N.M.F. Corticospinal tract conduction time in multiple sclerosis. Ann. Neurol., 1985, 18: 601-605. Penfield, W. and Boldrey, E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain, 1937, 60: 389-443. Philips, C.G. and Porter R. Corticospinal Neurons. Academic Press, London, 1977. Rossini, P.M., Marciani, M.G., Caramia, M., Roma, V. and Zarola, F. Nervous propagation along 'central' motor pathways in intact man: characteristics of motor responses to 'bifocal' and 'unifocar spine and scalp non-invasive stimulation. Electroenceph. clin. Neurophysiol., 1985, 61: 272-286. Rothwell, J.C., Thompson, P.D., Day, B.L., Dick, J.P.R., Kachi, T., Cowan, J.M.A. and Marsden, C.D. Motor cortex stimu-
172 lation in intact man. 1. General characteristics of EMG responses in different muscles. Brain, 1987, 110: 1173-1190. Snooks, S.J. and Swash, M. Motor conduction velocity of the human spinal cord: slowed conduction in multiple sclerosis and radiation myelopathy. J. Neurol. Neurosurg. Psychiat., 1985, 48: 1135-1139. Thompson, P.D., Day, B.L., Rothwell, J.C., Dick, J.P.R., Cowan, J.M.A., Asselman, P., Griffin, G.B., Sterky, M.P.
H. HOLMGREN ET AL. and Marsden, C.D. The interpretation of electromyographic responses to electrical stimulation of the motor cortex in diseases of the upper motor neuron. J. Neurol. Sci., 1987, 80: 91-110. Young, R.R. and Cracco, R.Q. Clinical neurophysiology of conduction in central motor pathways. Editorial. Ann. Neurol., 1985, 18: 606-610.
161
Elsevier Scientific Publishers Ireland, Ltd. EEG02145
Late muscular responses to transcranial cortical stimulation in man Helen Holmgren, Lars-Erik Larsson and Stephen Pedersen Dept. of Clinical Neurophysiology, University of Linki~ping, LinkiJping (Sweden) (Accepted for publication: 26 July 1989)
Summary
Transcranial cortical stimulation was performed in healthy human subjects. Electrical and magnetic stimuli were applied over the motor areas of the hand and the leg. Surface EMGs were recorded from the antagonistic extensor and flexor carpi radialis and the anterior tibial and triceps surae muscles. Except for the well known early (primary) responses several secondary responses have been recorded. (1) A response with a latency of approximately 100 msec and with higher threshold than the primary response. It was found in approximately half of the muscles recorded from except for the anterior tibial where it was comparatively rare. It could appear both contra- and ipsilaterally and increased in amplitude with preactivation of the target muscle. It is probably not of startle origin but its appearance may be influenced by an approximately simultaneous startle effect from the scalp stimulus. (2) In some subjects a response with a latency of 50-60 msec was recorded. Sometimes it was easily separated from the primary response and from S 100 but sometimes the differentiation was difficult. (3) It was found that the known inhibitory period which follows the primary response was contralateral and increased in duration with increasing stimulus strength. The increase in duration may be due to inhibitory feedback from receptors activated by the simultaneously increasing primary muscle twitch. (4) In relaxed muscle we also observed a contralateral response with a latency exceeding 150 msec. The latency increased with increasing stimulus strength. This increase seems to be related to discontinuance of the inhibitory period. Key words: Cortical stimulation; Man; Compound muscular responses
Ever since the classical experiment of Fritsch and Hitzig (1870), Ferrier (1873) and Leyton and Sherrington (1917), electrical stimulation of the exposed cortex has provided a wealth of information about cerebral function both in animals and in man (Penfield and Boldrey 1937; Philips and Porter 1977). Transcranial cortical electrical stimulation in man (TCCS) has also been introduced (Merton and Morton 1980a,b; Merton et al. 1982; Marsden et al. 1982, 1983; Levy et al. 1984; Hassan et al.
The financial assistance of The Vivian L. Smith Foundation, Houston, TX, is gratefully acknowledged.
Correspondence to: Dr. Helen Holmgren, Dept. of Clinical Neurophysiology, University Hospital, S-581 85 LinkiSping (Sweden).
1985; Rossini et al. 1985). In the papers hitherto published, the greatest interest has been directed towards an early muscular discharge which has been considered to represent a response to excitation of fast cortico-spinal fibres. Following this early response there is an inhibitory pause of long duration (Marsden et al. 1983). Normal values for latencies to different muscles after electrical stimulation, which take stimulus strength and body length into consideration, have been published (Hacke et al. 1987; Meyer et al. 1987). In addition, transcranial magnetic stimulation, which gives less discomfort, has been introduced (Hess et al. 1986, 1987; Barker et al. 1987). It has already been shown that TCCS may be a useful tool for clinical neurophysiological diagnosis (Young and Cracco 1985). Reports of prolonged latencies, which sometimes may be due to
0013-4649/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland, Ltd.
162 prolonged central conduction time, have been published for multiple sclerosis (Cowan et al. 1984; Mills and Murray 1985; Berardelli et al. 1988; Ingram et al. 1988), in motor neurone disease (Berardelli et al. 1987; Hugon et al. 1987; Ingram and Swash 1987) and in spinal cord disease (Snooks and Swash 1985; Dimitrijevi6 et al. 1988). It is obvious that a prolonged latency following TCCS is not confined to any specific condition or pathology (Thompson et al. 1987). In repeated experiments with TCCS under different conditions we have found at least 2 additional responses with comparatively long latencies. A preliminary report of our findings has been presented to the Swedish Society for Clinical Neurophysiology (Holmgren et al. 1988). Similar discharges were independently described in a paper by Calancie et al. (1987) for the extensor carpi radialis, and for triceps surae by Dimitrijevi6 et al. (1988). The aim of the present study was to describe and analyse the late E M G responses to electrical and magnetic TCCS in the forearm and lower leg at varying stimulus strengths.
Methods Forty experiments have been performed on ourselves and on healthy medical students (10 women and 19 men) who, after being fully informed, consented to participate. Approval for the experiments was obtained from the Ethical Board of Linkrping University. The subject was comfortably seated in a half reclining position with the lower legs supported and with the arms either on the elbow rests of the investigation chair or in the lap. The hands were pronated. C a r e was taken to make the position comfortable and the subject was told to relax except when otherwise instructed. The degree of relaxation in the target muscles was checked by the experimenters on a 4-channel oscilloscope. When preactivation of the target muscles in the hands was needed the subjects pinched small springloaded devices between thumb and forefingers. When used bilaterally they gave approximately the same resistance to the pinch. For
H. HOLMGREN ET AL. preactivation of the leg muscles the experimenter administered a slight degree of resistance by hand. In 20 experiments we used a Digitimer electrical stimulator (D180), which gives high voltage condensor shocks (max 750 V) with a short time constant (50-100/~sec). The stimulator strength is given in percent of maximum. Following preliminary trials with different electrode positions, including the ring electrode described by Rossini et al. (1985), we decided to administer the stimulus through Beckman electrodes (diameter 7 mm) fastened to the scalp with collodion. The anode was Cz, C3 or C4 (Jasper 1958). Originally we placed the cathode on Fz, F3 and F4, respectively. This gives an interelectrode distance of approximately 8 cm. In later experiments we moved the F electrode in the frontal direction, thereby increasing the interelectrode distance to 10-12 cm. According to measurements made in two cases, an increase of interelectrode distance lowered the stimulus threshold but, on the other hand, resulted in a slight loss of somatotopic specificity. The time interval between the stimuli varied but was around 5 min. With a stimulus duration of 100 /~sec the threshold for a peripheral motor response was generally found to be 20-40% of the maximum strength available. In 20 additional experiments we used a Cadwell magnetic stimulator (MES-10) with a coil measuring 9 cm. For strict contralateral stimulation of the hand area the upper circumference of the coil was placed approximately 5 cm lateral to the midline. For bilateral stimulation we usually placed the centre of the coil over the vertex. The stimulation gave a twitch of the scalp which, for the electrical stimulus, was found unpleasant by some subjects. There was considerable habituation after repeated stimuli. The magnetic stimulation was not considered to be unpleasant, neither were the peripheral motor responses. Recording in most cases was done from the two antagonistic muscle pairs, extensor and flexor carpi radialis and anterior tibial and triceps surae, respectively. Although several other muscles, both proximal and distal, were investigated, the results will not be presented in this context. Skin electrodes were taped over the thickest part of the muscle belly with an interelectrode distance of 2
TCCS: LATE M U S C U L A R RESPONSES
cm. Conventional AC amplifiers were used and were connected to a Nicolet 1170 averager and to an F M tape recorder for replay when necessary. The bandwidth of the recording system was from 1 to 1000 Hz ( - 6 dB at the ends).
163 TABLE I Latencies and standard deviations (range for S >150) for muscular responses in relaxed target muscles following electrical TCCS. In the fraction values the numerator gives the n u m b e r of muscles in which the response was recorded and the denominator the n u m b e r of muscles investigated.
Results Ext. carpi
Several different responses have been recorded, some of which have already been described by other authors. They are illustrated in Fig. 1. Latencies at submaximal electrical stimulation, suggested designations and frequencies of occurrence are presented in Table I. Except for latencies of the primary responses, which may be shorter for electrical stimulation, we have not found obvious differences between the results from electrical and magnetic stimulation. (1) Primary responses were obtained in all subjects; they were contralateral to the stimulus, increased in amplitude when the target muscle was preactivated and had shorter latency and longer duration with increasing stimulus strength. In the forearm they tended to be larger in the extensor compared to the flexor carpi; in the lower leg they
Prim (msec)
S 50 (msec)
S 100 (msec)
S > 150 (msec)
16.1 + 2.5
57.3 + 8.8 11/11 50.8___5.9 11/11
85.7 + 6.5 2/11 89.6 + 7.6 3/11 117.6+9.8 8/8 109.8-t-6.8 8/8
113-225 8/11 175-200 8/11 130-225 3/8 175 6/8
9/11 Flex. carpi Tib. ant. Tric. sur.
16.5 + 2.0 4/11 30.8+2.2 2/8 33.4+2.8 1/8
were clearly larger in the anterior tibial muscle compared to the triceps surae. Most of these characteristics have been described by other authors. They will not be further discussed. Several other responses were observed; they did not appear after each single stimulus but their appearances were so constant and the latencies so consistent within the given range, that their existence cannot be questioned.
$5o
S >
150 M. ext carpi Fad.
M. f l e x o r carpi rad.
0.1 mV I
50 ms Fig. 1. Single muscular responses to electrical TCCS in a relaxed subject. The stimulus was applied over the scalp at C3-F3 with C3 positive. The records were made from the extensor carpi (upper trace) and the flexor carpi radialis (lower trace) muscles.
H. H O L M G R E N ET AL.
164
I
A
M. t r t c e p s
S 100 could appear both contralaterally and ipsilaterally to the stimulus. In 7 subjects a bilateral investigation of the arm muscles was made using magnetic stimulation. In 6 of them an S 100 response could be observed, in 1 case, however, only after preactivation of the muscles. In these
surae
0. 5 mV
A ~_
Relaxed, M. flexor carpi rod. sin
M. ext. carpi rod. sin
M flexor carpi rad. dx
I lmV .50 m s
M. ext. carpi rad. dx Preactivation
Fig. 2. Appearance of S 100 in the anterior tibial muscle as a result of preactivation. Electrical stimulus at Cz-Fz with Cz positive. A: subject relaxed. B: slight preactivation. Note the different amplification in the two records. .
(2) S 50. A small response with a latency of 50-60 msec recorded in the forearm muscles in 4 subjects (Fig. 1). It was very close to the last part of the primary response but was clearly separated from it as well as from S 100 when both occurred. (3) S 100. A response with a latency of 80-110 msec recorded in slightly more than half of the investigated extensor and flexor carpi radialis and triceps surae muscles. In the relaxed anterior tibial muscle it was found only in a few subjects. Its threshold was usually slightly higher than that of the primary response. This response probably corresponds to the E2 response observed by Calancie et al. (1987) in the extensor carpi and by Dimitrijevi6 et al. (1988) in the triceps surae muscle. Although S 100 was definitely found in relaxed muscles, it clearly increased with preactivation and then was also often found in the anterior tibial muscle (Fig. 2). The average latency of S 100 was significantly longer in the legs than in the arms ( P < 0.0001, t test). The average difference was 22 msec, which is 5 msec more than the average difference between the primary responses recorded from the same muscles.
Relaxed
|
/ 10 mV |
50 ms Fig. 3. Single muscular responses to magnetic TCCS (90%). The upper circumference of the coil was applied over the left hand area. The records were made bilaterally from the flexor and extensor carpi radialis muscles as indicated. In A and C the subject was relaxed, in B he pinched a small springloaded device on both sides.
FCCS: LATE M U S C U L A R RESPONSES
165
.__
70 */,
9 0 */o
M. e x t . carpu r'ad. dx
-
n
~tlmm
... _.
-
. i
L
M. f l e x o r c a r p i tad. dx
M. e x t . carpi tad. sin T
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
~""
_'_L
-
" -
mL__
y
_Ln
-jL--'-_
- _-~--_
_--~.
~ r C
----
M, f l e x o r carpi rad. sin
..__
100 °Io
i
NNI I
J
I ----
T
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
t
. . . . . . . . . . .
j
I
J
I
n
• . . . . .
0 . 5 mV !
50 ms Fig. 4. Single muscular responses to magnetic TCCS at the stimulus strength indicated. The upper circumference of the coil was applied over the left hand area. The records were made bilaterally from the extensor carpi and flexor carpi radialis muscles. A secondary response appears first in the contralateral extensor muscle and then bilaterally with higher stimulus strength.
cases S 100 appeared both ipsi- and contralaterally, partly independent of each other. In 2 cases there was only an ipsilateral response. An illustration of the variability is given in Fig. 3. In Fig. 3A, B and C S 100 is clearly contralateral in the extensor carpi muscle. In Fig. 3C there is in addition a slight ipsilateral response. It is not seen in Fig. 3B in spite of the fact that the muscle is slightly preactivated. In 1 case we had difficulty in deciding whether the secondary response observed was a late S 50 or an early S 100 (Fig. 4). In a few individuals we sometimes found a
generalized muscle contraction at the beginning of the experimental session, even at very low stimulus strength. It disappeared on repeated stimulation, even if the stimulus strength was increased (Fig. 5). These responses had other characteristics than the S 100, which did not habituate to repeated stimulation and which demanded higher stimulation intensity to appear. We consider them to be startle reactions due to the scalp stimulus evoked in subjects who were a bit uncertain at the beginning of the experiment. Even if these possible startle reactions are sub-
H. HOLMGREN ET AL.
166 4
I
~
~
~
%
_
M. extensor carpi rad. I M. flexor carpi
I
rad.
M. abd. dig. V
25"/o
. . . . . .
14~,~ta/"-
!
20 */o
.
_
. . . . .
j
45 */, 0.1 mV I
100 ms Fig. 5. Possible startle reaction. Electrical TCCS. Records from the extensor and flexor carpi radialis muscles and from the abductor dig. V muscle. The figures show the results from the 1st, 2nd, 4th and 14th stimuli. Note that the response disappeared at repeated stimulation even though the stimulus strength was increased. Note also the different latency and configuration of the secondary response at 45% stimulation strength.
liminal in later parts of an experimental session, they might still have an influence on the responses recorded. (4) An inhibitory response. It was contralateral to the stimulus (Fig. 6) and so far seems to be related to the p r i m a r y response. We were, however, able to confirm that inhibitory periods could appear also at very low stimulation strength without: a discernible p r i m a r y response (Fig. 7; M a r s d e n et al. 1983; Calancie et al. 1987). In 3 subjects we investigated the relation between the duration of inhibition and stimulus strength. In all cases the length of the inhibitory period increased with increased intensity of the stimulus (Figs. 8 and 10B). In some cases the inhibition was divided into several parts, with interposed excitatory periods (Figs. 2 and 7).
(5) S > 150. A response with a latency usually longer than 150 msec and which increased with increasing stimulus strength (Fig. 9). It was definitely found in relaxed muscles contralateral to the stimulus. Fig. 10A shows examples of regressions between latencies and stimulus strength in 2 subjects. This response had a strong resemblance to the r e b o u n d effect f o u n d after inhibition of a previously contracted muscle. The latency increase with increasing stimulation for S > 150 is thus similar to the increasing duration of inhibition during the same conditions (Fig. 10A and B). Discussion It is clear that the stimuli given result in a complex sequence of excitability changes in the
TCCS: LATE M U S C U L A R RESPONSES
~
d
167
x rad. dx
J
~
__ 7 ext'carp' rad. s'n
i~ flexor carpi rod sin i . . . . . . . . . . . .
D
•Jl mV 50 ms Fig. 6. Single muscular responses to magnetic TCCS (100%). The upper circumference of the coil was applied over the left hand area.
The records were made bilaterally from the extensor and flexor carpi radialis muscles preactivated by bilateral pinching. From A to D the subject increased the degree of contraction. Note that the inhibitory response is strictly contralateral and that the duration tends to decrease with increasing contraction.
motoneurones for a long period after the primary response. We have not observed obvious differences in the late responses depending on whether electrical or magnetic stimulation has been used. We have found that the inhibitory process previously described is contralateral to the stimulus and that its duration is dependent on the stimulus strength. We have also found several excitatory episodes causing contra- and ipsilateral electromyographic responses. Even though these responses do not appear after each stimulus in a given subject and at a given stimulus strength, there can be no doubt of their existence. Possible
explanations of their somewhat capricious appearance may be a competition with the simultaneous inhibitory process and partly also variations in the subjects expectations before each single stimulus. A few words about terminology. The responses that we call primary, S 100 and S > 150 have been described by Calancie et al. (1987) as E 1, E 2 and E 3. This terminology was probably inspired by their method of recording the discharge frequency of single motor units. In the present paper we have also observed another possible response, S 50. Its latency will place it between E 1 and E:.
168
H. H O L M G R E N ET AL.
responses. Finally, even if overt startle effects due to the scalp stimulation seem comparatively easy to exclude (Fig. 5), there is still a possibility that
°Io
20 ~¢~--
. . . . . . .
Relaxed
.....
0.2 mV I lOO
ms
Fig. 7. Single muscular responses to electrical TCCS applied at Cz-Fz with Cz positive. Records from the preactivated anterior tibial muscle. In A the stimulus was supraliminal, in B and C there was no discernible primary response.
Instead of calling it Ea.5 or something similar we prefer to designate the different responses with approximate latencies. It is obviously difficult to draw conclusions regarding the origin of the late responses. The electrical field of the stimulus must cover a large area, which includes not only motor but also premotor and to some extent also parietal cortex, and the stimulus may influence spinal motoneurones not only via direct cortico-spinal connections but also via intermediary segmental spinal neurones, which by themselves may be sources of both facilitatory and inhibitory effects (Baldissera et al. 1981). It is also quite possible that there are influences mediated via the spinal projections from different nuclei in the brain-stem, which themselves receive cortical projections (Lundberg 1966). In addition, the distinct twitch which is connected to ihe primary response may give rise to secondary reflex
45
~%"~"~
Relaxed 50
ms
Fig. 8. Inhibitory responses as related to increasing stimulus strength. Electrical TCCS at C3-F3 with C3 positive. The percentage stimulus strength is given to the right. Records from the right extensor carpi radialis muscle. The upper and the lowermost records show responses when the subject was relaxed. In all other records the muscle was slightly preactivated.
169
TCCS: LATE M U S C U L A R RESPONSES
the scalp stimulus produces subliminal effects in the central tone producing systems. These alternatives cannot be thoroughly discussed without further experimental analysis. A few comments may, however, be justified on the basis of some of the characteristics described. The overt inhibitory process is very constant, it is clearly contralateral to the stimulus, its duration increases with increasing stimulus strength and there is some evidence that it is also deeper. It starts immediately after the primary response and therefore has a latency related to its duration. Since the primary response is also contralateral and also increases with increasing stimulus strength, the inhibitory process may be dependent on the existence of the primary twitch. The connection between the two could, for instance, be Golgi inhibition from the twitching muscle. The inhibition should in such a case increase with increasing stimulus strength, which gives an increasing twitch. The fact that it sometimes may be obtained without any discernible primary response (Fig. 7) is not necessarily contradictory, since it is difficult to exclude a minimal primary response hidden in the voluntary activity, which in the
Latency to S > 1 5 0
A
aLE o
220
E
v
1BO
140
100 20
25
30
35
M
Stimulus strength
ext. corp. rod.
M. ~lex. carp rod
30*/*
40 *lo
Contralaterul preactivotion
,
100 ms
1 o,1 mv I
Fig. 9. Latency increase of S > 1 5 0 (arrow) with increasing stimulus strength. Responses from right extensor carpi radialis (upper trace) and flexor carpi radialis (lower trace). Electrical TCCS at C3-F3 with C3 pos. Stimulus strength 20% in A and D; 30% in B and E; 40% in C and F. In A, B and C the subject was relaxed, in D, E and F he contracted muscles in his left forearm.
40
45
50
Stimulus (*/.)
L a t e n c y to enc} of inhibition
m a g n stim
O AM
• PB
190
o
170
E
0
e 0 ~
0
0
70
"
v
0
>~ 150
o, 130
o
0
•
50
0
"
dO"
Stimulus D~
electrical stim
4G
80
"
gO
"
I00
(%)
Fig. 10. Examples of regressions for the relation between stimulus strength and the latency of S > 1 5 0 (A) and the latency to the end of inhibition (B). In A the stimulus was electrical, in B magnetic. The target muscle was extensor carpi radialis in all cases.
present experiments has been the background necessary to demonstrate the inhibition. Neither can influences from minimal twitches in adjacent muscles, which were not recorded from, be completely excluded. It is known that inhibitory action may also occur between muscles not belonging to the same myotatic unit (Granit 1950). Calancie et al. (1987), who recorded from single motor units, noted a decrease in repetition frequency as an expression of the inhibitory period. With weak stimuli this decrease could be observed also without previous excitation as evidence of a contraction. The S > 150, which is also contralateral, seems to be a counterpart of the inhibitory response,
170
appearing at the end of inhibition, thus explaining why its latency increases with increasing stimulus strength. To avoid misunderstanding it should be stressed that even if a similar discharge is seen at the end of inhibition when the target muscle has been preactivated our present description builds upon observations in relaxed muscles. Calancie et al. (1987) call it a rebound effect, which may be a very appropriate designation. They suggest that it is due to the proprioceptive inflow from the contracting target muscle. This is an obvious possibility. The excitatory effect that seems to be released at the end of inhibition may, however, also be due to long-lasting discharges from tone-producing structures in the brain-stem, themselves activated by the cortical stimulation. In fact, it seems that the cortical stimulus gives rise to long-lasting excitatory and inhibitory processes competing for dominance in the engaged muscles and that this competition ends with a rebound phenomenon. The mechanisms behind S 100 may be complex. The average latency difference for S 100 between the arms and the legs was approximately 22 msec. This should be compared to the average latency difference of 17 msec for the primary responses in the same muscles. The difference of 5 msec is highly significant ( P < 0.0001, t test). Assuming that S 100 uses the same peripheral pathway as the primary response this would speak for a slower central conduction for fibres evoking S 100 than for those giving the primary response. The assumption of the same peripheral pathway for the two responses is to some extent supported by the illustrations provided by Calancie et al. (1987), showing that the same motor unit takes part in both the primary and the late response. Although the average latencies in arms and legs were calculated on different subjects, a hypothesis of a possibly slower central pathway is still not sufficiently founded. The latency of S 100 suggests a possible startle origin due to the somewhat unpleasant electrical scalp stimulus. However, the probable startle that we observed in some subjects at the beginning of an experiment had a somewhat longer latency and habituated rapidly, which is not typical for S 100. In addition, the response is readily obtainable also with magnetic stimulation, which gives much less
H. H O L M G R E N ET AL.
discomfort. The S 100 response sometimes appears bilaterally, which may speak for startle, but the appearance is asymmetrical in an unpredictable way. Sometimes the contralateral response is better developed, sometimes the ipsilateral. A probable explanation of ipsilateral dominance, when it occurs, is that a contralateral response is suppressed by the simultaneous contralateral inhibition. Such a mechanism may explain why a slight preactivation of the target muscle can result in the appearance of a previously unobserved S 100 discharge (Fig. 2). Nothing in the previous discussion explains why S 100 is sometimes contralateral or dominates contralaterally. Supposing that collaterals from the cortico-spinal tract activate the reticular formation one would expect a bilateral response. A similar activation of the rubro-spinal system would, on the other hand, further a contralateral response. Finally, backfiring receptors activated by the primary twitches both in agonists and antagonists might contribute to the response both in facilitatory and inhibitory ways. The latency of the response is sufficiently long to give all those afferents activated by the primary twitch (Ia, Ib and FRA) the possibility of a renewed influence at the segmental level. Can it be that the late responses described in this paper are some kind of epiphenomena, without real significance, or is it possible that they are related to feedback mechanisms, which facilitate or inhibit the effect of the cortico-spinal signal (Lundberg 1979)? The fact that these late responses are comparatively weak in relation to the primary responses may contradict this suggestion, but may have its explanation in the very unphysiological way in which the cortical stimuli are given.
References Baldissera, F., Hultborn, H. and lllert, M. Integration in spinal neuronal systems. In: H a n d b o o k of Physiology. The Nervous System. Motor Control. Part I. Williams and Wilkins, Baltimore, MD, 1981: 50-95. Barker, A.T., Freeston, I.L., Jalinous, R. and Jarrat, J.A. Magnetic stimulation of the h u m a n brain and peripheral nervous system: an introduction and the results of an initial clinical evaluation. Neurosurgery, 1987, 20: 100-109.
TCCS: LATE MUSCULAR RESPONSES Berardelli, A., Inghilleri, M., Formisano, R., Accornero, N. and Manfredi, M. Stimulation of motor tracts in motor neuron disease. J. Neurol. Neurosurg. Psychiat., 1987, 50: 732-737. Berardelli, A., Inghilleri, M., Cruccu, G., Fornarelli, M., Accomero, N. and Manfredi, M. Stimulation of motor tracts in multiple sclerosis. J. Neurol. Neurosurg. Psychiat., 1988, 51: 677-683. Calancie, B., Nordin, M., Wallin, U. and Hagbarth, K.E. Motor-unit responses in human wrist flexor and extensor muscles to transcranial cortical stimulation. J. Neurophysiol., 1987, 58: 1168-1185. Cowan, J.M.A., Rothwell, J.C., Dick, J.P.R., Day, B.B., Thompson, P.D. and Marsden, C.D. Abnormalities in the central motor pathways in multiple sclerosis. Lancet, 1984, ii: 304-307. Dimitrijevir, M.R., Eaton, W.J., Sherwood, A.M. and Van der Linden, C. Assessment of corticospinal tract integrity in human spinal cord injury. In: P.M. Rossini and C.D. Marsden (Eds.), Noninvasive Stimulation of Brain and Spinal Cord; Fundamentals and Clinical Applications. Neurology and Neurobiology. Alan R. Liss, New York, 1988: 243-253. Ferrier, D. Experimental researches in cerebral physiology and pathology. West Riding Lunatic Asylum Med. Rep., 1873, 3: 1-50. Fritsch, G. und Hitzig, E. Ober die elektrische Erregbarkeit des Grosshirns. Arch. Anat. Physiol. Wiss. Med., 1870, 37: 300-332. Granit, R. Reflex self-regulation of muscle contraction and autogenetic inhibition. J. Neurophysiol., 1950, 13: 351-372. I-Iacke, W., Buchner, H., Schnippering, H. und Karsten, C.H. Motorische Potentiale nach spinaler und transcranieller Stimulation: Normalwerte fiir die Ableitung ohne willkiJrliche Vorinnervation. Z. EEG-EMG, 1987, 18: 173-178. Hassan, N.F., Rossini, P.M., Cracco, R.Q. and Cracco J.B. Unexposed motor cortex activation by low voltage stimuli. In: C. Morocutti and P.A. Rizzo (Eds.), Evoked Potentials. Neurophysiological and Clinical Aspects. Elsevier, Amsterdam, 1985: 107-113. Hess, C.W., Mills, K.R. and Murray, N.M.F. Magnetic stimulation of the human brain: facilitation of motor responses by voluntary contraction of ipsilateral and contralateral muscles with additional observations on an amputee. Neurosci. Lett., 1986, 71: 235-240. Hess, C.W., Mills, K.R. and Murray, N.M.F. Responses in small hand muscles from magnetic stimulation of the human brain. J. Physiol. (Lond.), 1987, 388: 397-419. Holmgren, H., Larsson, L.E. and Pedersen, S. Muscular responses following transcranial cortical stimulation. Electroenceph, din. Neurophysiol., 1988, 70: 36P. Hugon, J., Lubeau, M., Tabaraud, F., Chazot, F., VaUat, J.M. and Dumas, M. Central motor conduction in motor neuron disease. Ann. Neurol., 1987, 22: 544-546. Ingram, D.A. and Swash, M. Cerebral motor conduction is abnormal in motor neuron disease. J. Neurol. Neurosurg. Psychiat., 1987, 50: 159-166.
171 Ingram, D.A., Thompson, A.J. and Swash, M. Central motor conduction in multiple sclerosis: evaluation of abnormalities revealed by transcutaneous magnetic stimulation of the brain. J. Neurol. Neurosurg. Psychiat., 1988, 51: 487-494. Jasper, H.H. Report of the committee on methods of clinical examination in electroencephalography. Electroenceph. clin. Neurophysiol., 1958, 10: 370-375, Levy, W.J., York, D.H., McCoffrey, M. and Tanzer, F. Motor evoked potentials from transcranial stimulation of the motor cortex in humans. Neurosurgery, 1984, 15: 287-302. Leyton, A.S,F. and Sherrington, C.S. Observations on the excitable cortex of the chimpanzee, orangutan and gorilla. Quart. J. Exp. Physiol., 1917, 11: 135-222. Lundberg, A, Integration in the Reflex Pathway. Almqvist and Wiksell, Stockholm, 1966: 275-305. Lundberg, A. Multisensory control of spinal reflex pathways. In: R. Granit and O. Pompeiano (Eds.), Reflex Control of Posture and Movement. Progress in Brain Research, Vol. 50. Elsevier, Amsterdam, 1979: 11-28. Marsden, C.D., Merton, P.A. and Morton, H.B. Percutaneous stimulation of spinal cord and brain: pyramidal tract conduction velocities in man. J. Physiol. (Lond.), 1982, 328: 6P. Marsden, C.D., Merton, P.A. and Morton, H.B. Direct electrical stimulation of the corticospinal pathways through the intact scalp in human subjects. In: J.E. Desmedt (Ed.), Motor Control Mechanisms in Health and Disease. Raven Press, New York, 1983. Merton, P.A. and Morton, I-I.B. Electrical stimulation of human motor and visual cortex through the scalp. J. Physiol. (Lond.), 1980a, 305: 9P-10P. Merton, P.A, and Morton, H.B. Stimulation of the cerebral cortex in the intact human subject. Nature, 1980b, 285: 227. Merton, P.A,, Morton, H.B., Hill, D.K. and Marsden, C.D. Scope of a technique for electrical stimulation of the human brain, spinal cord and muscle. Lancet, 1982, i: 597-600. Meyer, B.U., Benecke, R., G~hmann, M., Zipper, S. und Conrad, B. Mrglichkeiten und C;renzen der Bestimmung zentraler motorischer Leitungszeiten beim Menschen. Z. EEG-EMG, 1987, 18: 165-172. Mills, K.R. and Murray, N.M.F. Corticospinal tract conduction time in multiple sclerosis. Ann. Neurol., 1985, 18: 601-605. Penfield, W. and Boldrey, E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain, 1937, 60: 389-443. Philips, C.G. and Porter R. Corticospinal Neurons. Academic Press, London, 1977. Rossini, P.M., Marciani, M.G., Caramia, M., Roma, V. and Zarola, F. Nervous propagation along 'central' motor pathways in intact man: characteristics of motor responses to 'bifocal' and 'unifocar spine and scalp non-invasive stimulation. Electroenceph. clin. Neurophysiol., 1985, 61: 272-286. Rothwell, J.C., Thompson, P.D., Day, B.L., Dick, J.P.R., Kachi, T., Cowan, J.M.A. and Marsden, C.D. Motor cortex stimu-
172 lation in intact man. 1. General characteristics of EMG responses in different muscles. Brain, 1987, 110: 1173-1190. Snooks, S.J. and Swash, M. Motor conduction velocity of the human spinal cord: slowed conduction in multiple sclerosis and radiation myelopathy. J. Neurol. Neurosurg. Psychiat., 1985, 48: 1135-1139. Thompson, P.D., Day, B.L., Rothwell, J.C., Dick, J.P.R., Cowan, J.M.A., Asselman, P., Griffin, G.B., Sterky, M.P.
H. HOLMGREN ET AL. and Marsden, C.D. The interpretation of electromyographic responses to electrical stimulation of the motor cortex in diseases of the upper motor neuron. J. Neurol. Sci., 1987, 80: 91-110. Young, R.R. and Cracco, R.Q. Clinical neurophysiology of conduction in central motor pathways. Editorial. Ann. Neurol., 1985, 18: 606-610.