- Email: [email protected]
Effect of stimulus intensity and voluntary contraction on corticospinal potentials following transcranial magnetic stimulation
JOURNAL
OF THE
NEUROLOGICAL SCIENCES ELSEVIER
Journal of the Neurological Sciences 139(1996) 131- 136
Effect of stimulus intensity and voluntary contraction on corticospinal potentials following transcranial magnetic stimulation Kazuo Kaneko *, Shinya Kawai, Yasunori Fuchigami, Gen Shiraishi, Takashi Ito Department
of Orthopedic
Surgen,
Unicersity
of Yamaguchi,
School of Medicine,
Kogushi,
Ube City, Yamaguchi
755, Japan
Received 25 September 1995;revised 29 January 1996;accepted 18 February 1996
Abstract Following magnetic transcranial stimulation, motor-evoked potentials (MEPs) from the abductor digiti minimi muscle, and evoked spinal cord potentials (ESCPs) from the cervical epidural space were recorded simultaneously in 9 subjects in the awake and anesthetized condition. In the awake condition, during voluntary contraction, one (n = 5) or two (n = 4) components of the ESCPs were elicited at the threshold stimulus intensity of the MEPs. As the stimulus intensity increased, an early response (n = 7) and multiple late components were recorded. The first component at high stimulus output (average 80%) preceded the small potentials elicited at threshold stimulus intensity. The latency of each component of the ESCPs during voluntary contraction was the same as that during the resting condition. In addition, the enhancement of amplitude of the ESCPs during voluntary contraction was not significant compared with that recorded at rest. During general anesthesia with volatile anesthetics, the first component of the ESCPs could be elicited at high stimulus intensity, but later components were markedly attenuated. In paired transcranial magnetic stimulation, the amplitude of this first potential following the test stimulus completely recovered within the 2 ms interstimulus interval. From these results, we hypothesized that the first component was generated non-synaptically (D-wave), but later components were generated transsynaptically (I-waves). Compound muscle action potentials (CMAPS) and F-waves also were recorded following supramaximal ulnar nerve stimulation at the wrist. Peripheral conduction time, which included synaptic delay in spinal motor neurons, was measured as follows (latency of CMAPs + latency of F-wave + I)/2 (ms). The central motor conduction time (CMCT) was measured by subtracting the peripheral conduction time from the onset latency of the MEP at high stimulus intensity in the awake state. During voluntary contraction, the calculated CMCT (4.9 f 1.O ms) was the same as the onset latency of the second component of the ESCPs (I-wave, 4.3 & 0.2 ms) recorded from the C6-C6/7 epidural space. These results suggest that transcranial magnetic stimulation generates I-waves preferentially when the stimulus intensity was set at just the threshold level of the MEPs during voluntary contraction in the awake condition. At high stimulus intensity, transcranial magnetic stimulation can elicit both D- and I-waves, but most spinal cells require I-wave activation to fire. Facilitatory effects of voluntary contraction on the muscle response following transcranial magnetic stimulation mainly originates at a spinal level. Keywords:
Central motor conduction time; Spinal cord potential; Transcranial magnetic stimulation; Corticospinal tract; Motor evoked potential;
Facilitation
1. Introduction Magnetic and electrical stimulation of the human brain can elicit muscle action potentials (Barker et al., 1986). A single magnetic cortical stimulation can produce multiple descending potentials in the human pyramidal tracts (Day et al., 1987; Berardelli et al., 1990; Burke et al., 1992, 1993). These potentials are composed of D-waves (direct activation of corticospinal neurons) and I-waves (indirect or transsynaptic activation of corticospinal neurons) (Patton and Amassian, 1954). Voluntary contraction of the
* Corresponding author. Fax: + 8 1 836 222267. 0022-510X/96/$15.00 Published by Elsevier Science B.V. PII SOO22-5 10X(96)00050-0
target muscle enhances the amplitude and shortens the latency of the muscle response (Barker et al., 1986; Hess et al., 1987; Rothwell et al., 1987; Berardelli et al., 1990; Thompson et al., 1991). The reason why the latency of the muscle response following transcranial magnetic stimulation is prolonged
in the relaxed
state compared
with that
recorded during voluntary contraction can be explained by the temporal summation of multiple corticospinal descending potentials. The onset latency of the muscle response following transcranial electrical and magnetic stimulation differ in latency even among single motor units (Hess et al., 1987; Day et al., 1989). Day et al. (1989) have proposed the ‘Dand I-wave hypothesis’ to explain the difference of MEP
132
K. Kaneko et al./Journal
of the Neurological Sciences 139 (1996) 131-136
onset latency after transcranial electrical and that after magnetic stimulation. Transcranial electric stimulation generates D-waves preferentially, while magnetic stimulation preferentially generates I-waves. However, Burke et al. (1993) have demonstrated that the transcranial magnetic stimulation can produce a D-wave similar to that observed with electrical stimulation, although the amplitude of Dwave following magnetic stimulation was smaller than that following electrical stimulation. These data provided no support for the ‘D- and I-wave hypothesis’. The contradictory results of these studies may be due to the stimulus setting and the patients’ condition (awake or anesthetized). Considerable controversy regarding the ‘D- and I-waves hypothesis’ following transcranial magnetic stimulation persists. Although several prior studies have examined evoked spinal cord potentials (ESCPs) following magnetic stimulation of the human motor cortex, most have recorded ESCPs during neurosurgical procedures on the spinal cord with general anesthesia (Berardelli et al., 1990; Thompson et al., 1991; Hicks et al., 1992a,b). No systematic study has examined the effect of stimulus intensity and voluntary contraction on magnetic induced corticospinal descending volleys in the awake human. The purpose of this study was to investigate the effect of stimulus intensity on corticospinal potentials and clarify the reliability of the ‘D- and I-wave hypothesis’ through direct observation of ESCPs following transcranial magnetic stimulation in the awake and anesthetized condition. The origin of the facilitation mechanisms during voluntary contraction following transcranial magnetic stimulation was also investigated.
2. Materials
and methods
The study was performed in 9 patients (1 woman, 8 men, aged 19-61 years old). Two patients had spinal cord tumors in the upper thoracic region (T2/3 or T5/6), and 7 patients had unilateral brachial plexus injuries. No subjective or objective neurologic findings were observed in the healthy extremities of these patients. Informed consent for the procedure and the experiments was obtained from all subjects. Epidural electrodes (Unique Medical Co., Japan, UKGlOO-5PM) were inserted percutaneously in the dorsal midline of the cervical epidural space under fluoroscopic visualization. The purpose of inserting the epidural electrodes was to prevent the iatrogenic spinal cord injury during surgery utilizing evoked spinal cord potentials (Tamaki et al., 19811, to detect the continuity of nerve roots, and to determine which nerve root was available as a donor nerve for nerve grafting in the cases of brachial plexus repair (Fuchigami et al., 1995). Evoked spinal cord potentials (ESCPs) were recorded by bipolar recording with the active electrode in a cephalad position and an interelectrode distance of 1.5 cm. The epidural electrodes consist of five recording tips with an
interelectrode distance of 1.5 cm. This design permits the recording of ESCPs simultaneously at multiple levels. Motor evoked potentials (MEPs) from the abductor digiti minimi muscle (ADM) were recorded simultaneously using self-adhesive surface electrode in a standard bellytendon position. The right ADM was examined in cases of spinal cord tumor, and the ADM on the intact side was used in cases of brachial plexus injury. The relaxation and contraction of the target muscle was monitored by surface EMG recordings amplified via a loudspeaker. During voluntary contraction, the degree of muscle excitation was maintained at about 20-30% of maximum as measured by audio-visual biofeedback. Transcranial magnetic stimulation was delivered using a round 14 cm diameter coil (Magstim, Whitland, UK, Model 200). The center of the coil was held over the Cz position in the lo-20 international system. A clockwise current in the coil was delivered to stimulate the right hemisphere, and a counterclockwise current was used to stimulate the left hemisphere as viewed from above (Rossini et al., 1994). The threshold stimulus intensity was defined as the minimal stimulus that could elicit MEPs during voluntary contraction at a recording sensitivity of 0.5 mV/D. The high stimulus intensity was defined as the intensity 20 to 30% above the threshold stimulus intensity (from 70 to 90% maximal output). The MEPs and ESCPs were amplified, filtered at 203000 Hz, averaged (II = 5-lo), and stored with a standard electromyograph (Dantec, Denmark, Counterpoint). At least two waves were recorded and superimposed in recording the ESCPs and MEPs. The latencies and the peak-to-peak amplitudes of each component of the ESCPs and the onset latencies and peak-to-peak amplitude of the MEPs were measured, and the results reported as mean k S.D. Statistical comparisons of the amplitudes and latenties of the MEPs and ESCPs between the resting and voluntary condition were made by Wilcoxon signed-rank tests. For all tests, a P-value of less than 0.05 was considered significant. Compound muscle action potentials (CMAPS) and Fwaves following peripheral supramaximal electric stimulation (square wave 0.2 ms) of the ulnar nerve at the wrist also were recorded. Twenty responses were collected, and the shortest F-wave latency was measured. The conduction time from the motor cortex to the spinal motor neurons (central motor conduction time: CMCT) was calculated by subtracting the peripheral conduction time from the onset latency of the MEPs as follows: [latency of MEP - (latency of CMAP + shortest latency of F + 1)/2 ms] (Rossini et al., 1994). The CMCT refers to the conduction time required for the descending potentials to reach the spinal motor neurons, and does not include the synaptic time of the motor neurons. ESCPs were also recorded during general anesthesia with volatile anesthetics (50:50 nitrous oxide/oxygen mixture and end-tidal isoflurane concentration > 0.7%) at the
K. Kaneko et al./Joumal C516 Magnellc Stimulation
Voluntary
Awake
(“0)
10 uv !-
son uw
133
Sciences 139 (1996) 131-136
Rt ADM
epidural
P A
of the Neurological
Magnetic
Stimulation
50%
ContractIon 5oouv
L-1 c415 epidural
30
100 UVL 2 nls
Rt ADM 40
Fig. 1. ESCPs (left trace) and muscle responses (right trace) recorded from an awake 19-year-old man with a left brachial plexus injury. Two potentials of the ESCPs were recorded at the MEP threshold stimulus intensity during voluntary contraction. As the stimulus intensity was increased, an earlier component and several later potentials were elicited.
same stimulus ihtensity utilized in the awake condition. The refractory period of the ESCPs during general anesthesia was measured using two Magstim 200 magnetic stimulators which were connected to a Bistim system (Magstim, Whitland, UK). Paired magnetic cortical stimulation was delivered at a short time interstimulus interval (O-5 ms) at high stimulus intensity.
Fig. 2. Simultaneously recorded ESCPs and muscle responses from a 61.year-old woman with a thoracic spinal tumor. At the threshold stimulus intensity in the resting state, one small ESCP potential (I-wave) was recorded but no muscle response was observed (upper two traces). A muscle response could be elicited during voluntary contraction without any change of the ESCPs (lower two traces).
Negative peak-to-peak intervals of the multiple components ranged between 1 ms and 1.5 ms. 3.2. The effect of voluntary contraction on the ESCPs and MEPs
3. Results
At the previously determined threshold stimulus intensity, the ESCPs were recorded both in the relaxed or contracted muscle condition. However, the MEPs were recorded only in the contracted muscle condition. The latency and amplitude of the ESCPs during voluntary contraction were similar to those in relaxed condition (Fig. 2). At high stimulus intensity, the latency of the MEPs was shortened, on average 2.3 k 0.8 ms, and their amplitude was increased by 150-500% during voluntary contraction as compared with those recorded during the relaxed muscle condition. The shortened latencies of all components of the ESCPs were quite minimal (0.1-0.2 ms) during voluntary contraction. No statistically significant changes were observed in the mean amplitude of the ESCPs between the resting state and during voluntary contraction of the target
3.1. MEP and ESCP changes at various magnetic stimulus intensities At the threshold stimulus intensity of the MEPs (40 + 10% output) during voluntary contraction, the ESCPs consisted either of a single small (n = 5) or two responses (n = 4). As the magnetic stimulus intensity was increased, an earlier response (n = 7) and several later components (at least 3 components) were elicited. The threshold of the first and second components was equal in 2 cases, but that of the first component was higher than that of the second component in 7 cases. Several components of the ESCPs recorded at high stimulus intensity were named in the order of increasing latencies and their polarities (Fig. 1).
Table 1 Comparison of the latency and amplitude of the evoked spinal cord potentials during resting and voluntary contraction at middle cervical level Latency (ms)
Nl (n=9) N2(n=9) N3 (n = 8) N4 (n = 8) N5 (n = 3)
Amplitude (p,V)
Rest
Voluntary
3.2 + 0.4 4.3 + 0.5 5.8+0.4 7.3 +0.5 8.7 & 0.6
3.1 f0.4 4.2 f 0.5 5.8 f 0.4 7.2 f 0.5 8.5 f 0.7
NS NS NS NS NS
PI-N1 P2-N2 P3-N3 P4-N4 P5-N5
(n (n (n (n (n
= = = = =
9) 9) 8) 8) 3)
Rest
Voluntary
8.2 & 6.3 13.4+5.8 10.3+5.8 11.9k6.2 7.6 f4.0
8.8 i 5.7 14.0+6.2 10.9+4.8 13.8 + 7.6 1.5 + 2.5
NS NS NS NS NS
134
K. Kuneko et al. /Journal
Nevmlogicd
Sttmulatlon
Magne!ic
Voluntary
of the
Sciences
139 (19961 131- 136
go”o
Contraction
CMAP
3 9 ins
Fig. 3. ESCPs recorded from the same subjects as in Fig. 1 in the awake and anesthetized condition (Isoflurane). Only the first component (D-wave) was recorded at high stimulus intensity during anesthesia.
muscles. However, a slight enhancement of late ESCP components was observed (Table 1, Fig. 3). 3.3. Comparison of the central motor conduction time (CMCT) and the latencies of the ESCPs The latencies of each component of the ESCPs, which were recorded at the C6 to C6/7 to correspond to the spinal motor neuron level of the abductor digiti minimi muscles, were compared to the calculated CMCT. The averaged CMCT was 6.8 k 1.3 ms in the relaxed state, and 4.9 f 1.0 ms during voluntary contraction. During voluntary contraction, the mean value of the CMCT was similar to the P2 latency of the ESCPs, while in the resting state it correlated best with the P4 latency (Table 2, Fig. 3). In one case, the CMCT during voluntary contraction was similar to the Pl latency (Table 2, case 2). 3.4. EfSect of volatile anesthetics on the ESCPs
(Isoflurane). In all subjects, the ESCPs could not be recorded at the prior threshold stimulus intensity. At a high stimulus intensity, only the first component was recorded routinely, and the late components were markedly attenuated. The peak latencies of the first components were essentially equivalent (Fig. 4). The refractory period of the first ESCP component was measured in 4 cases at high stimulus intensity. The amplitude of this first component
Evoked Magnetic Stlmulatlon (“0)
30
1
Spinal Cord Potentials lsoflurane
Awake
C%.>-
--
1.5 % lo Yk ,._.,: .., y,
I 1e uv/o
40
In all patients, ESCPs also were recorded intraoperatively while the patients were receiving volatile anesthetics Table 2 Comparison of the latency of the evoked spinal cord potentials and central motor conduction time Latency of ESCPs (MS)
CMCT (ms)
Case
PI
P2
P3
P4
P5
Rest
Voluntary
I 2 3 4 5 6 7 8 9
3.0 2.7 2.9 3.1 3.3 3.1 3.1 3.4 2.9
4.0 4.1 4.5 4.2 4.5 4.4 4.3 4.5 3.8
5.3 5.4 5.7 5.8 6.2 5.9 6.0 NE 5.2
6.9 6.7 7.0 7.0 7.3 1.2 7.2 NE 6.6
8.4 8.9 NE NE NE NE NE NE 7.9
1.4 4.3 8.0 6.9 8.2 5.1 1.5 NE 6.3
3.9 3.0 5.6 5.0 5.1 4.8 6.5 5.0 5.3
60
C516
epldutal
Fig. 4. Simultaneously recorded ESCPs and muscle responses from a 61.year-old man with a thoracic spinal cord tumor. A shortening of the onset latency and a marked enlargement of the muscle response were observed during voluntary contraction compared with those during the resting state. The latency of each component of the ESCPs during voluntary contraction was the same as that recorded during the resting condition. However, a minimal enhancement of the late components of the ESCPs was observed during voluntary contraction. CMAP and Fwaves were recorded (right traces). Central conduction time in the resting state (7.4 ms) corresponded to the P4 latency of the ESCPs, during voluntary contraction (3.9 ms), it corresponded to the P2 latency (Table I, case I).
K. Kanrko
Paired Magnetic
et trl. / Journal
of the
Neurological
Stimulation C 4 epidural
I
2
0 Smgle
1 Ill
n
lsoflurane 1 .O% Fig. 5. ESCPs following paired transcranial magnetic stimulation while recieving volatile anesthetics. The amplitude of the first components of the ESCPs following a test stimulation was recovered within 2 ms of interstimulus inteval.
following a test stimulus completely recovered within a 2-ms interstimulus interval (Fig. 5).
4. Discussion In human studies, evoked spinal cord responses following cortical stimulation can be recorded from the epidural space. Single electric or magnetic cortical stimulation can elicit multiple descending potentials in the human pyramidal tract (Berardelli et al., 1990; Burke et al., 1992, 1993). This response is composed of an initial wave (D) followed by later waves (I) (Patton and Amassian, 1954). We recorded the multiple components of the ESCPs from the cervical epidural space at a high stimulus output. The effect of a volatile anesthetics (Isoflurane) on ESCPs following transcranial magnetic stimulation was also consistent with results from prior studies; the first component (D-wave) was unaffected but later components (I-waves) were attenuated with isoflurane (Hicks et al., 1992a,b; Burke et al.. 1993). In paired transcranial magnetic stimulation, the amplitude of the first component of the ESCPs following a test stimulus completely recovered within a 2-ms interstimulus interval (Fig. 5). We hypothesize that this potential was generated non-synaptically. From these results, although we did not perform a direct comparison between electrical and magnetic stimulation, we concluded that the first component represented a D-wave, and that later components were I-waves. The muscle response following magnetic stimulation has a longer latency than that following electric stimulation (Day et al., 1989). Day et al. (1989) have proposed the ‘D- and I-wave hypothesis’ to explain the difference in onset latency of the MEPs after
Sciences 139 (1996)
131-136
135
transcranial electrical and after magnetic stimulation. (Electric stimulation preferentially generates D-waves while magnetic stimulation preferentially generates I waves.) However, Burke et al. (1993) subsequently demonstrated that the transcranial magnetic stimulation can produce a D-wave similar to that observed following electrical stimulation. Our study demonstrated that transcranial magnetic stimulation at the threshold stimulus intensity of the MEPs during voluntary contraction did not always produce the first component (D-wave) of the ESCPs, although it was always recorded at high stimulus intensity. Theses results suggest that transcranial magnetic stimulation preferentially generates I-waves in the corticospinal tract, and that the ‘D- and I-wave hypothesis’ is reliable when the stimulus intensity is near threshold level of the MEPs during voluntary contraction. However, this hypothesis is not reasonable when the stimulus intensity is higher because both D- and I-waves can be elicited at a high stimulus intensity (Burke et al., 19931. Voluntary contraction in the target muscle shortens the onset latency, lowers the threshold, and increases the amplitude of the muscle response (Barker et al., 1986; Hess et al., 1987; Merton et al., 1982; Rothwell et al., 1987; Thompson et al., 1991). Our results demonstrated that the latency shortening can range from 2 ms to 3.5 ms, and that the amplitude during voluntary contraction can be increased to 150-500% of that recorded during the resting state. In the voluntary contraction state, the summation of descending voluntary impulses from cortical areas, afferent impulses from muscle spindles and descending potentials secondary to magnetic brain stimulation can activate the spinal motor neurons earlier than under resting conditions and shorten the MEP onset latency (Claus et al., 1988; Rossini et al., 1987a,b). Our results support this hypothesis because the shortening of the MEP latency was observed without attenuation of the ESCP latency during voluntary contraction. This observation suggests that the shortened latency of the MEPs during voluntary contraction originate in the spinal summation mechanism and not a supraspinal mechanism. This hypothesis was also supported by the results that demonstrated no significant difference between the amplitudes of the evoked muscle response following electric cervical column stimulation compared to those following magnetic cortical stimulation (Maertens de Noordhout et al., 1992). The present study also showed that the mean calculated CMCT of the abductor digiti minimi corresponded best to the measured mean onset latency of the second component of the ESCPs during voluntary contraction. In only one case (case 21, was the calculated CMCT during voluntary contraction similar to the onset latency of the first response (D-wave). These results suggest that magnetic cortical stimulation can produce D-waves but that spinal motor neurons require I-wave stimulation in order to fire in most subjects. Two possible mechanisms may explain why spinal motor neurons were not activated by the D-wave following
136
K. Kaneko et al./Joumal
of the Neurological Sciences 139 (1996) 131-136
magnetic stimulation: (1) The amplitude of the D-wave was too small to recruit spinal motor neurons; and (2) D-waves which were elicited in this manner did not stimulate spinal motor neurons of the hand muscles, but did activate those innervating the muscles of the trunk or lower extremities. Although D-waves were elicited following magnetic stimulation at a high stimulus output, they are most effectively produced by electrical stimulation. (Amassian et al., 1990; Berardelli et al., 1990; Burke et al., 1993). Indeed, the first component (D-waves) had a smaller amplitude than the second or third components (I-waves) in our study. The hypothesis that D-waves, which were elicited in this coil position, did not distribute to spinal motor neurons of the hand muscles may explain this phenomenon. To resolve this hypothesis, we plan to record ESCPs simultaneously from the cervical and thoracic epidural space. In conclusion, the present study supports the hypothesis that transcranial magnetic stimulation preferentially generates I-waves at the threshold stimulus intensity of the MEPs during voluntary contraction, but it generates both D- and I-waves at a high stimulus intensity. The ‘D- and I-wave hypothesis’ is reliable only when the stimulus intensity is near the threshold of the MEPs during voluntary contraction. Our results also demonstrate that the shortened MEPs onset latency can be explained by a temporal summation mechanism associated with voluntary muscle contraction at the spinal level.
References Amassian, V.E., Quirk, G. and Stewart, M.A. (1990) A comparison of corticospinal activation by magnetic coil and electrical stimulation of monkey cortex. Electroenceph. Clin. Neurophysiol., 77: 390-401. Barker, A.T., Freeston, I.L., Jalinous, R. and Jarratt, J. (1986) Clinical evaluation of conduction time measurements in central motor pathways using magnetic stimulation of the human brain. Lance%, i: 132% 1326. Berardelli, A., Inghilleri, M., Cruccu, G. and Manfredi, M. (1990) Descending volley after electrical and magnetic transcranial stimulation in man. Neurosci. Lett., 112: 54-58. Burke, D., Hicks, R., Stephen, J., Woodforth, I. and Crawford, M. (1992) Assessment of corticospinal and somatosensory conduction simultaneously during scoliosis surgery. Electroencephalogr. Clin. Neurophysiol., 83: 388-396. Burke, D., Hicks, R., Gandevia, S.C., Stephen, J., Woodforth, I. and Crawford, M. (1993) Direct comparison of corticospinal volleys in human subjects to transcranial magnetic and electrical stimulation. J. Physiol., 470: 383-393. Claus, D., Mills, K.R. and Murray, N.M.F. (1988) Facilitation of muscle
responses to magnetic brain stimulation by mechanical stimuli in man. Exp. Brain Res., 71: 273-278. Day, B.L., Rothwell, J.C., Thompson, P.D., Dick, J.P.R., Cowman, J.M.A., Berardelli, A. and Marsden, C.D. (1987) Motor cortex stimulation in intact man. I. Multiple descending volleys. Brain, 110: 1191-1209. Day, B.L., Dressier, D., Maertens de Noordhout, A., Marsden, C.D., Nakashima, K., Rothwell, J.C. and Thompson, P.D. (1989) Electric and magnetic stimulation of human motor cortex: surface EMG and single motor unit responses. J. Physiol., 412: 449-473. Fuchigami, Y., Kawai, S., Doi, K., Shiraishi, G., Ito, T., Kaneko, K., Hashida, T., Kawamura, H. and Oofuji, A. (1995) Intraoperative electrodiagnosis for brachial plexus injury. Electroencephalogr. Clin. Neurophysiol. 97; S177. Hess, C.W., Mills, K.R. and Murray, N.M.F. (1987) Responses in small hand muscles from magnetic stimulator of the human brain. J. Physiol., 388: 397-419. Hicks, R., Burke. D., Stephen, J., Woodforth, I. and Crawford, M. (1992al Corticospinal volleys evoked by electrical stimulation of human motor cortex after withdrawal of volatile anesthetics. J. Physiol., 456: 393-404. Hicks, R., Woodforth, I., Crawford, M., Stephen, J. and Burke, D. (1992b) Some effects of isoflurane on I waves of the motor evoked potentials. Br. J. Anaesthiol., 69: 130-136. Maertens de Noordhout, A., Pepin. L., Gerard, P., Delweide, P.J. (1992) Facilitation of responses to motor cortex stimulation: effect of isometric voluntary contraction. Ann. Neural. 32: 365-370. Merton, P.A., Morton, H.B., Hill, D.K. and Marsden, C.D. (1982) Scope of a technique for electrical stimulation of human brain, spinal cord, and muscle. Lancet, ii: 597-600. Patton, H.D. and Amassian, V.E. (1954) Single and multiple unit analysis of cortical stage of pyramidal tract activation. J. Neurophysiol., 17: 345-363. Rossini, P.M., Caramia, M.D. and Zarola, F. (1987a) Mechanisms of nervous propagation along central motor pathways: non-invasive evaluation in healthy subjects and in patients with neurological disease. Neurosurgery, 20: 183- 19 1. Rossini, P.M., Caramia, M.D., Zarola, F. (1987b) Central motor tract propagation in man: studies with non-invasive, unifocal, scalp stimulation. Brain Res., 415: 21 l-225. Rossini, P.M., Barker, A.T., Berardelli, A., Caramia, M.D., Caruso, G., Cracco, R.Q., Dimitrijevic, M.R., Hallet, M., Katayama, Y., Lucking, C.H., Maertens de Noordhout, A.L., Marsden, C.D., Murray, N.M.F, Rothwell, J.C., Swash, M. and Tomberg, C. (1994) Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application, Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 91: 79-92. Rothwell, J.C., Thompson, P.D., Day, B.L., Dick, J.P.R., Kachi, T., Cowan, J.M.A. and Marsden, C.D. (1987) Motor cortical stimulation in intact man. I. General characteristics of EMG responses in different muscles. Brain, 110: 1173-I 190. Tamaki, T., Tsuji, H., Inoue S. and Kobayashi, H. (1981) The prevention of iatrogenic spinal cord injury utilizing the evoked spinal cord potentials. Int. Orthop., 4: 313-3 17 Thompson, P.D., Day, B.L., Rothwell, J.C., Dressier, D., Maertens de Noordhout, A.L. and Marsden, C.D. (1991) Further observations on the facilitation of muscle responses to cortical stimulation by voluntary contraction. Electroenceph. Clin. Neurophysiol., 81: 397-402.
OF THE
NEUROLOGICAL SCIENCES ELSEVIER
Journal of the Neurological Sciences 139(1996) 131- 136
Effect of stimulus intensity and voluntary contraction on corticospinal potentials following transcranial magnetic stimulation Kazuo Kaneko *, Shinya Kawai, Yasunori Fuchigami, Gen Shiraishi, Takashi Ito Department
of Orthopedic
Surgen,
Unicersity
of Yamaguchi,
School of Medicine,
Kogushi,
Ube City, Yamaguchi
755, Japan
Received 25 September 1995;revised 29 January 1996;accepted 18 February 1996
Abstract Following magnetic transcranial stimulation, motor-evoked potentials (MEPs) from the abductor digiti minimi muscle, and evoked spinal cord potentials (ESCPs) from the cervical epidural space were recorded simultaneously in 9 subjects in the awake and anesthetized condition. In the awake condition, during voluntary contraction, one (n = 5) or two (n = 4) components of the ESCPs were elicited at the threshold stimulus intensity of the MEPs. As the stimulus intensity increased, an early response (n = 7) and multiple late components were recorded. The first component at high stimulus output (average 80%) preceded the small potentials elicited at threshold stimulus intensity. The latency of each component of the ESCPs during voluntary contraction was the same as that during the resting condition. In addition, the enhancement of amplitude of the ESCPs during voluntary contraction was not significant compared with that recorded at rest. During general anesthesia with volatile anesthetics, the first component of the ESCPs could be elicited at high stimulus intensity, but later components were markedly attenuated. In paired transcranial magnetic stimulation, the amplitude of this first potential following the test stimulus completely recovered within the 2 ms interstimulus interval. From these results, we hypothesized that the first component was generated non-synaptically (D-wave), but later components were generated transsynaptically (I-waves). Compound muscle action potentials (CMAPS) and F-waves also were recorded following supramaximal ulnar nerve stimulation at the wrist. Peripheral conduction time, which included synaptic delay in spinal motor neurons, was measured as follows (latency of CMAPs + latency of F-wave + I)/2 (ms). The central motor conduction time (CMCT) was measured by subtracting the peripheral conduction time from the onset latency of the MEP at high stimulus intensity in the awake state. During voluntary contraction, the calculated CMCT (4.9 f 1.O ms) was the same as the onset latency of the second component of the ESCPs (I-wave, 4.3 & 0.2 ms) recorded from the C6-C6/7 epidural space. These results suggest that transcranial magnetic stimulation generates I-waves preferentially when the stimulus intensity was set at just the threshold level of the MEPs during voluntary contraction in the awake condition. At high stimulus intensity, transcranial magnetic stimulation can elicit both D- and I-waves, but most spinal cells require I-wave activation to fire. Facilitatory effects of voluntary contraction on the muscle response following transcranial magnetic stimulation mainly originates at a spinal level. Keywords:
Central motor conduction time; Spinal cord potential; Transcranial magnetic stimulation; Corticospinal tract; Motor evoked potential;
Facilitation
1. Introduction Magnetic and electrical stimulation of the human brain can elicit muscle action potentials (Barker et al., 1986). A single magnetic cortical stimulation can produce multiple descending potentials in the human pyramidal tracts (Day et al., 1987; Berardelli et al., 1990; Burke et al., 1992, 1993). These potentials are composed of D-waves (direct activation of corticospinal neurons) and I-waves (indirect or transsynaptic activation of corticospinal neurons) (Patton and Amassian, 1954). Voluntary contraction of the
* Corresponding author. Fax: + 8 1 836 222267. 0022-510X/96/$15.00 Published by Elsevier Science B.V. PII SOO22-5 10X(96)00050-0
target muscle enhances the amplitude and shortens the latency of the muscle response (Barker et al., 1986; Hess et al., 1987; Rothwell et al., 1987; Berardelli et al., 1990; Thompson et al., 1991). The reason why the latency of the muscle response following transcranial magnetic stimulation is prolonged
in the relaxed
state compared
with that
recorded during voluntary contraction can be explained by the temporal summation of multiple corticospinal descending potentials. The onset latency of the muscle response following transcranial electrical and magnetic stimulation differ in latency even among single motor units (Hess et al., 1987; Day et al., 1989). Day et al. (1989) have proposed the ‘Dand I-wave hypothesis’ to explain the difference of MEP
132
K. Kaneko et al./Journal
of the Neurological Sciences 139 (1996) 131-136
onset latency after transcranial electrical and that after magnetic stimulation. Transcranial electric stimulation generates D-waves preferentially, while magnetic stimulation preferentially generates I-waves. However, Burke et al. (1993) have demonstrated that the transcranial magnetic stimulation can produce a D-wave similar to that observed with electrical stimulation, although the amplitude of Dwave following magnetic stimulation was smaller than that following electrical stimulation. These data provided no support for the ‘D- and I-wave hypothesis’. The contradictory results of these studies may be due to the stimulus setting and the patients’ condition (awake or anesthetized). Considerable controversy regarding the ‘D- and I-waves hypothesis’ following transcranial magnetic stimulation persists. Although several prior studies have examined evoked spinal cord potentials (ESCPs) following magnetic stimulation of the human motor cortex, most have recorded ESCPs during neurosurgical procedures on the spinal cord with general anesthesia (Berardelli et al., 1990; Thompson et al., 1991; Hicks et al., 1992a,b). No systematic study has examined the effect of stimulus intensity and voluntary contraction on magnetic induced corticospinal descending volleys in the awake human. The purpose of this study was to investigate the effect of stimulus intensity on corticospinal potentials and clarify the reliability of the ‘D- and I-wave hypothesis’ through direct observation of ESCPs following transcranial magnetic stimulation in the awake and anesthetized condition. The origin of the facilitation mechanisms during voluntary contraction following transcranial magnetic stimulation was also investigated.
2. Materials
and methods
The study was performed in 9 patients (1 woman, 8 men, aged 19-61 years old). Two patients had spinal cord tumors in the upper thoracic region (T2/3 or T5/6), and 7 patients had unilateral brachial plexus injuries. No subjective or objective neurologic findings were observed in the healthy extremities of these patients. Informed consent for the procedure and the experiments was obtained from all subjects. Epidural electrodes (Unique Medical Co., Japan, UKGlOO-5PM) were inserted percutaneously in the dorsal midline of the cervical epidural space under fluoroscopic visualization. The purpose of inserting the epidural electrodes was to prevent the iatrogenic spinal cord injury during surgery utilizing evoked spinal cord potentials (Tamaki et al., 19811, to detect the continuity of nerve roots, and to determine which nerve root was available as a donor nerve for nerve grafting in the cases of brachial plexus repair (Fuchigami et al., 1995). Evoked spinal cord potentials (ESCPs) were recorded by bipolar recording with the active electrode in a cephalad position and an interelectrode distance of 1.5 cm. The epidural electrodes consist of five recording tips with an
interelectrode distance of 1.5 cm. This design permits the recording of ESCPs simultaneously at multiple levels. Motor evoked potentials (MEPs) from the abductor digiti minimi muscle (ADM) were recorded simultaneously using self-adhesive surface electrode in a standard bellytendon position. The right ADM was examined in cases of spinal cord tumor, and the ADM on the intact side was used in cases of brachial plexus injury. The relaxation and contraction of the target muscle was monitored by surface EMG recordings amplified via a loudspeaker. During voluntary contraction, the degree of muscle excitation was maintained at about 20-30% of maximum as measured by audio-visual biofeedback. Transcranial magnetic stimulation was delivered using a round 14 cm diameter coil (Magstim, Whitland, UK, Model 200). The center of the coil was held over the Cz position in the lo-20 international system. A clockwise current in the coil was delivered to stimulate the right hemisphere, and a counterclockwise current was used to stimulate the left hemisphere as viewed from above (Rossini et al., 1994). The threshold stimulus intensity was defined as the minimal stimulus that could elicit MEPs during voluntary contraction at a recording sensitivity of 0.5 mV/D. The high stimulus intensity was defined as the intensity 20 to 30% above the threshold stimulus intensity (from 70 to 90% maximal output). The MEPs and ESCPs were amplified, filtered at 203000 Hz, averaged (II = 5-lo), and stored with a standard electromyograph (Dantec, Denmark, Counterpoint). At least two waves were recorded and superimposed in recording the ESCPs and MEPs. The latencies and the peak-to-peak amplitudes of each component of the ESCPs and the onset latencies and peak-to-peak amplitude of the MEPs were measured, and the results reported as mean k S.D. Statistical comparisons of the amplitudes and latenties of the MEPs and ESCPs between the resting and voluntary condition were made by Wilcoxon signed-rank tests. For all tests, a P-value of less than 0.05 was considered significant. Compound muscle action potentials (CMAPS) and Fwaves following peripheral supramaximal electric stimulation (square wave 0.2 ms) of the ulnar nerve at the wrist also were recorded. Twenty responses were collected, and the shortest F-wave latency was measured. The conduction time from the motor cortex to the spinal motor neurons (central motor conduction time: CMCT) was calculated by subtracting the peripheral conduction time from the onset latency of the MEPs as follows: [latency of MEP - (latency of CMAP + shortest latency of F + 1)/2 ms] (Rossini et al., 1994). The CMCT refers to the conduction time required for the descending potentials to reach the spinal motor neurons, and does not include the synaptic time of the motor neurons. ESCPs were also recorded during general anesthesia with volatile anesthetics (50:50 nitrous oxide/oxygen mixture and end-tidal isoflurane concentration > 0.7%) at the
K. Kaneko et al./Joumal C516 Magnellc Stimulation
Voluntary
Awake
(“0)
10 uv !-
son uw
133
Sciences 139 (1996) 131-136
Rt ADM
epidural
P A
of the Neurological
Magnetic
Stimulation
50%
ContractIon 5oouv
L-1 c415 epidural
30
100 UVL 2 nls
Rt ADM 40
Fig. 1. ESCPs (left trace) and muscle responses (right trace) recorded from an awake 19-year-old man with a left brachial plexus injury. Two potentials of the ESCPs were recorded at the MEP threshold stimulus intensity during voluntary contraction. As the stimulus intensity was increased, an earlier component and several later potentials were elicited.
same stimulus ihtensity utilized in the awake condition. The refractory period of the ESCPs during general anesthesia was measured using two Magstim 200 magnetic stimulators which were connected to a Bistim system (Magstim, Whitland, UK). Paired magnetic cortical stimulation was delivered at a short time interstimulus interval (O-5 ms) at high stimulus intensity.
Fig. 2. Simultaneously recorded ESCPs and muscle responses from a 61.year-old woman with a thoracic spinal tumor. At the threshold stimulus intensity in the resting state, one small ESCP potential (I-wave) was recorded but no muscle response was observed (upper two traces). A muscle response could be elicited during voluntary contraction without any change of the ESCPs (lower two traces).
Negative peak-to-peak intervals of the multiple components ranged between 1 ms and 1.5 ms. 3.2. The effect of voluntary contraction on the ESCPs and MEPs
3. Results
At the previously determined threshold stimulus intensity, the ESCPs were recorded both in the relaxed or contracted muscle condition. However, the MEPs were recorded only in the contracted muscle condition. The latency and amplitude of the ESCPs during voluntary contraction were similar to those in relaxed condition (Fig. 2). At high stimulus intensity, the latency of the MEPs was shortened, on average 2.3 k 0.8 ms, and their amplitude was increased by 150-500% during voluntary contraction as compared with those recorded during the relaxed muscle condition. The shortened latencies of all components of the ESCPs were quite minimal (0.1-0.2 ms) during voluntary contraction. No statistically significant changes were observed in the mean amplitude of the ESCPs between the resting state and during voluntary contraction of the target
3.1. MEP and ESCP changes at various magnetic stimulus intensities At the threshold stimulus intensity of the MEPs (40 + 10% output) during voluntary contraction, the ESCPs consisted either of a single small (n = 5) or two responses (n = 4). As the magnetic stimulus intensity was increased, an earlier response (n = 7) and several later components (at least 3 components) were elicited. The threshold of the first and second components was equal in 2 cases, but that of the first component was higher than that of the second component in 7 cases. Several components of the ESCPs recorded at high stimulus intensity were named in the order of increasing latencies and their polarities (Fig. 1).
Table 1 Comparison of the latency and amplitude of the evoked spinal cord potentials during resting and voluntary contraction at middle cervical level Latency (ms)
Nl (n=9) N2(n=9) N3 (n = 8) N4 (n = 8) N5 (n = 3)
Amplitude (p,V)
Rest
Voluntary
3.2 + 0.4 4.3 + 0.5 5.8+0.4 7.3 +0.5 8.7 & 0.6
3.1 f0.4 4.2 f 0.5 5.8 f 0.4 7.2 f 0.5 8.5 f 0.7
NS NS NS NS NS
PI-N1 P2-N2 P3-N3 P4-N4 P5-N5
(n (n (n (n (n
= = = = =
9) 9) 8) 8) 3)
Rest
Voluntary
8.2 & 6.3 13.4+5.8 10.3+5.8 11.9k6.2 7.6 f4.0
8.8 i 5.7 14.0+6.2 10.9+4.8 13.8 + 7.6 1.5 + 2.5
NS NS NS NS NS
134
K. Kuneko et al. /Journal
Nevmlogicd
Sttmulatlon
Magne!ic
Voluntary
of the
Sciences
139 (19961 131- 136
go”o
Contraction
CMAP
3 9 ins
Fig. 3. ESCPs recorded from the same subjects as in Fig. 1 in the awake and anesthetized condition (Isoflurane). Only the first component (D-wave) was recorded at high stimulus intensity during anesthesia.
muscles. However, a slight enhancement of late ESCP components was observed (Table 1, Fig. 3). 3.3. Comparison of the central motor conduction time (CMCT) and the latencies of the ESCPs The latencies of each component of the ESCPs, which were recorded at the C6 to C6/7 to correspond to the spinal motor neuron level of the abductor digiti minimi muscles, were compared to the calculated CMCT. The averaged CMCT was 6.8 k 1.3 ms in the relaxed state, and 4.9 f 1.0 ms during voluntary contraction. During voluntary contraction, the mean value of the CMCT was similar to the P2 latency of the ESCPs, while in the resting state it correlated best with the P4 latency (Table 2, Fig. 3). In one case, the CMCT during voluntary contraction was similar to the Pl latency (Table 2, case 2). 3.4. EfSect of volatile anesthetics on the ESCPs
(Isoflurane). In all subjects, the ESCPs could not be recorded at the prior threshold stimulus intensity. At a high stimulus intensity, only the first component was recorded routinely, and the late components were markedly attenuated. The peak latencies of the first components were essentially equivalent (Fig. 4). The refractory period of the first ESCP component was measured in 4 cases at high stimulus intensity. The amplitude of this first component
Evoked Magnetic Stlmulatlon (“0)
30
1
Spinal Cord Potentials lsoflurane
Awake
C%.>-
--
1.5 % lo Yk ,._.,: .., y,
I 1e uv/o
40
In all patients, ESCPs also were recorded intraoperatively while the patients were receiving volatile anesthetics Table 2 Comparison of the latency of the evoked spinal cord potentials and central motor conduction time Latency of ESCPs (MS)
CMCT (ms)
Case
PI
P2
P3
P4
P5
Rest
Voluntary
I 2 3 4 5 6 7 8 9
3.0 2.7 2.9 3.1 3.3 3.1 3.1 3.4 2.9
4.0 4.1 4.5 4.2 4.5 4.4 4.3 4.5 3.8
5.3 5.4 5.7 5.8 6.2 5.9 6.0 NE 5.2
6.9 6.7 7.0 7.0 7.3 1.2 7.2 NE 6.6
8.4 8.9 NE NE NE NE NE NE 7.9
1.4 4.3 8.0 6.9 8.2 5.1 1.5 NE 6.3
3.9 3.0 5.6 5.0 5.1 4.8 6.5 5.0 5.3
60
C516
epldutal
Fig. 4. Simultaneously recorded ESCPs and muscle responses from a 61.year-old man with a thoracic spinal cord tumor. A shortening of the onset latency and a marked enlargement of the muscle response were observed during voluntary contraction compared with those during the resting state. The latency of each component of the ESCPs during voluntary contraction was the same as that recorded during the resting condition. However, a minimal enhancement of the late components of the ESCPs was observed during voluntary contraction. CMAP and Fwaves were recorded (right traces). Central conduction time in the resting state (7.4 ms) corresponded to the P4 latency of the ESCPs, during voluntary contraction (3.9 ms), it corresponded to the P2 latency (Table I, case I).
K. Kanrko
Paired Magnetic
et trl. / Journal
of the
Neurological
Stimulation C 4 epidural
I
2
0 Smgle
1 Ill
n
lsoflurane 1 .O% Fig. 5. ESCPs following paired transcranial magnetic stimulation while recieving volatile anesthetics. The amplitude of the first components of the ESCPs following a test stimulation was recovered within 2 ms of interstimulus inteval.
following a test stimulus completely recovered within a 2-ms interstimulus interval (Fig. 5).
4. Discussion In human studies, evoked spinal cord responses following cortical stimulation can be recorded from the epidural space. Single electric or magnetic cortical stimulation can elicit multiple descending potentials in the human pyramidal tract (Berardelli et al., 1990; Burke et al., 1992, 1993). This response is composed of an initial wave (D) followed by later waves (I) (Patton and Amassian, 1954). We recorded the multiple components of the ESCPs from the cervical epidural space at a high stimulus output. The effect of a volatile anesthetics (Isoflurane) on ESCPs following transcranial magnetic stimulation was also consistent with results from prior studies; the first component (D-wave) was unaffected but later components (I-waves) were attenuated with isoflurane (Hicks et al., 1992a,b; Burke et al.. 1993). In paired transcranial magnetic stimulation, the amplitude of the first component of the ESCPs following a test stimulus completely recovered within a 2-ms interstimulus interval (Fig. 5). We hypothesize that this potential was generated non-synaptically. From these results, although we did not perform a direct comparison between electrical and magnetic stimulation, we concluded that the first component represented a D-wave, and that later components were I-waves. The muscle response following magnetic stimulation has a longer latency than that following electric stimulation (Day et al., 1989). Day et al. (1989) have proposed the ‘D- and I-wave hypothesis’ to explain the difference in onset latency of the MEPs after
Sciences 139 (1996)
131-136
135
transcranial electrical and after magnetic stimulation. (Electric stimulation preferentially generates D-waves while magnetic stimulation preferentially generates I waves.) However, Burke et al. (1993) subsequently demonstrated that the transcranial magnetic stimulation can produce a D-wave similar to that observed following electrical stimulation. Our study demonstrated that transcranial magnetic stimulation at the threshold stimulus intensity of the MEPs during voluntary contraction did not always produce the first component (D-wave) of the ESCPs, although it was always recorded at high stimulus intensity. Theses results suggest that transcranial magnetic stimulation preferentially generates I-waves in the corticospinal tract, and that the ‘D- and I-wave hypothesis’ is reliable when the stimulus intensity is near threshold level of the MEPs during voluntary contraction. However, this hypothesis is not reasonable when the stimulus intensity is higher because both D- and I-waves can be elicited at a high stimulus intensity (Burke et al., 19931. Voluntary contraction in the target muscle shortens the onset latency, lowers the threshold, and increases the amplitude of the muscle response (Barker et al., 1986; Hess et al., 1987; Merton et al., 1982; Rothwell et al., 1987; Thompson et al., 1991). Our results demonstrated that the latency shortening can range from 2 ms to 3.5 ms, and that the amplitude during voluntary contraction can be increased to 150-500% of that recorded during the resting state. In the voluntary contraction state, the summation of descending voluntary impulses from cortical areas, afferent impulses from muscle spindles and descending potentials secondary to magnetic brain stimulation can activate the spinal motor neurons earlier than under resting conditions and shorten the MEP onset latency (Claus et al., 1988; Rossini et al., 1987a,b). Our results support this hypothesis because the shortening of the MEP latency was observed without attenuation of the ESCP latency during voluntary contraction. This observation suggests that the shortened latency of the MEPs during voluntary contraction originate in the spinal summation mechanism and not a supraspinal mechanism. This hypothesis was also supported by the results that demonstrated no significant difference between the amplitudes of the evoked muscle response following electric cervical column stimulation compared to those following magnetic cortical stimulation (Maertens de Noordhout et al., 1992). The present study also showed that the mean calculated CMCT of the abductor digiti minimi corresponded best to the measured mean onset latency of the second component of the ESCPs during voluntary contraction. In only one case (case 21, was the calculated CMCT during voluntary contraction similar to the onset latency of the first response (D-wave). These results suggest that magnetic cortical stimulation can produce D-waves but that spinal motor neurons require I-wave stimulation in order to fire in most subjects. Two possible mechanisms may explain why spinal motor neurons were not activated by the D-wave following
136
K. Kaneko et al./Joumal
of the Neurological Sciences 139 (1996) 131-136
magnetic stimulation: (1) The amplitude of the D-wave was too small to recruit spinal motor neurons; and (2) D-waves which were elicited in this manner did not stimulate spinal motor neurons of the hand muscles, but did activate those innervating the muscles of the trunk or lower extremities. Although D-waves were elicited following magnetic stimulation at a high stimulus output, they are most effectively produced by electrical stimulation. (Amassian et al., 1990; Berardelli et al., 1990; Burke et al., 1993). Indeed, the first component (D-waves) had a smaller amplitude than the second or third components (I-waves) in our study. The hypothesis that D-waves, which were elicited in this coil position, did not distribute to spinal motor neurons of the hand muscles may explain this phenomenon. To resolve this hypothesis, we plan to record ESCPs simultaneously from the cervical and thoracic epidural space. In conclusion, the present study supports the hypothesis that transcranial magnetic stimulation preferentially generates I-waves at the threshold stimulus intensity of the MEPs during voluntary contraction, but it generates both D- and I-waves at a high stimulus intensity. The ‘D- and I-wave hypothesis’ is reliable only when the stimulus intensity is near the threshold of the MEPs during voluntary contraction. Our results also demonstrate that the shortened MEPs onset latency can be explained by a temporal summation mechanism associated with voluntary muscle contraction at the spinal level.
References Amassian, V.E., Quirk, G. and Stewart, M.A. (1990) A comparison of corticospinal activation by magnetic coil and electrical stimulation of monkey cortex. Electroenceph. Clin. Neurophysiol., 77: 390-401. Barker, A.T., Freeston, I.L., Jalinous, R. and Jarratt, J. (1986) Clinical evaluation of conduction time measurements in central motor pathways using magnetic stimulation of the human brain. Lance%, i: 132% 1326. Berardelli, A., Inghilleri, M., Cruccu, G. and Manfredi, M. (1990) Descending volley after electrical and magnetic transcranial stimulation in man. Neurosci. Lett., 112: 54-58. Burke, D., Hicks, R., Stephen, J., Woodforth, I. and Crawford, M. (1992) Assessment of corticospinal and somatosensory conduction simultaneously during scoliosis surgery. Electroencephalogr. Clin. Neurophysiol., 83: 388-396. Burke, D., Hicks, R., Gandevia, S.C., Stephen, J., Woodforth, I. and Crawford, M. (1993) Direct comparison of corticospinal volleys in human subjects to transcranial magnetic and electrical stimulation. J. Physiol., 470: 383-393. Claus, D., Mills, K.R. and Murray, N.M.F. (1988) Facilitation of muscle
responses to magnetic brain stimulation by mechanical stimuli in man. Exp. Brain Res., 71: 273-278. Day, B.L., Rothwell, J.C., Thompson, P.D., Dick, J.P.R., Cowman, J.M.A., Berardelli, A. and Marsden, C.D. (1987) Motor cortex stimulation in intact man. I. Multiple descending volleys. Brain, 110: 1191-1209. Day, B.L., Dressier, D., Maertens de Noordhout, A., Marsden, C.D., Nakashima, K., Rothwell, J.C. and Thompson, P.D. (1989) Electric and magnetic stimulation of human motor cortex: surface EMG and single motor unit responses. J. Physiol., 412: 449-473. Fuchigami, Y., Kawai, S., Doi, K., Shiraishi, G., Ito, T., Kaneko, K., Hashida, T., Kawamura, H. and Oofuji, A. (1995) Intraoperative electrodiagnosis for brachial plexus injury. Electroencephalogr. Clin. Neurophysiol. 97; S177. Hess, C.W., Mills, K.R. and Murray, N.M.F. (1987) Responses in small hand muscles from magnetic stimulator of the human brain. J. Physiol., 388: 397-419. Hicks, R., Burke. D., Stephen, J., Woodforth, I. and Crawford, M. (1992al Corticospinal volleys evoked by electrical stimulation of human motor cortex after withdrawal of volatile anesthetics. J. Physiol., 456: 393-404. Hicks, R., Woodforth, I., Crawford, M., Stephen, J. and Burke, D. (1992b) Some effects of isoflurane on I waves of the motor evoked potentials. Br. J. Anaesthiol., 69: 130-136. Maertens de Noordhout, A., Pepin. L., Gerard, P., Delweide, P.J. (1992) Facilitation of responses to motor cortex stimulation: effect of isometric voluntary contraction. Ann. Neural. 32: 365-370. Merton, P.A., Morton, H.B., Hill, D.K. and Marsden, C.D. (1982) Scope of a technique for electrical stimulation of human brain, spinal cord, and muscle. Lancet, ii: 597-600. Patton, H.D. and Amassian, V.E. (1954) Single and multiple unit analysis of cortical stage of pyramidal tract activation. J. Neurophysiol., 17: 345-363. Rossini, P.M., Caramia, M.D. and Zarola, F. (1987a) Mechanisms of nervous propagation along central motor pathways: non-invasive evaluation in healthy subjects and in patients with neurological disease. Neurosurgery, 20: 183- 19 1. Rossini, P.M., Caramia, M.D., Zarola, F. (1987b) Central motor tract propagation in man: studies with non-invasive, unifocal, scalp stimulation. Brain Res., 415: 21 l-225. Rossini, P.M., Barker, A.T., Berardelli, A., Caramia, M.D., Caruso, G., Cracco, R.Q., Dimitrijevic, M.R., Hallet, M., Katayama, Y., Lucking, C.H., Maertens de Noordhout, A.L., Marsden, C.D., Murray, N.M.F, Rothwell, J.C., Swash, M. and Tomberg, C. (1994) Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application, Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 91: 79-92. Rothwell, J.C., Thompson, P.D., Day, B.L., Dick, J.P.R., Kachi, T., Cowan, J.M.A. and Marsden, C.D. (1987) Motor cortical stimulation in intact man. I. General characteristics of EMG responses in different muscles. Brain, 110: 1173-I 190. Tamaki, T., Tsuji, H., Inoue S. and Kobayashi, H. (1981) The prevention of iatrogenic spinal cord injury utilizing the evoked spinal cord potentials. Int. Orthop., 4: 313-3 17 Thompson, P.D., Day, B.L., Rothwell, J.C., Dressier, D., Maertens de Noordhout, A.L. and Marsden, C.D. (1991) Further observations on the facilitation of muscle responses to cortical stimulation by voluntary contraction. Electroenceph. Clin. Neurophysiol., 81: 397-402.