54
Neuroscience Letters, 112 (1990) 54-58 Elsevier Scientific Publishers Ireland Ltd.
NSL 06802
Descending volley after electrical and magnetic transcranial stimulation in man A. Berardelli, M . Inghilleri, G . C r u c c u a n d M . M a n f r e d i Dipartimento Scienze Neurologiche, Universitlt di Roma 'La Sapienza', Rome (Italy)
(Received 10 October 1989; Revised version received 29 December 1989; Accepted 3 January 1990) Key words: Motor cortex; Stimulation
The descending volley evoked by electrical and magnetic transeranial stimulation was recorded with spinal electrodes in 3 subjects undergoing spinal surgery. The descending volley evoked by electrical stimulation, as previously described, was composed by a short-latency initial wave followed by later waves. In two subjects magnetic stimulation evoked an initial wave of slightly longer latency (0.2-0.3 ms), smaller amplitude and higher threshold than the initial wave evoked by electrical stimulation. In these two subjects, magnetic stimuli probably activated the pyramidal axons directly. In the third subject the initial wave evoked by magnetic stimulation had a latency of 1.4 ms longer and a considerably smaller amplitude than that evoked by electrical stimulation. In this case magnetic stimulation may activate the pyramidal axons indirectly. In the intact m a n muscle responses can be evoked after electrical and magnetic stimulation o f the scalp. The responses after magnetic stimulation have longer latency and larger amplitude than those after electrical stimulation [5]. To explain the difference in amplitude and latency the hypothesis has been put forward that electrical stimulation mainly activates pyramidal axons whereas magnetic stimulation activates the pyramidal neurons presynaptically [4, 5]. In animal experiments the descending volley evoked by stimulation o f exposed motor cortex has been recorded from the cervical spinal cord and it is c o m p o s e d o f an initial wave (D) followed by later waves (I) [2, 9, 10]. With epidural and spinal recordings, Boyd et al. [3] and Inghilleri et at. [7, 8] have demonstrated that also in h u m a n s it is possible to record the descending volley evoked by electrical stimulation o f the m o t o r cortex. In this paper we have studied the descending volley after magnetic stimulation and c o m p a r e d it with that evoked by electrical stimulation. The study was performed in three patients (33, 45 and 62 years old) with tumors in the thoracic spinal cord, undergoing resection o f the tumors. The dura mater was open and the spinal cord exposed at cervical and upper thoracic level. All the subjects Correspondence." A. Berardelli, Dipartimento Scienze Neurologiche, Viale Universith 30, 00185 Rome,
Italy. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
55 were curarized (pancuronium 0.04 mg/kg) and anesthetized (70% nitrous oxide, fenthanyl 1-3 #g/kg/h). The surgical procedure did not allow us to make recordings at different levels of anaesthesia. Patients gave informed consent to the study, which was approved by the hospital ethics committee. Electrical stimulation of the scalp was performed with a Digitimer D 180 model, which delivered stimuli up to 750 V (100% output) at 50 #s time constant. The anode was placed on the vertex and the cathode 7 cm anterior to it. Magnetic stimulation was performed with a Novametrix Model 200 with a 10 cm diameter flat coil, generating a magnetic field of up to 1.5 Tesla (100% output) placed on the vertex. Magnetic stimulation was performed with 80-100% of the output. Signals were recorded bipolarly (interelectrode distance 2.5 cm) by means of Sigma Medtronic 3483 epidural electrodes placed on the spinal cord between the dura and the lateral surface of the cord, rostral to the lesion in the spinal cord. The proximal electrode was connected to the inverting input of the amplifier (negativity upward). Signals were amplified (2 Hz to 5 kHz), averaged (16 trials) and stored on a Basis (OTE) signal analyzer. The latency was measured at the negative peak and the amplitude peak to peak [3, 8]. To calculate conduction velocity, recordings were performed at two different sites along the spinal cord, as previously described [3, 7, 8]. Electrical and magnetic stimulation was performed in all the subjects before operation. The responses were normal in the upper and absent in the lower limbs. In two subjects (A, B), (Fig. 1) electrical stimulation of the scalp evoked an initial wave, at low stimulus intensity (30-40% of the output) which plateaued at 70-80% of the output. The latency depended on the position of the recording electrode and was 2.2 ms in patient A (C5-C6 level) and 1.2 ms in patient B (CrC2 level). The maximum amplitude was 9 #V in patient A and 57 #V in patient B. Magnetic stimulation of the scalp also evoked an initial wave but only with high stimulation intensity (80%). In patient A the latency was 2.5 ms and the amplitude 2.5 #V, in B 1.4 ms and 30 #V using maximum output of the stimulator. In subject A, later waves followed the initial wave after both types of stimulation. In subject C, electrical stimulation evoked an initial wave with a latency of 2.4 m s ( C 6 - C 7 level) and an amplitude of 47 #V; the first wave evoked by magnetic stimulation using maximum output of the stimulator, had a latency of 3.8 ms and an amplitude of 8 #V. In this subject we calculated the conduction velocity of the first waves after electrical and magnetic stimulation and a conduction velocity of 62.2 ms/s was found for both types of stimulation (Fig. 1). In the present paper we have confirmed that electrical stimuli to the scalp evoke a descending volley composed by early and later waves. On the basis of the conduction velocity and recovery cycle, we previously observed that in the intact man the early wave of the descending volley after transcranial electrical stimulation is due to activation of the pyramidal axons and that it is similar to the D wave recorded in animal experiments. We have also measured the conduction velocity of the later waves and found that they have a conduction velocity similar to that of the initial wave [8]. We now show that a descending volley composed by early and later waves can also be evoked by magnetic stimulation. In subject A and B the latency of the initial mag-
56
A
;
v
B
/
C
Fig. 1. Descending volley after electrical (on the left) and magnetic (on the right) stimulation in three subjects (A-C). In subject A and B the descending volley has been recorded at one site. Initial and later waves are present in subject A and only initial waves in subject B. In subject C the descending volley has been recorded at two different sites over the cervical spinal cord. The conduction velocity of the first recorded wave is 62 m/s after both electrical and magnetic stimulation (distal recording bottom trace). Each trace is the average of 16 trials. Horizontal calibration is 2 ms for subject A and C and 1 ms for subject B. Vertical calibration is 1.8/~V for subject A, 30/~V for subject B and 14/tV for subject C.
netic wave is slightly longer (0.2-0.3 ms) than that after electrical stimulation, and the amplitude is considerably (by 30--50%) smaller and the threshold higher. In these two subjects, the site o f activation o f magnetic stimuli seems to be similar to that o f electrical stimuli, i.e. at the level o f the pyramidal axon [8]. The reason for the small amplitude and latency delay o f the initial magnetic wave m a y be explained by the excitation o f pyramidal axons at a slightly more proximal site than that reached by electrical stimulation, owing to a smaller spread o f the current field induced by magnetic stimulation. This is supported by the finding o f Boyd et al. [3] that the amplitude o f the initial wave increased and the latency decreased by 0.3 ms on average over the full range o f intensity. O u r previous findings [8] also support this view. The smaller size o f the initial wave evoked in subjects A and B by magnetic stimulation as c o m p a r e d to that evoked by electrical stimulation, implies that fewer axons were excited by the magnetic stimulus. The electric field induced by magnetic shocks was p r o b a b l y weaker since the amplitude o f the electrically evoked D waves did reach a plateau, whereas magnetic D waves did not, even at m a x i m u m o u t p u t o f the stimu-
57
lator. It is likely that the descending volley is mainly produced by the corticomotoneuronal component of the corticospinal tract, but other fibres can contribute to it. The different amplitude could be due to magnetic stimulation activating fewer corticomotoneuronal axons, or electrical stimulation also activating fibres other than corticomotoneuronal. If fewer corticomotoneuron axons are activated, the synaptic times at the lower motoneurones are probably longer because of a reduced spatial summation and the need for temporal summation by later descending volleys. This would explain the longer latency of muscle responses to magnetic stimulation than electrical stimulation. In case C, the first wave evoked by magnetic stimulation has a latency 1.4 ms longer and a considerably smaller amplitude than that evoked by electrical stimulation. The difference in latency between the initial wave evoked by electrical and magnetic stimulaton in this subject could be explained by at least 3 mechanisms (1) excitation of axons with slower conduction velocity (2) activation of axons at a more proximal site (3) excitation of pyramidal dendrites or cortical interneurones and activation of the output elements transsynaptically. The first hypothesis can be excluded by the observation that in patient C, the conduction velocity is similar after electrical and magnetic stimulation. The second hypothesis seems unlikely since a decrease in latency of 1.4 ms in fibres conducting at 60 m/s implies that the electrical stimulus would spread 9 cm deeper into the brain than the magnetic. The third hypothesis that the first wave recorded after magnetic stimulation is due to transsynaptic rather than to direct excitation of the pyramidal axons is most likely. The difference in latency between the first wave evoked by electrical and magnetic stimulation is similar to the latency difference between the initial wave and the first of the later waves measured in animal and human experiments [2, 8-10]. In conclusion, we have for the first time shown that in man magnetic stimulation is capable of activating the pyramidal axons directly but less effectively than electrical stimulation. With a high stimulus intensity the tendency of a stimulus to reach into the depth of the brain increases, making possible both a transsynaptic activation (patient C) and a direct excitation of the axons (patient A and B). Studies in man [6] and in monkeys [1] have suggested that at high stimulus intensities magnetic stimulation may produce both types of activation. We cannot comment on the number and the size of the later waves because it is known that transsynaptic excitation suffers from anesthesia [2]. 1 Amassian, V.E., Quirk, G.J. and Stewart, M., Magnetic coil versus electrical stimulation of monkey motor cortex, J. Physiol. (Lond.), 394 (1987) 119P. 2 Amassian, V.E., Stewart, M., Quirk, G.J. and Rosenthal, J.L., Physiological basis of motor effects of a transient stimulus to cerebral cortex, Neurosurgery, 20 (1987) 74-93. 3 Boyd, S.G., Rothwell, J.C., Cowan, J.M.A., Webb, P.J., Morley, T., Asselman, P. and Marsden, C.D., A method of monitoring function in corticospinal pathways during scoliosis surgery with a note on motor conduction, J. Neurol. Neurosurg. Psychiatry, 49 (1986) 251-257. 4 Day, B.L., Thompson, P.D., Dick, J.P., Nakashima, K. and Marsden, C.D., Different sites of action of electrical and magnetic stimulation of the human brain, Neurosci. Lett., 75 (1987) 101-106.
58 5 Hess, Ch.W., Mills, K.R. and Murray, N.M.F., Responses in small hand muscles from magnetic stimulation of the human brain, J. Physiol. (Lond.), 388 (1987) 397-419. 6 Hess, Ch.W. and Ludin, H.P., Transcranial brain stimulation by magnetic pulses. Methodological and physiological considerations, EEG- EMG, 19 (I 988) 209-215. 7 Inghilleri, M., Berardelli, A., Cioni, B., Cruccu, G., Meglio, M. and Manfredi, M,, The conduction velocity of the corticospinal tract in man. In A.R. Liss (Ed.), Non-invasive Stimulation of Brain and Spinal Cord: Fundamentals and Clinical Applications, 1988, pp. 123-130. 8 Inghilleri, M., Berardelli, A., Cruccu, G., Priori, A. and Manfredi, M., Corticospinal potentials after transcranial stimulation in the intact man, J. Neurol. Neurosurg. Psychiatry, 52 (1989) 970--974. 9 Kernell, D. and Chien-Ping, W.V., Responses of pyramidal tract to stimulation of the baboon's motor cortex, J. Physiol. (Lond.), 191 (1967) 653-672. 10 Patton, H.D. and Amassian, V.E., Single and multiple unit analysis of cortical stage of pyramidal tract activation, J. Neurophysiol., 17 (1954) 345 363.
Neuroscience Letters, 112 (1990) 54-58 Elsevier Scientific Publishers Ireland Ltd.
NSL 06802
Descending volley after electrical and magnetic transcranial stimulation in man A. Berardelli, M . Inghilleri, G . C r u c c u a n d M . M a n f r e d i Dipartimento Scienze Neurologiche, Universitlt di Roma 'La Sapienza', Rome (Italy)
(Received 10 October 1989; Revised version received 29 December 1989; Accepted 3 January 1990) Key words: Motor cortex; Stimulation
The descending volley evoked by electrical and magnetic transeranial stimulation was recorded with spinal electrodes in 3 subjects undergoing spinal surgery. The descending volley evoked by electrical stimulation, as previously described, was composed by a short-latency initial wave followed by later waves. In two subjects magnetic stimulation evoked an initial wave of slightly longer latency (0.2-0.3 ms), smaller amplitude and higher threshold than the initial wave evoked by electrical stimulation. In these two subjects, magnetic stimuli probably activated the pyramidal axons directly. In the third subject the initial wave evoked by magnetic stimulation had a latency of 1.4 ms longer and a considerably smaller amplitude than that evoked by electrical stimulation. In this case magnetic stimulation may activate the pyramidal axons indirectly. In the intact m a n muscle responses can be evoked after electrical and magnetic stimulation o f the scalp. The responses after magnetic stimulation have longer latency and larger amplitude than those after electrical stimulation [5]. To explain the difference in amplitude and latency the hypothesis has been put forward that electrical stimulation mainly activates pyramidal axons whereas magnetic stimulation activates the pyramidal neurons presynaptically [4, 5]. In animal experiments the descending volley evoked by stimulation o f exposed motor cortex has been recorded from the cervical spinal cord and it is c o m p o s e d o f an initial wave (D) followed by later waves (I) [2, 9, 10]. With epidural and spinal recordings, Boyd et al. [3] and Inghilleri et at. [7, 8] have demonstrated that also in h u m a n s it is possible to record the descending volley evoked by electrical stimulation o f the m o t o r cortex. In this paper we have studied the descending volley after magnetic stimulation and c o m p a r e d it with that evoked by electrical stimulation. The study was performed in three patients (33, 45 and 62 years old) with tumors in the thoracic spinal cord, undergoing resection o f the tumors. The dura mater was open and the spinal cord exposed at cervical and upper thoracic level. All the subjects Correspondence." A. Berardelli, Dipartimento Scienze Neurologiche, Viale Universith 30, 00185 Rome,
Italy. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
55 were curarized (pancuronium 0.04 mg/kg) and anesthetized (70% nitrous oxide, fenthanyl 1-3 #g/kg/h). The surgical procedure did not allow us to make recordings at different levels of anaesthesia. Patients gave informed consent to the study, which was approved by the hospital ethics committee. Electrical stimulation of the scalp was performed with a Digitimer D 180 model, which delivered stimuli up to 750 V (100% output) at 50 #s time constant. The anode was placed on the vertex and the cathode 7 cm anterior to it. Magnetic stimulation was performed with a Novametrix Model 200 with a 10 cm diameter flat coil, generating a magnetic field of up to 1.5 Tesla (100% output) placed on the vertex. Magnetic stimulation was performed with 80-100% of the output. Signals were recorded bipolarly (interelectrode distance 2.5 cm) by means of Sigma Medtronic 3483 epidural electrodes placed on the spinal cord between the dura and the lateral surface of the cord, rostral to the lesion in the spinal cord. The proximal electrode was connected to the inverting input of the amplifier (negativity upward). Signals were amplified (2 Hz to 5 kHz), averaged (16 trials) and stored on a Basis (OTE) signal analyzer. The latency was measured at the negative peak and the amplitude peak to peak [3, 8]. To calculate conduction velocity, recordings were performed at two different sites along the spinal cord, as previously described [3, 7, 8]. Electrical and magnetic stimulation was performed in all the subjects before operation. The responses were normal in the upper and absent in the lower limbs. In two subjects (A, B), (Fig. 1) electrical stimulation of the scalp evoked an initial wave, at low stimulus intensity (30-40% of the output) which plateaued at 70-80% of the output. The latency depended on the position of the recording electrode and was 2.2 ms in patient A (C5-C6 level) and 1.2 ms in patient B (CrC2 level). The maximum amplitude was 9 #V in patient A and 57 #V in patient B. Magnetic stimulation of the scalp also evoked an initial wave but only with high stimulation intensity (80%). In patient A the latency was 2.5 ms and the amplitude 2.5 #V, in B 1.4 ms and 30 #V using maximum output of the stimulator. In subject A, later waves followed the initial wave after both types of stimulation. In subject C, electrical stimulation evoked an initial wave with a latency of 2.4 m s ( C 6 - C 7 level) and an amplitude of 47 #V; the first wave evoked by magnetic stimulation using maximum output of the stimulator, had a latency of 3.8 ms and an amplitude of 8 #V. In this subject we calculated the conduction velocity of the first waves after electrical and magnetic stimulation and a conduction velocity of 62.2 ms/s was found for both types of stimulation (Fig. 1). In the present paper we have confirmed that electrical stimuli to the scalp evoke a descending volley composed by early and later waves. On the basis of the conduction velocity and recovery cycle, we previously observed that in the intact man the early wave of the descending volley after transcranial electrical stimulation is due to activation of the pyramidal axons and that it is similar to the D wave recorded in animal experiments. We have also measured the conduction velocity of the later waves and found that they have a conduction velocity similar to that of the initial wave [8]. We now show that a descending volley composed by early and later waves can also be evoked by magnetic stimulation. In subject A and B the latency of the initial mag-
56
A
;
v
B
/
C
Fig. 1. Descending volley after electrical (on the left) and magnetic (on the right) stimulation in three subjects (A-C). In subject A and B the descending volley has been recorded at one site. Initial and later waves are present in subject A and only initial waves in subject B. In subject C the descending volley has been recorded at two different sites over the cervical spinal cord. The conduction velocity of the first recorded wave is 62 m/s after both electrical and magnetic stimulation (distal recording bottom trace). Each trace is the average of 16 trials. Horizontal calibration is 2 ms for subject A and C and 1 ms for subject B. Vertical calibration is 1.8/~V for subject A, 30/~V for subject B and 14/tV for subject C.
netic wave is slightly longer (0.2-0.3 ms) than that after electrical stimulation, and the amplitude is considerably (by 30--50%) smaller and the threshold higher. In these two subjects, the site o f activation o f magnetic stimuli seems to be similar to that o f electrical stimuli, i.e. at the level o f the pyramidal axon [8]. The reason for the small amplitude and latency delay o f the initial magnetic wave m a y be explained by the excitation o f pyramidal axons at a slightly more proximal site than that reached by electrical stimulation, owing to a smaller spread o f the current field induced by magnetic stimulation. This is supported by the finding o f Boyd et al. [3] that the amplitude o f the initial wave increased and the latency decreased by 0.3 ms on average over the full range o f intensity. O u r previous findings [8] also support this view. The smaller size o f the initial wave evoked in subjects A and B by magnetic stimulation as c o m p a r e d to that evoked by electrical stimulation, implies that fewer axons were excited by the magnetic stimulus. The electric field induced by magnetic shocks was p r o b a b l y weaker since the amplitude o f the electrically evoked D waves did reach a plateau, whereas magnetic D waves did not, even at m a x i m u m o u t p u t o f the stimu-
57
lator. It is likely that the descending volley is mainly produced by the corticomotoneuronal component of the corticospinal tract, but other fibres can contribute to it. The different amplitude could be due to magnetic stimulation activating fewer corticomotoneuronal axons, or electrical stimulation also activating fibres other than corticomotoneuronal. If fewer corticomotoneuron axons are activated, the synaptic times at the lower motoneurones are probably longer because of a reduced spatial summation and the need for temporal summation by later descending volleys. This would explain the longer latency of muscle responses to magnetic stimulation than electrical stimulation. In case C, the first wave evoked by magnetic stimulation has a latency 1.4 ms longer and a considerably smaller amplitude than that evoked by electrical stimulation. The difference in latency between the initial wave evoked by electrical and magnetic stimulaton in this subject could be explained by at least 3 mechanisms (1) excitation of axons with slower conduction velocity (2) activation of axons at a more proximal site (3) excitation of pyramidal dendrites or cortical interneurones and activation of the output elements transsynaptically. The first hypothesis can be excluded by the observation that in patient C, the conduction velocity is similar after electrical and magnetic stimulation. The second hypothesis seems unlikely since a decrease in latency of 1.4 ms in fibres conducting at 60 m/s implies that the electrical stimulus would spread 9 cm deeper into the brain than the magnetic. The third hypothesis that the first wave recorded after magnetic stimulation is due to transsynaptic rather than to direct excitation of the pyramidal axons is most likely. The difference in latency between the first wave evoked by electrical and magnetic stimulation is similar to the latency difference between the initial wave and the first of the later waves measured in animal and human experiments [2, 8-10]. In conclusion, we have for the first time shown that in man magnetic stimulation is capable of activating the pyramidal axons directly but less effectively than electrical stimulation. With a high stimulus intensity the tendency of a stimulus to reach into the depth of the brain increases, making possible both a transsynaptic activation (patient C) and a direct excitation of the axons (patient A and B). Studies in man [6] and in monkeys [1] have suggested that at high stimulus intensities magnetic stimulation may produce both types of activation. We cannot comment on the number and the size of the later waves because it is known that transsynaptic excitation suffers from anesthesia [2]. 1 Amassian, V.E., Quirk, G.J. and Stewart, M., Magnetic coil versus electrical stimulation of monkey motor cortex, J. Physiol. (Lond.), 394 (1987) 119P. 2 Amassian, V.E., Stewart, M., Quirk, G.J. and Rosenthal, J.L., Physiological basis of motor effects of a transient stimulus to cerebral cortex, Neurosurgery, 20 (1987) 74-93. 3 Boyd, S.G., Rothwell, J.C., Cowan, J.M.A., Webb, P.J., Morley, T., Asselman, P. and Marsden, C.D., A method of monitoring function in corticospinal pathways during scoliosis surgery with a note on motor conduction, J. Neurol. Neurosurg. Psychiatry, 49 (1986) 251-257. 4 Day, B.L., Thompson, P.D., Dick, J.P., Nakashima, K. and Marsden, C.D., Different sites of action of electrical and magnetic stimulation of the human brain, Neurosci. Lett., 75 (1987) 101-106.
58 5 Hess, Ch.W., Mills, K.R. and Murray, N.M.F., Responses in small hand muscles from magnetic stimulation of the human brain, J. Physiol. (Lond.), 388 (1987) 397-419. 6 Hess, Ch.W. and Ludin, H.P., Transcranial brain stimulation by magnetic pulses. Methodological and physiological considerations, EEG- EMG, 19 (I 988) 209-215. 7 Inghilleri, M., Berardelli, A., Cioni, B., Cruccu, G., Meglio, M. and Manfredi, M,, The conduction velocity of the corticospinal tract in man. In A.R. Liss (Ed.), Non-invasive Stimulation of Brain and Spinal Cord: Fundamentals and Clinical Applications, 1988, pp. 123-130. 8 Inghilleri, M., Berardelli, A., Cruccu, G., Priori, A. and Manfredi, M., Corticospinal potentials after transcranial stimulation in the intact man, J. Neurol. Neurosurg. Psychiatry, 52 (1989) 970--974. 9 Kernell, D. and Chien-Ping, W.V., Responses of pyramidal tract to stimulation of the baboon's motor cortex, J. Physiol. (Lond.), 191 (1967) 653-672. 10 Patton, H.D. and Amassian, V.E., Single and multiple unit analysis of cortical stage of pyramidal tract activation, J. Neurophysiol., 17 (1954) 345 363.
