Electroencephalography and cfinical Neurophysiology, 85 (1992) 365-373

365

© 1992 Elsevier Scientific Publishers Ireland, Ltd. 0924-980X/92/$05.00

ELMOCO 92011

Early and late lower limb motor evoked potentials elicited by transcranial magnetic motor cortex stimulation * M.R. Dimitrijevi6, M. Kofler a, W.B. McKay, A.M. Sherwood, C. Van der Linden b and M.A. Lissens c Restorative Neurology and Human Neurobiology, Baylor College of Medic'me, Houston TX 77030-3498 (U.S.A.), ~ University Clinic"for Neurology, A-6020 Innsbruck (Austria), b Department of Neurology, University Hospital, Ghent (Belgium), and c Department of Physical Medicine and Rehabilitation, University HospBals of Lout,ain, Lout,ain (Belgium) (Accepted for publication: 24 August 1992)

Summary

Transcranial magnetic motor cortex stimulation can elicit a series of responses recorded with different latencies from relaxed muscles of the lower limbs. In 7 healthy subjects, ranging in age from 16 to 62 years, stimulation was delivered by a 9 cm coil centered over Cz with the subject in the supine position. Surface polyelectromyography was used to record motor evoked potentials (MEPs) from the quadriceps (QD), hamstrings (HS), tibialis anterior (TA) and triceps surae (TS) muscles bilaterally. Three characteristic responses were identified in each muscle group on the basis of amplitude and latency criteria, identified by latencies: the direct oligosynaptic response MEP30 appeared with a latency of 24.3 msec in the QD, 26.3 msec in the HS, 30.5 msec in the TA and 31.3 msec in the TS; MEP70 with latencies of 64 msec in the QD, 59 msec in the HS, 79 msec in the TA and 72 msec in the TS; MEPt20 with latencies of 115 msec in the QD, 126 msec in the HS, 117 msec in the TA and 124 msec in the TS. These 3 responses have distinct latencies, amplitudes and durations. MEP70 appears to be the result of activation of long descending tracts which end on spinal interneuronal circuits. As MEPt20 has different features, it may have a different mechanism.

Key words: Transcranial magnetic cortical stimulation; Transcranial stimulation; Motor cortex; Motor evoked potentials; MEPs; Electromyography

Motor evoked potentials resulting from either transcranial electric or magnetic motor cortex stimulation may be used to demonstrate the functional integrity and conduction properties of cortico-spinal tracts in healthy human nervous systems as well as those affected by neurological disorders (Merton and Morton 1980; Mills et al. 1987; Rothwell et al. 1987). When this technique is used to measure central conduction time, responses of the fastest cortico-spinal fibers are recorded from the earliest compound motor action potentials (Thompson et al. 1987), single motor units (Calancie et al. 1987; Boniface et al. 1991) or single muscle fiber potentials (Zidar et al. 1987; Zarola et al. 1989). With few exceptions, physiological and clinical studies have focused on these shortest latency responses, neglecting later events, to provide an estimation of conduction velocities in the fastest descending

Correspondence to." Milan R. Dimitrijevi6, M.D., D.Sc., Restorative Neurology and Human Neurobiology, One Baylor Plaza, Suite S-800, Baylor College of Medicine, Houston, TX 77030-3498 (U.S.A.). Tel.: (713) 798-3649; Fax: (713) 798-3683. * Funded by the Vivian L. Smith Foundation for Restorative Neurology, Houston, TX.

spinal tracts. These fastest descending tracts have been shown to connect oligosynaptically to spinal motor cells (Cheney et al. 1985; Porter 1987). There have been reports of longer latency responses being recorded from extensor and flexor carpi radialis muscles of the upper limbs and from tibialis anterior and triceps surae muscles of the lower limbs in healthy subjects (Holmgren et al. 1990). These are distinct from the prolonged central conduction time seen in, for example, subjects with multiple sclerosis (Cowan et al. 1984). In preliminary studies of cortico-spinal tract integrity in chronic spinal cord injury, we found late responses to transcranial stimulation (Dimitrijevi6 et al. 1988). Late responses have also been reported in Wilson's disease (Chu 1990). In addition to the fastest descending cortico-spinal fibers, there exist other indirect descending cortico-spinal or cortico-brain-stem pathways with fibers that terminate on the spinal interneurons in the intermediate zone of the spinal grey matter (Kuypers 1981). These descending pathways to spinal motor cells via polysynaptic networks (Peterson et al. 1979) may mediate MEPs with longer latencies. Also, if transcranial stimuli cover a large area of the motor and surrounding cortices (e.g., frontal and parietal lobes) it is not unrea-

366

sortable to expect that more than one MEP response could be recorded for a single transcranial stimulus. We therefore decided to study the characteristics of later responses following transcranial stimulation in relaxed, healthy, adult subjects. In this report, we describe the electrophysioiogical characteristics of MEPs recorded from lower limb muscles during the first 200 msec following delivery of transcranial magnetic motor cortex stimulation.

Material and methods

With approval of the local Institutional Review Board for Human Research, 7 healthy subjects (1 female, 6 males) ranging in age from 16 to 62 years (mean 32.6 _+ 16.1 years) were examined. Each subject was asked to lie supine on a comfortable examination bed, with care taken to achieve full relaxation of all muscles of the lower limbs. Surface E M G electrode pairs were placed with 3 cm center-to-center spacing, oriented parallel to the long axis of the muscle and centered over the muscle belly of the right and left abductor pollicis brevis, quadriceps (QD), hamstring (HS), tibialis anterior (TA) and triceps surae (TS) muscles. Inter-electrode impedance was reduced to less than 5 k~2. The EMG signals were amplified with a gain of 5000, a filter bandpass of 30-1000 Hz and digitized at 2000 samples/sec with 12-bit resolution and stored for computer analysis. Transcranial magnetic motor cortex stimulation (TCCS) was delivered by a Cadwell MES 10 stimulator with a peak magnetic flux density of 2.2 Tesla and a pulse duration of 80 p, sec, delivered through a 14 turn coil (7.5 cm ID; 9.0 cm OD). The magnitude of the induced field was proportional to the current discharged through the stimulating coil (Baker et al. 1987) and was set as a percentage of the maximum output. The coil was centered over the Cz (international 10-20 E E G electrode placement system) location on the scalp, oriented so that the current flow was clockwise (viewed from above). Stimuli were triggered manually with a minimum interval of 5 sec between stimuli. Responses recorded from the right and left abductor pollicis brevis muscles were used to confirm symmetrical stimulus delivery. After the subject was fully relaxed, 3 stimuli each were delivered at 60%, 80% and 100% intensity. Unless otherwise indicated, the data presented were taken from trials using maximum stimulus strength. An additional recording was performed in one subject to study the time relationship between electromyographic and mechanical events. The subject was prone with a reflective target disk (2.5 cm in diameter) placed on the plantar surface of the foot for the recording of ankle movement using a photoelectric motion sensor (Dimitrijevi6 et al. 1986) in addition to the recording

M.R. D1M1TRIJEVII~ ET AI_

and transcranial stimulation previously described, h another subject, we applied transcranial stimulation it the prone as well as the supine position to examine i the direction of gravity had any effect on MEP: recorded from anterior or posterior compartment mus cles. Analysis of the responses was carried out using computer program that reported latency, duration, are~ and peak-to-peak amplitude after manual selection ol the beginning and ending of each response. Response, were recognized using superimposition of repeated tri. als and were subsequently grouped (first, second ol third response) on the basis of latency. In order tc qualify as a separate response, the trace must have returned to baseline from the previous response ane have a peak amplitude of 10 p.V. In one subject, the hamstring muscles were poorly relaxed and therefore their MEPs were not included in the calculation ol mean values or frequency of appearance for the hamstrings. In another subject, there was excessive stimulu~

RHS

RTA

RTS LQD LHS ~

~

~

LTA

LTS

_

_ ~

~

25 m s

Fig. 1. Superimposed responses from the right quadriceps (RQD), right hamstring (RHS), right tibialis anterior (RTA), right triceps surae (RTS), left quadriceps (LQD), left hamstring (LHS), left tibialis anterior (LTA) and left triceps surae (LTS) muscles are shown. Short-latency MEPs are apparent in all recorded muscles. Additional responses are also identifiable in some but not all muscles. In this and subsequent figures, superimposed traces depict responses to 3 sequential, maximum strength, stimuli unless otherwise indicated.

EARLY AND LATE MEPs

367

artifact in the left quadriceps channdl, leading to its exclusion.

Results In relaxed', healthy volunteers, it was possible to record motor evoked potentials (MEPs) from QD, HS, T A and TS muscles from both lower limbs to repeated transcranial magnetic motor cortex stimulation. In addition to the early, short latency responses, we were able to identify two later responses in all muscles, although not in all trials in all subjects (Fig. 1). For example, later responses were low or absent in QD, but as large or larger than the early responses in HS in this subject. With the exception of the HS, the MEP30 responses were generally larger than the later responses (Table I). We have chosen for the convenience of discussion to call these responses MEP~o, MEP7o and MEP~2o (Fig. 2) to correspond with the nominal latency as they usually appeared (Table I). The shortest latency response, MEP30, occurred with a mean onset latency of 24 msec in the QD, 26 msec in the HS, 30 msec in the T A and 31 msec in the TS (see Table I). This response was also the first to appear as stimulus strength was increased, at strengths as low as 60% in the T A and TS (Fig. 3). MEP30 was present in 100% of stimulus trials in these muscles when stimulus was delivered at maximum intensity (Fig. 3). MEP30 mean peak amplitudes were largest in the T A muscle (773 /xV) with an amplitude in the QD of i52 /zV, in the HS of 123/zV and in the TS of 114/xV. Also, the

TABLE I Latencies (L, in msec), durations (D, in msec) and peak-to-peak amplitudes (A, in #V) are shown for the 3 responses (MEP3o, MEP7o and MEP~e o) to transcranial cortical stimulation at maximum strength. As not all subjects generated all 3 responses to each stimulus, the indicated number of responses in the appropriate time interval (n) was used to calculate the mean values ( ~ ) and standard deviations (or) shown. MEP30 L30

QD cr

n = 39 24.3 24.2 2.5 6.4

HS cr

L70

MEPI20 D70

A70

LI20

152 128

n = 40 64.6 23.6 8.1 7.9

55 111

n = 24 115.3 52.7 28.3 32.0

D(20

AI20 74 62

n = 36 27.5 4.5

1213 101

59.2 3.5

n = 36 34.4 8.6

286 210

n = 30 126.5 56.3 30.1 21.8

92 65

30.5 2.9

n = 42 44.3 14.6

733 527

79.1 9.8

n = 14 22.2 8.5

41 26

n = 22 117.5 45.1 17.1 16.9

101 98

31.3 3.4

n = 42 30.4 6.7

11,1 73

72.2 7.7

n = 34 25.3 12.8

64 69

n = 34 124.2 51.5 29.7 26.6

157 164

TS o-

A30

26.3 4.9

TA cr

MEP70 D30

80~_~

_-

-.

I

100%

lOOms Avg. Onset,

Duration:

[

I MEP30

/

MEPT0

/

MEPt20 ~

Fig. 2. Selected examples of MEP3o, MEP70 and MEPx2 o (shaded areas) from the QD, HS, TA and TS muscles are shown for 80% and 100% stimulus strengths. Examples were taken from different subjects, but both 80% and 100% stimulus strength responses were from the same subject and at the same scale (200/zV). Because the MEP7o response in the QD was so small, it is shown with 5