вход по аккаунту


Corticospinal tract conduction time in multiple sclerosis.

код для вставкиСкачать
Corticospinal Tract
Conduction Time in Multiple Sclerosis
Kerry R. Mills, MRCP, and Nicholas M. F. Murray, MRCP
Anodal shocks of 400 to 700 V from a low-output impedance stimulator applied percutaneously over the motor cortex
evoke muscle action potentials in partially voluntarily activated contralateral muscles. Cathodal shocks from the same
device applied to the cervical spinal cord produce maximal ipsilateral muscle action potentials in a relaxed limb. This
technique was used to study the central motor pathway in 15 healthy subjects and 8 patients with clinically definite
multiple sclerosis. As stimuli were applied in the axilla, over the C7 vertebral level, and over the arm area of the motor
cortex, recordings were made of muscle action potentials of forearm flexor muscles. In controls, cord-to-axilla conduction time was 4.1 T 0.61 ms, and cortex-to-cord time was 4.4 2 0.75 ms. In patients, cord-to-axilla conduction times
were normal, while central conduction times were either markedly prolonged (6.4 to 31 ms) or absent. This technique
is a potentially powerful tool for the investigation of central motor pathways in healthy subjects and patients with
neurological disease.
Mills KR, Murray NMF: Corticospinal tract conduction time in multiple sclerosis.
Ann Neurol 18:601-605, 1985
Direct investigation of conduction in central motor
pathways has not hitherto been possible in conscious
human beings. A technique that allows percutaneous
electrical stimulation of the brain and spinal cord has
now been described 112-171, enabling estimates to be
made of conduction times in the pyramidal tract.
The early work on animals {91 and on exposed human brain 1181 has established that the cortex has a
relatively high threshold to electrical stimulation. It is
therefore not surprising that initial attempts to stimulate the brain by passing currents through the intact
scalp failed because the procedure was too painful {9].
The current must penetrate the scalp, skull, and dura
mater and then achieve a sufficient density at the cortex to activate the neural elements.
Merton and Morton 116, 177 discovered that by using a high-voltage capacitative discharge through a lowoutput impedance circuit, current density at the cortex
could be raised above the threshold for excitation.
They described how stimulation of the posterior scalp
area produced phosphenes in the appropriate visual
field and how stimulation over the motor area produced twitching on the contralateral side of the body.
Shocks were large (2,000 V), and although tolerated
by healthy subjects, probably could not be used in
patients with multiple sclerosis (MS).
It has long been known that cortical thresholds are
less for anodal shocks than for cathodal ones 16, 71. It
was not until cortical stimulation could be applied to
conscious human subjects, however, that the seminal
observation was made that cortical thresholds were
dramatically reduced if the subject made a small voluntary contraction of the muscle of interest 1111. The
procedure then required only 400 to 700 V and could
be applied to both healthy subjects and patients with
neurological disease. The site of cortical stimulation is
difficult to control, but by selectively lowering its cortical threshold, a particular muscle group can be studied
in isolation.
In 1982, Marsden and co-workers 1131 also discovered, apparently by accident, that shocks applied with
the low-output impedance stimulator over the spinal
cord could also evoke muscle responses. This finding
immediately led to a technique to measure conduction
time in central motor fibers by estimating the latency
difference between muscle responses from the cortex
and the spinal cord. This technique has been used to
demonstrate the normal connection between cortex
and muscle in severe Parkinson’s disease [47 and to
show slowing of conduction in central motor fibers in
MS C5, 191.
We report here the initial results of measurements
of central conduction times in healthy subjects and in
patients with clinically definite MS.
From the Department of Clinical Neurophysiology, National Hospital for Nervous Diseases, Queen Square, London W C l N 3BG,
Received Jan 23, 1985, and in revised form Mar 20. Accepted for
publication Apr 5 , 1985.
Address reprint requests to Dr Mills.
Patients and Methods
Patients and Controls
The healthy control group consisted of 15 laboratory and
medical personnel who gave their informed consent to the
procedure, which had local ethical committee approval.
60 1
10 20 30 40 50 60 70 80 90 100
There were 12 males (age range, 19 to 37 years) and 3
females (age range, 19 to 34 years). None had any history or
evidence of neurological disease. Patients or controls with a
history of convulsive seizures or cardiac disease were
specifically excluded from the study.
The 8 patients, all with clinically definite MS, consisted of
3 males (age range, 25 to 62 years) and 5 females (age range,
17 to 37 years). All had severe pyramidal weakness and
spasticity in the legs. All patients had hyperreflexia but only
mild weakness in the arms. In addition, 4 patients had cerebellar ataxia of the arms. Visual evoked responses to patternreversal stimuli were delayed in all patients, and all patients
had oligoclonal band patterns demonstrable in the cerebrospinal fluid. All patients gave informed consent to the procedure.
Compound muscle action potentials were recorded from the
forearm flexor muscles in response to stimulation of the
motor pathway at three sites: the peripheral nerve in the
axilla, the spinal cord at the C7 vertebral level, and the arm
area of the contralateral motor cortex.
In all controls and in 7 of the 8 patients, both sides of the
body were studied. Recordings were made with surface electrodes of the saddle type with an interelectrode distance of
3.5 cm. Responses were amplified (the band-pass of the
amplifier was 20 Hz to 2 kHt) and displayed on a conventional electromyograph machine (Medelec model MS8),
which was also interfaced to a microcomputer (Research Machines, model 3802) to allow processing, storage, and plotting of the responses. Stimuli giving maximal responses were
applied to the axilla, and the position of the recording electrodes was adjusted to give as simple a waveform as possible
with a sharp take-off. Stimuli were delivered either with a
conventional stimulator (Medelec model MS8) or with a specially constructed prototype stimulator; results were identical
in each case.
Initial studies of responses evoked by cord and cortical
stimulation were carried out with the prototype stimulator;
later, a D180 isolated stimulator (Digitimer Ltd, Welwyn
Garden City, Hertfordshire, England) was used. Stimuli
from this low-output impedance stimulator were applied to
the spinal cord by placing the cathode between the spines of
the C6 and C7 vertebrae with the anode 6 cm lateral to the
602 Annals of Neurology
Axi lla
10 20 3 0 40 50 60 7 0 80 90 100
F ig 1. Muscle action potentials recorded from forearmjexor muscles of a healthy subject (left) and a patient with multiple sclerosis (right) in response to stimulation in the axilh, spinal cord,
and motor cortex. In the patient, the onset latency of the response
to cortical stimulation is clearb delayed (25 ms) compared with
that of the healthy subject (12.2 ms).
cathode. Stimulating electrodes were oval silver strips covered in saline-soaked lint with an approximate contact area
with the skin of 1 cm2. Subjects were tested on an examination couch, sitting at an angle of about 45 degrees with the
legs extended. Voltage was increased until a response of
maximal amplitude was obtained.
The anode for cortical stimulation was placed over the
motor area for the arm. This was estimated as a point 7 cm
down the interaural line from the vertex and 1 cm anterior to
that line. The cathode was placed 6 cm anterior to rhe anode.
Subjects on the examination couch were instructed to relax
as fully as possible. They were then requested to perform a
moderate wrist flexion by lifting a 1-kg weight while the
cortical shock was given. N o more than five shocks were
given on any one side. Voltage was increased until the latency of the response did not change; although maximal muscle responses can be obtained [12], this was not attempted in
the present study.
In all subjects, estimates of conduction distances were
made by means of a flexible tape, but as it is well recognized
that this measurement is inaccurate, we decided to quote
conduction times rather than attempt to calculate conduction
Typical muscle action potentials evoked in the forearm
flexor muscles of a healthy subject by stimulation at
the axilla, spinal cord, and motor cortex are seen in
Figure 1. Allowing for a small degree of temporal dispersion between the cord and the axilla, it can be seen
that the muscle action potentials from these two sites
are very similar in waveform and amplitude. The muscle action potential evoked by motor cortex stimulation is an order of magnitude smaller, but then no
Vol 18 N o 5 November 1985
St 1m.
Ax1 11a
St in. Cord (C7)
5 t h Cortex
10 20 30 40 55 60 70 80 90 100
Stinr. Cortex
Fig 2. Muscle action potentials recorded from forearm $exor muscles of a patient w i t h multiple sclerosis in response to stimulation
of the motor patbway in the axiiia, spinal cord, and motor cortex. The responses to cortical stimulation (three are superimposed)
are delayed and of small amplitude.
attempt was made to make this response maximal in
amplitude. In all healthy subjects it was easy to obtain
similar responses from the axilla and the cord, as well
as a response from the cortex that had a reproducible
onset latency.
In healthy controls, the mean conduction time ( ? 1
SD) from the C7 level of the spinal cord to the axilla
was 4.1 ? 0.61 ms and from the motor cortex to the
C7 level of the cord was 4.4 & 0.75 ms.
In the patients with MS, it was also easy to obtain
responses from stimulation of the axilla and the cord
that were similar in waveform and amplitude (Figs 13). Responses from stimulation of the cortex, however, were either not obtainable (Fig 4) (both sides in
one patient aod one side in another), or were small,
dispersed, and delayed (Figs 1-3). Stimuli were of the
same magnitude or larger in the patients in whom no
responses could be found.
The mean conduction time ( & 1 SD) between the
cord and axilla in patients was 4.2 -+ 0.7 ms; this value
is not significantly different @ > 0.5) from that in
healthy subjects. Conduction times from motor cortex
to cord in the patients varied from 6.4 to 31.1 ms. All
values were well outside the mean ( ? 2 SD) in healthy
subjects (Fig 5). N o correlation was found between the
degree of central motor slowing and the degree of
pyramidal motor involvement as judged on clinical
All study participants tolerated the procedure well,
and all controls and 7 of the 8 patients consented to
the second side being studied after completion of the
study on one side. The whole procedure took about 30
minutes to complete on each person. All subjects were
asked to report on the discomfort produced; approximately half regarded cord stimulation as equally or
10 20 30 40 50 60 70 00 90 100
F ig 3 . Muscle action potentials recorded from forearm$exor muscles of another patient w i t h multiple sclerosis in response to
siimulation of the motor pathway in the axiila, spinal cord, and
motor cortex. The responses to cortical stimulation (two are
superimposed) are dekzyed and of small a m p l i t d e .
more comfortable than peripheral nerve stimulation.
All subjects regarded cortical stimulation as more uncomfortable than either cord or peripheral nerve
stimulation, but in no case was it reported as severe or
intolerable. Much of the discomfort from cortical
stimulation derives from the concurrent contraction of
the temporalis muscle, which is often partially activated in anticipation of the shock; it is for this reason
that subjects were encouraged to relax as much as possible and to concentrate only on producing the wrist
flexion prior to the shock.
Whenever new techniques are introduced, the question of their safety must be borne in mind. The
stimulator produces a large instantaneous flow of current (about l amp) with a time constant of decay of 50
or 100 ps. It is clearly important that the stimulator
remain isolated from ground and that current flow,
other than that between its terminals, is minimized.
For this reason, the terminals of the stimulator must
never be connected to any other piece of grounded
apparatus (e.g., a load resistor to measure the current
flow). Grounding of the recording side of the system is
done with a plate near the recording site. Cortical and
spinal cord stimulation have now been applied to a
considerable number of healthy volunteers [ 4 , 5 , 12171 including our own 15 controls. No untoward side
effects have so far been reported. One of our healthy
subjects developed a migraine headache for 3 to 4
hours after the procedure, but this may have been
fortuitous. The amount of energy entering the brain
during a cortical shock is about 0.02 of that occurring
Mills and Murray: Corticospinal Conduction in Multiple Sclerosis 603
Stir Axilla
- / ,
Stim. Cortex
10 20 30 40 50 60 70 80 90 100
10 20 30 40 50 60 70 80 90 100
604 Annals of Neurology Vol 18 No 5
during electroconvulsive therapy, but in the latter, repetitive shocks are given to provoke a seizure. It has
been shown that the electroencephalograms of patients
receiving single cortical shocks are unchanged by the
procedure [l I}. Evidence from animal experiments
suggests that repetitive shocks would be required to
provoke a seizure El].
The possibility that electric shock might predispose
to exacerbation of MS was considered. A recent prospective and retrospective study 121 did not implicate
accidental electric shock as an etiological factor, and
although exacerbation within 6 months of electric
shock occurred more frequently than expected, this
was not significant at the p = 0.05 level. Electric shock
was not associated with more rapid deterioration than
expected: of 23 patients, 5 followed the predicted
course, 10 showed more rapid deterioration over 6
months than expected, and in 8, progression was less
rapid than predicted {2). We believed that the evidence against electric shock was insufficient to preclude its use in MS and that a pilot study was justifiable
in light of the technique's possible advantages as a new
diagnostic tool.
The questions as to what neural elements are excited
and what pathways are involved must be addressed. In
healthy controls, the average conduction time of 4.4
ms between the cortex and the cord indicates conduction in large myelinated fibers, which would be consistent with conduction in the pyramidal tract. However,
we cannot rule out other routes such as the corticalcerebellar-spinal pathway, but it seems unlikely that
there are any interposed synapses. Whether cortical
cell bodies, dendrites, or axons are excited is still uncertain, but there is evidence that surface anodal stimulation produces depolarization of pyramidal cell axons
and hyperpolarization of apical-cell dendrites and horizontally running axons in the outer layers of the cortex
Fig 4. Muscle action potentials recorded fromfEexor muscles of the
lt$t and right forean of a third patient with multiple sclerosis
in response to stimulation of the motor pathway in the axilla,
spinal cord, and motor cortex. No consistent responses to cortical
stimulation are seen on either side, yet stimuli were larger than
those normally used t o evoke responses in healthy subjects.
Spinal Cord ( C 7 ) to Ax1110
Conduction t i m e (ma)
Cortex-Spinal Cord (C7)
Conduction T i m e (me)
25 .
Fig 5 . Conduction times from spinal cord t o axilla (left)and
from motor cortex t o spinal cord (right) in healthy controh and
in patients with multiple sclerosis (MS). Horizontal lines indicate the mean value 2 standard deviations.
November 1985
P O ) . This would tend to focus the effects of anodal
stimulation on corticofugal fibers with direct excitation
of the impulse-generating regions of pyramidal cells.
The striking enhancement of the response to cortical
stimulation that occurs during voluntary activation of
the muscle suggests that the excitability of conical cells
is raised and that the shock excites the cell body or the
first node of its descending axon. There is recent evidence, however, that facilitation produced by voluntary activation may take place, at least partly, in the
spinal cord 131, despite the absence of enhancement
with voluntary contraction during spinal cord stimulation.
It is probable that stimulation over the cervical cord
excites either anterior horn-cell bodies or their axons
within or immediately adjacent to the cord. Response
latencies with the cathode over the cord and with the
anode 6 cm lateral to the cathode are similar to those
obtained when the anode and cathode are aligned vertically over the lower cervical cord in the midline, making excitation of any but the most proximal segments
of the motor roots unlikely. It is conceivable that cervical stimulation in fact excites descending corticospinal
tract fibers, but we have found that in order to stimulate motor fibers to the legs, the voltage applied over
the cervical region has to be appreciably higher than
that required to evoke maximal responses in arm muscles (unpublished observations, 1984). Overall, excitation of segmental neural elements-perhaps alpha
motor axons-within or near the cord seems most
Our findings clearly demonstrate that in severe MS,
corticospinal conduction time may be markedly prolonged, whereas conduction in the peripheral (cord-toaxilla) segment is normal. The delays seen are similar
to those encountered in MS when sensory evoked potentials are used to study afferent pathways, and it
seems likely that they reflect demyelination in motor
tracts. However, it would be unrealistic to expect these
findings to be specific for MS; it is entirely possible
that similar abnormalities could be found in other disorders such as compressive or vascular lesions.
Our findings in healthy subjects and MS patients are
similar to those obtained independently by Cowan and
colleagues IS}. Like them, we found that the procedure could be used in a clinical setting with no untoward effects. In a manner analogous to visual evoked
potentials that provide evidence of asymptomatic optic
neuritis, this technique may be valuable in detecting
clinically silent lesions involving the pyramidal tracts.
The technique undoubtedly has great potential for the
further study of central and proximal peripheral motor
function in many neurological conditions.
A preliminary report of this work was presented at the American
Association of Electromyography and Electrodiagnosis meeting at
Kansas City, September 1984 [19].
We are grateful to Prof P. A. Merton and Mr H . B. Morton for
advice and helpful discussion and to the physicians of The National
Hospital for permission to study their patients.
1. Ajmone Marsaon C: Focal electrical stimulation. In Purpura DP,
Penry JK, Tower D, et al (eds): Experimental Models of
Epilepsy. A Manual for the Laboratory Worker. New York,
Raven, 1972, pp 147-172
2. Bamford CR, Sibley WA, Cole T, et al: Trauma as an etiologic
and aggravating factor in multiple sclerosis. Neurology (NY)
31~1229-1234, 1981
3. Beradelli A, Cowan JMA, Day BL, et al: The site of facilitation
of the response to cortical stimulation during voluntary contraction in man. J Physiol (in press)
4. Beradelli A, Cowan JMA, Day BL, et al: Motor cortical excitability in Parkinson’s disease. J Physiol 353:33P, 1984
5. Cowan JMA, Dick JPR, Day BL, et al: Abnormalities in central
motor pathway conduction in multiple sclerosis. Lancet 2:304307, 1984
6. Ferrier D: The Functions of the Brain, 1st ed. London, Smith,
Elder. 1876
7. Fritsch T, Hitzig E O n the electrical excirabiliry of the cerebrum. Arch Anat Physiol Wiss Med 37:300-333, 1870
8. Gorman ALF: Differential patterns of activation of the pyramidal system elicited by surface anodal and cathodal cortical stimulation. J Neurophysiol 29:547-564, 1966
9. Gualtierotti T, Paterson AS: Electrical stimulation of the unexposed cerebral cortex. J Physiol 125:278-291, 1954
10. Hern JEC, Landgren S, Phillips CG, Porter R Selective excitation of corticofugal neurones by surface-anodal stimulation of
the baboon’s motor cortex. J Physiol 161:73-90, 1962
11. Levy WJ, York D H , McCaffrey M, Tanzer F Motor evoked
potentials from transcranial stimulation of the motor cortex in
humans. Neurosurgery 15:287-302, 1984
12. Marsden CD, Merton PA, Morton HB: Maximal twitches from
stimulation of the motor cortex in man. J Physiol 312:5P, 1980
13. Marsden CD, Merton PA, Morton HB: Percutaneous stimulation of spinal cord and brain: pyramidal tract conduction velocities in man. J Physiol 328:6P, 1982
14. Marsden CD, Merton PA, Morton HB: Direct electrical stimulation of corticospinal pathways through the intact scalp in human subjects. In Desmedt JE (ed). Motor Control Mechanisms
in Health and Disease. New York, Raven, 1983, pp 387-391
15. Merton PA, Morton HB: Stimulation of the cerebral cortex in
the intact human subject. Nature 285:227, 1980
16. Merton PA, Morton HB: Electrical stimulation of human motor
and visual cortex through the intact scalp. J Physiol 305:9-10P,
17. Merton PA, Morton HB, Hill DK, Marsden CD: Scope of a
technique for electrical stimulation of human brain, spinal cord,
and muscle. Lancet 2:597-600, 1982
18. Milner-Brown SH, Girvin JP, Brown W F The effects of motor
cortical stimulation on the excitability of spinal rnotoneurons in
man. Can J Neurol Sci 2:245-253, 1975
19. Murray NMF, Mills KR: Pyramidal tract conduction time in
normal subjects and multiple sclerosis patients. Muscle Nerve
7:563, 1984
Mills and Murray: Corticospinal Conduction in Multiple Sclerosis
Без категории
Размер файла
481 Кб
times, trace, sclerosis, multiple, conducting, corticospinal
Пожаловаться на содержимое документа