вход по аккаунту


Electroencephalographic response to transcranial magnetic stimulation in children Evidence for giant inhibitory potentials.

код для вставкиСкачать
Transcranial Magnetic Stimulation Evokes
Giant Inhibitory Potentials in Children
Stephan Bender, MD,1 Kristine Basseler,1 Imke Sebastian,2 Franz Resch, MD,1 Thomas Kammer, MD,3
Rieke Oelkers-Ax, MD,1 and Matthias Weisbrod, MD2
The electroencephalographic response to transcranial magnetic stimulation (TMS) recently has been established as a
direct parameter of motor cortex excitability. Its N100 component was suggested to reflect an inhibitory response. We
investigated influences of cerebral maturation on TMS-evoked N100 in 6- to 10-year-old healthy children. We used a
forewarned reaction time (contingent negative variation) task to test the effects of response preparation and sensory
attention on N100 amplitude. Single-pulse TMS of motor cortex at 105% motor threshold intensity evoked N100
amplitudes of more than 100␮V in resting children (visible in single trials), which correlated negatively with age and
positively with absolute stimulation intensity. During late contingent negative variation, which involves preactivation of
the cortical structures necessary for a fast response, N100 amplitude was significantly reduced. We conclude that (1)
N100 amplitude reduction during late contingent negative variation provides further evidence that TMS-evoked N100
reflects inhibitory processes, (2) response preparation and attention modulate N100, and (3) TMS-evoked N100 undergoes maturational changes and could serve to test cortical integrity and inhibitory function in children. Parallels between
the inhibitory N100 after TMS (provoking massive synchronous excitation) and the inhibitory wave component of
epileptic spike wave complexes are suggested.
Ann Neurol 2005;58:58 – 67
The electroencephalographic (EEG) response to transcranial magnetic stimulation (TMS) has been established as a new approach to characterize reactivity and
connectivity of the brain.1– 4 The TMS-evoked potential was separated from the artifact caused directly by
TMS-induced currents, the auditory-evoked potential
(AEP) produced by the coil click, the somatosensory
stimulation on the scalp, and reafferent proprioceptive
input from muscle contraction.3–5 It was interpreted as
a correlate of cortical excitability, which, in contrast
with the compound motor-evoked potential (MEP), is
able to separate cerebral from spinal cord or peripheral
The TMS-evoked N100 component has been suggested to reflect inhibitory processes3: It was reduced
immediately before movement onset compared with a
relaxed resting condition. In paired pulse paradigms,
inhibitory processes have been described with a peak
100 to 150msec after the conditioning stimulus in
adults6 – 8; that is, when TMS-evoked N100 of the
conditioning stimulus peaks.3
Inhibitory deficits are believed to play an important
role in the pathophysiology of epilepsy,9 attention deficit hyperactivity disorder10,11 or migraine,12,13 to
name but a few disorders that originate in childhood.
However, in contrast with MEP,14,15 maturation of the
TMS-evoked EEG response has not been studied in
children so far, though maturation of the motor system
is not completed until the late teenage years: The
readiness potential changes its polarity from positive to
negative,16 –18 and the late component of contingent
negative variation (late CNV, or lCNV) increases during childhood and adolescence in healthy subjects.19,20
During lCNV, the cortical areas involved in subsequent rapid processing of the imperative stimulus and
response execution are supposed to be preactivated.21
This study sought to extend the findings of the
TMS-evoked EEG response to the immature motor
system of children. Furthermore, we tested the hypothesis that TMS-evoked N100 is an inhibitory potential
that is reduced when cortex is preactivated.3 Therefore,
we examined TMS-evoked N100 during resting conditions and during lCNV in 6- to 10-year-old children.
From the 1Department for Child and Adolescent Psychiatry, and
Section for Experimental Psychopathology, Psychiatric Hospital,
University of Heidelberg, Heidelberg; and 3Psychiatric Hospital,
University of Ulm, Ulm, Germany.
Published online Jun 27, 2005 in Wiley InterScience
( DOI: 10.1002/ana.20521
Received Oct 12, 2004, and in revised form Mar 7, 2005. Accepted
for publication Apr 19, 2005.
Subjects and Methods
Seventeen right-handed (Edinburgh Handedness Inventory22), healthy children (10 girls, 7 boys) between 6.8 and
Address correspondence to Dr Bender, Department for Child and
Adolescent Psychiatry, University of Heidelberg, Blumenstrasse 8,
D-69115 Heidelberg, Germany.
© 2005 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services
10.0 (mean age ⫾ standard deviation, 8.8 ⫾ 0.8) years old
were examined. Seven right-handed, healthy young adults
(three female and four male adults; aged 25.0 ⫾ 1.8 years)
served as the control group. Subjects were screened for visual
impairments (corrected visus ⱖ 0.8) and neurological or psychiatric diseases. An individual or a family history of epilepsy
and an individual history of a convulsive attack of any type
were exclusion criteria. No subject took any psychoactive
drugs. All children were required a reliable motor threshold
below or equal to 100% of maximum stimulator output
(MSO; Magstim 200; Magstim, Wales, UK). The study was
approved by the local ethics committee. All participants and
the parents of all children provided informed written consent
in accordance with the Declaration of Helsinki. One subject
could not be included because of discomfort to TMS.
Contingent Negative Variation Task
We recorded 60 CNV trials using a visual warning stimulus
S1 (a black exclamation mark on 11.5 ⫻ 10cm white background [width ⫻ height], which was presented for 150msec)
and a visual imperative stimulus S2 (a black line drawing of
a hand on 11.5 ⫻ 10cm white background, which was presented for 150msec) on a computer screen (black background) at 0.8m distance. Intertrial intervals varied randomly
from 7 to 11 seconds. Stimulus onset asynchrony between
S1 and S2 was 3 seconds. In the 20 trials with TMS during
lCNV (2.8 seconds after the onset of S1), S2 was delayed
until 0.5 second after TMS (3.3 seconds after onset of S1) to
avoid interferences between processes related to TMS and to
presentation of S2. Subjects were instructed to respond as
fast as possible when S2 occurred by pressing a mouse button with the index finger of their nondominant left hand.
lCNV amplitude was determined as the mean amplitude
200msec before TMS in trials with magnetic stimulation.
Transcranial Magnetic Stimulation
Monophasic single-pulse TMS was applied using a Magstim
200 stimulator with a 9cm diameter circular coil. The coil’s
center was positioned over the vertex, and the coil was oriented tangentially to the skull on the right hemisphere so
that the maximum MEP amplitude was produced in the left
first dorsal interosseal muscle. The current flow within the
coil was clockwise to stimulate the right motor cortex. The
weight of the coil was carried by a tripod stand; the coil’s
position relative to the subject’s head was fixed with tape, as
well as an elastic band. An assistant assured manually that
the position of head and coil remained constant during the
recordings. Resting motor threshold (RMT) was determined
as the lowest intensity that produced an MEP amplitude of
at least 50␮V in at least 5 of 10 trials in the relaxed left first
dorsal interosseal muscle.23 Relaxation was controlled by
acoustic feedback and electromyogram (EMG) offline analysis. For TMS during the CNV task, intensity was set to
105% RMT to obtain reliable MEPs. When 105% RMT
exceeded MSO (three subjects, RMT 96, 98, and 100%
MSO), intensity was set to 100%. For both ethical and technical reasons, we refrained from applying stronger intensities.
Because this analysis focused on intraindividual but not on
between-subject comparisons, these subjects were not ex-
cluded (their inclusion did not crucially affect the presented
TMS was synchronized with the EEG recordings by TTL
triggers (Gentask, Stim software package; Neuroscan, El
Paso, TX): 20 magnetic stimulations were triggered during
lCNV, and 20 stimulations were triggered randomly during
the intertrial intervals (resting condition; see Fig 4). Stimulations during the resting condition and lCNV were randomly intercalated.
COMPARISONS BETWEEN ADULT SUBJECTS AND CHILDREN. Ten stimulations with 50% MSO were applied to
adults and children to compare maturational effects at the
same absolute intensity.
dian nerve was stimulated magnetically on the left forearm at
an intensity that produced a somatosensory sensation on the
forearm subjectively superior to the sensation TMS had produced on the scalp.
Electromyographic Recordings
Surface EMG (compound muscle action potential) was recorded from the left first dorsal interosseal muscle with AgAgCl electrodes (impedances ⬍ 10k⍀, filter 20 –10,000Hz;
Toennies Neuroscreen; Jaeger-Toennies, Hoechberg, Germany). Epochs of 600msec (180msec baseline before and
420msec after TMS) were analyzed offline: MEP interpeak
amplitudes at rest and during lCNV were calculated. Rectified background EMG 180msec before TMS was determined
in addition to acoustic EMG feedback.24
Electroencephalographic Recordings
We recorded continuous direct current EEG at a sampling
rate of 500Hz (Neuroscan Synamp; Neuroscan) against
linked mastoid reference. An antialiasing, low-pass filter of
70Hz and a 50Hz notch filter were used. Recordings were
taken at CP6⬘, CP5⬘, Cz, PO1⬘, PO2⬘, O1⬘, O2⬘, Oz, FP1‘,
FP10‘, and FP9‘ (extended international 10-20 system, minor deviations are indicated by ‘) by sintered Ag-AgCl disc
electrodes (impedances ⬍ 5k⍀). There was no electrode
heating25 under the single-pulse conditions used. Deblocking
with a sample-and-hold circuit was applied by a trigger from
the magnetic stimulator to the EEG amplifier. CP6⬘ was
chosen for most analyses because previous reports point toward a maximum of the TMS-evoked EEG response of motor cortex slightly posterior to and below C3/C4 ipsilateral to
Signal Preprocessing
For the analysis of CNV, recordings 1 second before S1
served as baseline. The EEG-signal was low-pass filtered offline (20Hz high cutoff, zero-phase Butterworth filter, slope
24 dB/octave) and segmented into epochs of 7.5 seconds (1
second before S1 to 3.5 seconds after S2). Artifacts were rejected by visual inspection. TMS artifacts made automatic
artifact detection or ocular correction procedures impossible.
Overall, less than one third of the trials had to be removed.
For the analysis of TMS-evoked potentials, a baseline
Bender et al: EEG Response to TMS in Children
500msec before TMS was applied; the EEG was segmented
into intervals from 500msec before to 500msec after TMS.
The timing of the event triggers was controlled for latency
jitter (tview function of Gentask Stim software package;
Neuroscan). Data were low-pass filtered (same parameters as
above), and artifacts were removed manually by visual inspection (less than 10% of trials had to be rejected).
Data Analysis
TMS-evoked N100 was determined as the highest negative
peak at CP6⬘ in the interval 80 to 200msec after TMS.3,4
Visual inspection confirmed that there was only one prominent negative peak in this latency range. N100 amplitude
was calculated as the mean amplitude in the time window
40msec (⫾20msec) around the N100 peak at CP6⬘. N100
latency was determined as the time from TMS to the peak
maximum at CP6⬘. N100 was separated from the TMSinduced artifact and the “N100” that was produced when
only the electrodes were stimulated on a glass head dummy,
which was covered by a cloth soaked with water (simulating
the impedances of skull and scalp, respectively) so that impedances of about 5k⍀ were yielded.
Linear regression analysis was used to determine the children’s stimulus–intensity slope for TMS-evoked N100 and
its maturational development with increasing age. The development of RMT was also examined by linear regression.
The following further comparisons were performed
(paired t tests). First, N100 amplitude at CP6⬘ and CP5⬘
was compared to test for lateralization. Second, N100 topography and amplitudes were compared with known topographies and amplitudes of AEP (produced by the coil click)5
and somatosensory-evoked potentials (produced by the haptic sensation on the scalp)4 to find out whether other evoked
potentials apart from direct motor cortex activation by TMS
could account for N100. Third, mean N100 amplitudes in
trials with MEP amplitudes above and below the median3,4
were compared to elucidate influences of proprioceptive feedback on N100 and the relation between TMS-evoked N100
and the EMG response to TMS. Fourth, influences of the CNV
task were examined: N100 amplitude at CP6⬘ was tested for
task influences by comparing TMS-evoked N100 at rest and
during lCNV. Fifth, the difference between TMS-evoked
N100 amplitude at rest and during lCNV was compared with
lCNV amplitude to test whether a baseline shift could account
for reduced N100 amplitudes during lCNV. Sixth, N100 amplitudes during the first and the second half of trails with
TMS at rest (during the intertrial intervals) were compared in
order to test for significant habituation/potentiation.
Separation of N100 from Transcranial Magnetic
Stimulation–Induced Artifacts
Figure 1 shows the differentiation of N100 from the
TMS-induced artifact that remained in the recordings
despite the deblocking device. The short sharp EEG
artifact produced by the TMS pulse was transformed
by the amplifier’s hardware and filter characteristics to
a slightly delayed high-frequency oscillation with decreasing amplitude. The characteristics of the head
Annals of Neurology
Vol 58
No 1
July 2005
dummy did not critically change the electromagnetic
artifact. Mean N100 amplitude at CP6⬘ at 105%
RMT for 6- to10-year-old children was ⫺130.6 ⫾
71.9␮V at rest (values are given mean ⫾ standard deviation unless indicated otherwise). This value differed
significantly from the “N100” amplitude when the
electrodes were stimulated on a head dummy: 4.8 ⫾
0.9␮V (t ⫽ 7.8; p ⬍ 0.00001; N100 was absent).
We replicated that the EEG responses to left median
nerve stimulation yielded no such prominent negativity
in the N100 latency range (not shown).
Latencies and Topography of N100
Figure 2 shows the topography of TMS-evoked N100.
N100 was significantly lateralized, as demonstrated by
the difference of N100 amplitude at CP6⬘ versus
CP5⬘: ⫺37.7 ⫾ 61.8␮V (t ⫽ 2.5; p ⫽ 0.02). Mean
“N100” latency for 6- to 10-year-old children was
168 ⫾ 12msec at 105% RMT. Reduction of stimulation intensity to 50% MSO resulted in a decrease of
mean “N100” latency to 147 ⫾ 17msec. In adults,
N100 latency at 50% MSO was significantly shorter:
112 ⫾ 23msec (t ⫽ 4.9; p ⬍ 0.0001).
Maturation of N100
EFFECTS OF AGE ON N100 AMPLITUDE. The steep regression slope when N100 amplitude was predicted by
age (Fig 3) almost reached significance within our limited age range (r ⫽ 0.47; p ⫽ 0.056; regression slope,
41.4 ⫾ 20.0␮V/year [mean ⫾ standard error]; the
positive value indicates a decrease of the negative N100
RMT also decreased significantly with increasing age
(see Fig 3). The regression slope was ⫺12.3 ⫾ 4.4%
per year (mean ⫾ standard error) in the children’s
group (r ⫽ 0.58; p ⫽ 0.01). The children’s mean
RMT was 76.7 ⫾ 17.2%; the adults’ mean RMT was
49.1 ⫾ 12.1%.
DEPENDENCE. The stimulus–
intensity slope for N100 amplitude was ⫺2.8 ⫾
0.8␮V per percentage point of MSO (mean ⫾ standard error; r ⫽ 0.69; p ⫽ 0.002; see Fig 3) for 6- to
10-year-old children. The negative slope indicates an
increase of the negative surface potential. Intraindividual stimulus-intensity dependence is also illustrated
in Figure 3.
After correcting for different absolute stimulation intensities, the correlation between age and TMS-evoked
N100 amplitude disappeared (partial correlation r⬘ ⫽
0.12; p ⫽ 0.67). However, when adults and children
were stimulated by the same absolute intensity (50%
Fig 1. Electroencephalographic (EEG) response to transcranial magnetic stimulation (TMS) of right motor cortex (right) in contrast
with the artifact induced by TMS when the electrodes were stimulated on a head dummy with the same coil position (left). The
response shown at the right was recorded at CP6⬘ in one child (compare with the TMS-induced artifact in the same lead). The
arrows indicate TMS application (left 100% maximum stimulator output [MSO], right 74% MSO). Electrode impedances were
5k⍀ in both cases. Note that the N100 component from about 100msec on was not distorted by the TMS artifact and that 20Hz
low-pass filtering eliminated only artifactual high-frequency oscillations without affecting TMS-evoked N100. The EEG artifact
produced by the short TMS pulse was transformed by amplifier characteristics and filter settings to a rapidly decreasing highfrequency oscillation. The deblocking device (sample-and-hold circuit) introduced a short delay after TMS but was not sufficient to
eliminate the TMS-induced artifact.
MSO), mean N100 amplitude at CP6⬘ at rest was
⫺8.7 ⫾ 4.1␮V for adults and ⫺26.1 ⫾ 20.1␮V for
children (t ⫽ 2.2; p ⫽ 0.035).
Effects of the Contingent Negative Variation Task
Mean lCNV amplitude was small but significant in 6to 10-year-old children: ⫺5.0 ⫾ 8.6␮V at Cz (t ⫽ 2.4;
p ⫽ 0.03), and ⫺2.6 ⫾ 4.8␮V at CP6⬘ (t ⫽ 2.2; p ⫽
0.04). For N100 amplitude reduction during lCNV, see
Figure 4 and the Table. Though the effect was small
regarding absolute average N100 amplitude, it was
found reliably in the intraindividual comparisons.
N100 amplitude difference at CP6⬘ between TMS at
rest and during lCNV was significantly larger than
lCNV: ⫺9.1 ⫾ 12.2␮V (t ⫽ 3.1; p ⫽ 0.008). The
correlation between mean individual lCNV amplitudes
and the mean individual decreases in TMS-evoked N100
amplitude was not significant (r ⫽ 0.05; p ⫽ 0.85).
Comparisons of Motor-Evoked Potential Amplitude
and Transcranial Magnetic Stimulation–Evoked
N100 Amplitude
See the Table for MEP amplitudes at rest and during
lCNV. There was no significant correlation between
mean MEP amplitude and mean TMS-evoked N100
amplitude at CP6⬘ (r ⫽ 0.09; p ⫽ 0.74), nor was there
a significant correlation between rectified pre-TMS
EMG and N100 amplitude (r ⫽ 0.1; p ⫽ 0.7). The
difference of N100 amplitude at CP6⬘ between trials
with MEP amplitudes above and below the median
was 3.8 ⫾ 15.6␮V (t ⫽ 1.0; p ⫽ 0.33).
Mean N100 amplitude at CP6⬘ for the first and second halves of the trials were ⫺125.1 ⫾ 70.2␮V and
⫺136.1 ⫾ 73.8␮V, respectively. The difference was
⫺11.0 ⫾ 11.6␮V (t ⫽ 3.9; p ⫽ 0.001; Fig 5).
Methodological Issues: Origin of Transcranial
Magnetic Stimulation–Evoked N100
Children aged 6 to 10 years had giant, TMS-evoked,
N100 amplitudes of more than 100␮V at 105%
RMT. N100 could be easily distinguished visually in
single trials. Its topography, lateralized ipsilaterally to
the side of stimulation with a centroparietal negative
maximum and frontopolar positivity, was in clear
Bender et al: EEG Response to TMS in Children
Fig 2. Topography of the electroencephalographic (EEG) response (grand average) to transcranial magnetic stimulation (TMS) at
105% motor threshold averaged over all healthy children (linked mastoid reference). Left electrodes have odd numbers. The vertical
dashed line indicates when TMS was applied. Note the lateralization ipsilateral to the side of the stimulation (right), the centroparietal negative maximum, and the frontopolar positivity. EOG (VEOG indicates an electrode 1cm below the left eye) did not critically influence the presented results (see VEOG and frontopolar leads).
agreement with an origin in the stimulated ipsilateral
motor cortex26 (see Fig 2), though the time-course of
the potential at centroparietal and frontopolar sites was
not perfectly “mirrored.”
The TMS-evoked N100 could be well separated
from the electromagnetically induced artifact or artifacts produced by coil vibration (see Fig 1) using a
common direct current amplifier. We could not assess
the early components of the TMS-evoked potential3,4,27 because of insufficient deblocking of the
TMS-induced artifact, though in grand averages a peak
that could correspond to N452– 4 (see Fig 4) appeared.
The TMS-induced artifact, but not TMS-evoked
N100, showed considerable variability and was strongly
influenced by the relative position of the coil with respect to electrodes and cables.
No white noise was used to mask the coil click4 because of its arousing effects (blurring task condition effects) and because children would not have tolerated
the additional stress. However, N100 could not be explained by an AEP after the coil click5,28 because AEPs
show a less lateralized topography,29 much lower amplitudes especially at central leads in children, and
shorter N1 latencies.29 –32
The influence of proprioceptive reafferences was
small: We replicated that TMS-evoked N100 amplitude did not correlate with MEP amplitudes of the target muscle.3,4 This suggested that MEP and TMSevoked N100 are two independent parameters.
The somatosensory-evoked potentials resulting from
the haptic sensation on the scalp would have produced
Annals of Neurology
Vol 58
No 1
July 2005
a potential lateralized contralaterally to the side of
stimulation3,4 with much lower amplitudes.33 This is
in concordance with a control stimulation of the medianus nerve yielding much lower amplitudes in the
N100 latency range (AEP, somatosensory-evoked potentials, proprioceptive reafferences).3,34
Moreover, known stimulus-intensity dependencies
for these sensory-evoked potentials are far less pronounced than in our data.35–39
Stimulus-Intensity Dependence and Maturation
TMS-evoked N100 amplitude depended strongly on
the intensity of stimulation. Contradictory previous results in adults2– 4 could be attributed to smaller N100
amplitudes in adult subjects and lower absolute stimulus intensities used.
TMS-evoked N100 amplitude decreased with increasing age in the narrow age range examined (see Fig
3). When TMS-evoked N100 amplitude values of
adults are considered,2– 4 a decrease of N100 amplitude
through school age is further supported. This agerelated decrease might be attributed not only to maturational effects, but mainly to an age-related decrease of
RMT15,40 – 42 and the different absolute stimulus intensities applied. The comparison of N100 amplitude at
the same absolute intensity between children and
adults showed also true maturational differences. The
effect appears too large for a simple increase in skull
thickness when compared with the development of
other evoked potentials such as auditory N1.30 –32 Latency shortening with increasing age is a well-known
Fig 3. Scatterplots illustrating the age-dependent decrease of transcranial magnetic stimulation (TMS)–evoked N100 amplitude, the
decrease of motor threshold with increasing age, and interindividual and intraindividual stimulus-intensity dependence of TMSevoked N100 at CP6⬘. (Top left) Maturation of TMS-evoked N100 at CP6⬘ at an intensity of 105% resting motor threshold
(RMT; r ⫽ 0.47; p ⫽ 0.056). Values in adults further support an age-related amplitude decrease. (Top right) RMT (percentage
of maximum stimulator output [MSO]) as a function of age (r ⫽ 0.58; p ⫽ 0.01). (Bottom left) Stimulus-intensity dependence of
TMS-evoked N100. On the abscissa, absolute stimulator output values are given. Note that except for three subjects (100% MSO),
the applied intensity was adjusted to 105% RMT (same relative but not absolute stimulus intensity; r ⫽ 0.69; p ⫽ 0.002). (Bottom right) The TMS-evoked potential at CP6⬘ is shown at 40, 50, 60, and 70% MSO (57, 71, 86, and 100% RMT) for a
representative child. Between-subject stimulus-intensity dependence was not accidentally produced by interindividual variability. The
arrow indicates TMS application. Note that TMS-evoked N100 showed a more pronounced stimulus-intensity dependence than the
TMS-induced artifact, which, in contrast with TMS-evoked N100 amplitude, showed considerable variability (see Fig 2).
finding in developmental evoked potential/eventrelated potential (EP/ERP) research.43
Effects of Attention and Motor Preparation during
Late Contingent Negative Variation
The children’s lCNV amplitude was small but significant.19,20 TMS-evoked N100 amplitude was reduced
during motor preparation and stimulus anticipation
during lCNV when the cortex was preactivated.21 This
clearly points toward N100 as inhibitory potential.3 Inhibitory processes in deep cortical layers produce
surface-negative potentials.44
The random sequence of TMS during intertrial intervals and lCNV avoided systematic confounding effects of external conditions on N100 results.
Mean MEP amplitude during lCNV was significantly larger than at rest (previous findings in adults
had not always reached significance24,45,46). EMG (increased MEP amplitude during lCNV) and the cortical
response to TMS (decreased N100 amplitude) were
again dissociated. EMG preactivation did not differ for
the rest and lCNV conditions; therefore, changes in
MEP amplitude were not likely to be caused by taskinduced differences in muscle precontraction.24
N100 amplitude reduction during lCNV could not
be explained by effects of lCNV on other evoked potential components: Probe stimuli during CNV have
been shown to increase the N1/P3 complex47,48 so that
effects of CNV on AEPs, somatosensory-, or
proprioceptive-evoked potentials would have resulted
Bender et al: EEG Response to TMS in Children
Fig 4. (A) Experimental setup. A schematic contingent negative variation (CNV) waveform is given. The vertical dashed line indicates presentation of S2; transcranial magnetic stimulations (TMSs; indicated by a flash) during the intertrial interval were randomly interspersed with stimulations during late CNV (lCNV; 2.8 seconds after S1). (B) Grand averages (n ⫽ 17 children) of
TMS-evoked N100 at CP6⬘ for the resting (TMS during the intertrial interval) and the lCNV condition. The vertical dashed
line indicates TMS. Note that the small peak between the TMS-induced artifact (first negative peak) and the TMS-evoked N100
(third broad negative peak) may represent a small N45/55,2– 4 which is shadowed in our recordings by the TMS-induced artifact
(see Fig 1). (C) Scatterplot illustrating the highly consistent (in 15/17 children) decrease in TMS-evoked N100 amplitude during
lCNV (although it was small with respect to absolute N100 amplitude). Note that for only 2 of 17 children (7.0 and 8.5 years
old) equal or greater N100 amplitudes were found during lCNV.
Annals of Neurology
Vol 58
No 1
July 2005
Table. EEG and EMG Responses to TMS at Rest and during Late CNV
N100 amplitude at CP6⬘ (⫾SD)
MEP amplitude
Background EMG pre TMS
Resting Condition
Late CNV
⫺130.6 ⫾ 71.9␮V
218 ⫾ 143␮V
6.9 ⫾ 2.2␮V
⫺118.9 ⫾ 72.1␮V
279 ⫾ 174␮V
7.3 ⫾ 2.5␮V
11.7 ⫾ 11.0␮V (t ⫽ 4.4; p ⬍ 0.001)
61 ⫾ 80␮V (t ⫽ 3.2; p ⫽ 0.006)
0.5 ⫾ 1.9␮V (n.s.)
EEG ⫽ electroencephalogram; EMG ⫽ electromyogram; TMS ⫽ transcranial magnetic stimulation; CNV ⫽ contingent negative variation;
MEP ⫽ motor-evoked potential.
in an increase of N100 amplitude and in a different
biphasic waveform (N1/P3). However, the established
differences corresponded in their time course to the
waveform of the TMS-evoked N100 at rest.
TMS-evoked N100 appears sensitive to changes in
attention and motor preparation and may prove to be a
promising tool to investigate cortical activity in cognitive tasks. Nikulin and colleagues3 recently described
that a visual stimulus influenced N100 amplitude.
There was a significant potentiation of N100 amplitude during the recording session, although intervals
between subsequent TMS exceeded 5 seconds. Lowfrequency TMS exerts inhibitory effects on the cortex49 –51; however, no influences on MEP amplitudes42
have been described for frequencies less than 0.2Hz.
Explicatory Mechanisms
The TMS-evoked N100 component could result from
inhibitory cortical interneurons directly excited by
TMS3; however, even for long-lasting IPSP,3,52 the
long peak latency is difficult to explain.
Transcranial magnetic stimulation produces a sudden synchronized neuronal discharge that in this re-
spect is similar to an epileptic spike. To prevent a sudden synchronized discharge from generalization,
intracortical, as well as thalamocortical, inhibitory
mechanisms are activated via the thalamic reticular nucleus as established in animal models,53,54 producing
typical spike wave complexes in the EEG.44,55,56
Therefore, TMS-evoked N100 could be interpreted as
“wave” response to an externally generated “spike.” It
might provide an in vivo model to assess thalamocortical inhibitory processes in human subjects.
During lCNV, the GABAergic cells in the thalamic
reticular nucleus are inhibited,21 thus N100 may be
reduced during lCNV because this inhibition decreases
the effect of synchronized corticothalamic input. This
hypothetical model needs confirmation by further research.
In summary, we found an easily distinguishable TMSevoked N100 amplitude by near–motor threshold
stimulation in 6- to 10-year-old children, which correlated with absolute rather than relative stimulus intensity and showed age-dependent maturation. TMSevoked N100 was independent from MEP amplitudes
and showed significant potentiation throughout a trial
block. During lCNV (cortical preactivation), TMSevoked N100 was significantly reduced, pointing toward a surface-negative inhibitory potential. TMSevoked N100 may be a valuable tool to investigate
inhibitory functioning in children.
This work was supported by the Medical Young Investigator Award
of the Medical Faculty of the University of Heidelberg (S.B.).
We thank Professor H.-M. Meinck for providing technical support
and Dr S. Zimmermann for editing the manuscript for fluency in
Fig 5. Potentiation of transcranial magnetic stimulation–
evoked N100. The grand averages for 6- to 10-year-old children for the first (thin line) and the second (thick line) 10
stimulations during the rest condition at CP6⬘ are presented.
TMS ⫽ transcranial magnetic stimulation.
1. Ilmoniemi RJ, Virtanen J, Ruohonen J, et al. Neuronal responses to magnetic stimulation reveal cortical reactivity and
connectivity. Neuroreport 1997;8:3537–3540.
2. Komssi S, Kahkonen S, Ilmoniemi RJ. The effect of stimulus
intensity on brain responses evoked by transcranial magnetic
stimulation. Hum Brain Mapp 2004;21:154 –164.
Bender et al: EEG Response to TMS in Children
3. Nikulin VV, Kicic D, Kahkonen S, Ilmoniemi RJ. Modulation
of electroencephalographic responses to transcranial magnetic
stimulation: evidence for changes in cortical excitability related
to movement. Eur J Neurosci 2003;18:1206 –1212.
4. Paus T, Sipila PK, Strafella AP. Synchronization of neuronal
activity in the human primary motor cortex by transcranial
magnetic stimulation: an EEG study. J Neurophysiol 2001;86:
5. Nikouline V, Ruohonen J, Ilmoniemi RJ. The role of the coil
click in TMS assessed with simultaneous EEG. Clin Neurophysiol 1999;110:1325–1328.
6. Nakamura H, Kitagawa H, Kawaguchi Y, Tsuji H. Intracortical
facilitation and inhibition after transcranial magnetic stimulation in conscious humans. J Physiol 1997;498:817– 823.
7. Roick H, von Giesen HJ, Benecke R. On the origin of the
postexcitatory inhibition seen after transcranial magnetic brain
stimulation in awake human subjects. Exp Brain Res 1993;94:
489 – 498.
8. Valls-Sole J, Pascual-Leone A, Wassermann EM, Hallett M.
Human motor evoked responses to paired transcranial magnetic
stimuli. Electroencephalogr Clin Neurophysiol 1992;85:
9. Bernard C, Cossart R, Hirsch JC, et al. What is GABAergic
inhibition? How is it modified in epilepsy? Epilepsia 2000;
41(suppl 6):S90 –S95.
10. Overtoom CC, Kenemans JL, Verbaten MN, et al. Inhibition
in children with attention-deficit/hyperactivity disorder: a psychophysiological study of the stop task. Biol Psychiatry 2002;
51:668 – 676.
11. Moll GH, Heinrich H, Trott G, et al. Deficient intracortical
inhibition in drug-naive children with attention-deficit hyperactivity disorder is enhanced by methylphenidate. Neurosci Lett
12. Cutrer FM, Moskowitz MA. Wolff Award 1996. The actions of
valproate and neurosteroids in a model of trigeminal pain.
Headache 1996;36:579 –585.
13. Palmer JE, Chronicle EP, Rolan P, Mulleners WM. Cortical
hyperexcitability is cortical under-inhibition: evidence from a
novel functional test of migraine patients. Cephalalgia 2000;20:
14. Muller K, Kass-Iliyya F, Reitz M. Ontogeny of ipsilateral corticospinal projections: a developmental study with transcranial
magnetic stimulation. Ann Neurol 1997;42:705–711.
15. Nezu A, Kimura S, Uehara S, et al. Magnetic stimulation of
motor cortex in children: maturity of corticospinal pathway and
problem of clinical application. Brain Dev 1997;19:176 –180.
16. Otto D, Reiter L. Developmental changes in slow cortical potentials of young children with elevated body lead burden. Neurophysiological considerations. Ann N Y Acad Sci 1984;425:
17. Chiarenza GA, Papakostopoulos D, Giordana F, GuareschiCazzullo A. Movement-related brain macropotentials during
skilled performances. A developmental study. Electroencephalogr Clin Neurophysiol 1983;56:373–383.
18. Warren C, Karrer R. Movement-related potentials in children.
A replication of waveforms, and their relationships to age, performance and cognitive development. Ann N Y Acad Sci 1984;
425:489 – 495.
19. Bender S, Weisbrod M, Just U, et al. Lack of age-dependent
development of the contingent negative variation (CNV) in migraine children? Cephalalgia 2002;22:132–136.
20. Timsit-Berthier M, Hausmann J. [Study of the CNV and of
the phenomenon of motor preparation in 5-15- year-old children]. Rev Electroencephalogr Neurophysiol Clin 1972;2:
124 –136.
Annals of Neurology
Vol 58
No 1
July 2005
21. Rockstroh B, Elbert T, Canavan A, et al. Slow cortical potentials and behavior. Baltimore, MD: Urban and Schwarzenberg,
22. Oldfield RC. The assessment and analysis of handedness: the
Edinburgh inventory. Neuropsychologia 1971;9:97–113.
23. Rossini PM, Barker AT, Berardelli A, et al. 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 1994;91:79 –92.
24. Hoshiyama M, Kakigi R. Shortening of the cortical silent period following transcranial magnetic brain stimulation during
an experimental paradigm for generating contingent negative
variation (CNV). Clin Neurophysiol 1999;110:1394 –1398.
25. Roth BJ, Pascual-Leone A, Cohen LG, Hallett M. The heating
of metal electrodes during rapid-rate magnetic stimulation: a
possible safety hazard. Electroencephalogr Clin Neurophysiol
1992;85:116 –123.
26. Gerloff C, Uenishi N, Hallett M. Cortical activation during fast
repetitive finger movements in humans: dipole sources of
steady-state movement-related cortical potentials. J Clin Neurophysiol 1998;15:502–513.
27. Virtanen J, Ruohonen J, Naatanen R, Ilmoniemi RJ. Instrumentation for the measurement of electric brain responses to
transcranial magnetic stimulation. Med Biol Eng Comput
28. Tiitinen H, Virtanen J, Ilmoniemi RJ, et al. Separation of contamination caused by coil clicks from responses elicited by
transcranial magnetic stimulation. Clin Neurophysiol 1999;
29. Naatanen R, Picton T. The N1 wave of the human electric and
magnetic response to sound: a review and an analysis of the
component structure. Psychophysiology 1987;24:375– 425.
30. Pang EW, Taylor MJ. Tracking the development of the N1
from age 3 to adulthood: an examination of speech and nonspeech stimuli. Clin Neurophysiol 2000;111:388 –397.
31. Ceponiene R, Rinne T, Naatanen R. Maturation of cortical
sound processing as indexed by event-related potentials. Clin
Neurophysiol 2002;113:870 – 882.
32. Gomes H, Dunn M, Ritter W, et al. Spatiotemporal maturation of the central and lateral N1 components to tones. Brain
Res Dev Brain Res 2001;129:147–155.
33. Kemner C, Verbaten MN, Cuperus JM, et al. Visual and somatosensory event-related brain potentials in autistic children
and three different control groups. Electroencephalogr Clin
Neurophysiol 1994;92:225–237.
34. Paus T, Castro-Alamancos MA, Petrides M. Cortico-cortical
connectivity of the human mid-dorsolateral frontal cortex and
its modulation by repetitive transcranial magnetic stimulation.
Eur J Neurosci 2001;14:1405–1411.
35. Afra J, Proietti Cecchini A, Sandor PS, Schoenen J. Comparison of visual and auditory evoked cortical potentials in migraine
patients between attacks. Clin Neurophysiol 2000;111:
1124 –1129.
36. Massey AE, Marsh VR, McAllister-Williams RH. Lack of effect
of tryptophan depletion on the loudness dependency of auditory event related potentials in healthy volunteers. Biol Psychol
37. Hegerl U, Prochno I, Ulrich G, Muller-Oerlinghausen B. Sensation seeking and auditory evoked potentials. Biol Psychiatry
1989;25:179 –190.
38. Hegerl U, Wulff H, Muller-Oerlinghausen B. Intensity dependence of auditory evoked potentials and clinical response to
prophylactic lithium medication: a replication study. Psychiatry
Res 1992;44:181–190.
39. Nakajima Y, Imamura N. Relationships between attention effects and intensity effects on the cognitive N140 and P300
components of somatosensory ERPs. Clin Neurophysiol 2000;
40. Garvey MA, Ziemann U, Bartko JJ, et al. Cortical correlates of
neuromotor development in healthy children. Clin Neurophysiol 2003;114:1662–1670.
41. Rossini PM, Desiato MT, Caramia MD. Age-related changes of
motor evoked potentials in healthy humans: non-invasive evaluation of central and peripheral motor tracts excitability and
conductivity. Brain Res 1992;593:14 –19.
42. Kammer T, Beck S, Thielscher A, et al. Motor thresholds in
humans: a transcranial magnetic stimulation study comparing
different pulse waveforms, current directions and stimulator
types. Clin Neurophysiol 2001;112:250 –258.
43. Oelkers-Ax R, Bender S, Just U, et al. Pattern-reversal visualevoked potentials in children with migraine and other primary
headache: evidence for maturation disorder? Pain 2004;108:
44. Caspers H, Speckmann EJ, Lehmenkuhler A. Electrogenesis of
cortical DC potentials. Prog Brain Res 1980;54:3–15.
45. Hausler UH, Lutzenberger W, Birbaumer N. Simultaneous recording of slow brain potentials and transcranial magnetic stimulation of hand area in human motor cortex. Neurosci Lett
46. Seyal M, Mull B, Bhullar N, Ahmad T, et al. Anticipation and
execution of a simple reading task enhance corticospinal excitability. Clin Neurophysiol 1999;110:424 – 429.
47. Rockstroh B, Muller M, Wagner M, et al. Event-related and
motor responses to probes in a forewarned reaction time task in
schizophrenic patients. Schizophr Res 1994;13:23–34.
48. Rockstroh B, Muller M, Wagner M, et al. “Probing” the nature
of the CNV. Electroencephalogr Clin Neurophysiol 1993;87:
49. Ogawa A, Ukai S, Shinosaki K, et al. Slow repetitive transcranial
magnetic stimulation increases somatosensory high-frequency oscillations in humans. Neurosci Lett 2004;358:193–196.
50. Weber M, Eisen AA. Magnetic stimulation of the central and
peripheral nervous systems. Muscle Nerve 2002;25:160 –175.
51. Khedr EM, Gilio F, Rothwell J. Effects of low frequency and
low intensity repetitive paired pulse stimulation of the primary
motor cortex. Clin Neurophysiol 2004;115:1259 –1263.
52. Krnjevic K, Randic M, Straughan DW. An inhibitory process
in the cerebral cortex. J Physiol 1966;184:16 – 48.
53. Steriade M, Amzica F. Sleep oscillations developing into seizures
in corticothalamic systems. Epilepsia 2003;44(suppl 12):9 –20.
54. Blumenfeld H. From molecules to networks: cortical/
subcortical interactions in the pathophysiology of idiopathic
generalized epilepsy. Epilepsia 2003;44(suppl 2):7–15.
55. Niedermeyer E, Lopes da Silva F. Electroencephalography: basic
principles, clincal applications and related fields. 4th ed.
Philadelphia: Lippincott Williams and Wilkins, 1999.
56. Blumenfeld H, McCormick DA. Corticothalamic inputs control the pattern of activity generated in thalamocortical networks. J Neurosci 2000;20:5153–5162.
Bender et al: EEG Response to TMS in Children
Без категории
Размер файла
280 Кб
potential, electroencephalographic, transcranial, response, giants, inhibitors, evidence, magnetic, children, stimulating
Пожаловаться на содержимое документа