Journal of Neuroscience Methods 171 (2008) 132–139
Concurrent excitation of the opposite motor cortex during transcranial
magnetic stimulation to activate the abdominal muscles
Henry Tsao a , Mary P. Galea b , Paul W. Hodges a,∗
a
NHMRC Centre of Clinical Research Excellence in Spinal Pain, Injury and Health, School of Health of Rehabilitation Sciences,
The University of Queensland, Brisbane, QLD 4072, Australia
b School of Physiotherapy, The University of Melbourne, Melbourne, Australia
Received 10 January 2008; received in revised form 3 February 2008; accepted 4 February 2008
Abstract
The study investigated the potential for stimulation of both motor cortices during transcranial magnetic stimulation (TMS) to evoke abdominal
muscle responses. Electromyographic activity (EMG) of transversus abdominis (TrA) was recorded bilaterally in eleven healthy volunteers using
fine-wire electrodes. TMS at 120% motor threshold (MT) was delivered at rest and during 10% activation at 1 cm intervals from the midline
to 5 cm lateral, along a line 2 cm anterior to the vertex. The optimal site to evoke responses in TrA is located 2 cm lateral to the vertex. When
bilateral abdominal responses were evoked at or lateral to this site, onset of ipsilateral motor evoked potentials (MEPs) were ∼3–4 ms longer than
contralateral MEPs. The difference between latencies is consistent with activation of faster crossed-, and slower uncrossed-corticospinal pathways
from one hemisphere. However, latencies of MEPs were similar between sides when stimulation was applied more medially and were consistent
with concurrent activation of crossed corticospinal tracts on both sides. The findings suggest that stimulation of both motor cortices is possible
when TMS is delivered less than 2 cm from midline. Concurrent stimulation of both motor cortices can be minimised if TMS is delivered at least
2 cm lateral to midline.
© 2008 Elsevier B.V. All rights reserved.
Keywords: Transcranial magnetic stimulation; Abdominal muscle; Corticospinal pathways; Motor evoked potential; Motor cortex
1. Introduction
The abdominal muscles can be activated by the motor cortex via rapid-conducting corticospinal projections to the spinal
motoneurons (Plassman and Gandevia, 1989). One method to
investigate excitability of these pathways is transcranial magnetic stimulation (TMS). However, corticospinal cells in the
motor cortex that project to the abdominal motoneurons are
located medially near the central fissure (Penfield and Boldrey,
1937). Thus it is unclear whether induced-currents from magnetic stimulation over the motor cortex can directly stimulate
the opposite hemisphere. If both cortices are stimulated concurrently, it would be impossible to determine the origin of
ipsilateral and contralateral evoked responses.
Few studies have investigated the potential for concurrent
stimulation of both motor cortices during TMS to elicit acti-
∗
Corresponding author. Tel.: +61 7 3365 4567; fax: +61 7 3365 4567.
E-mail address: [email protected] (P.W. Hodges).
0165-0270/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jneumeth.2008.02.005
vation of muscles with midline motor representations. Several
studies have reported that stimulation at scalp sites 2 cm lateral to the midline can evoke responses of the contralateral
abdominal muscles with a latency that is ∼2–5 ms faster than
responses on the ipsilateral side (Fujiwara et al., 2001; Strutton
et al., 2004; Tunstill et al., 2001). The latency of contralateral
responses is consistent with activation of crossed corticospinal
pathways (Gandevia and Plassman, 1988). The slower ipsilateral
responses however are argued to be unlikely due to excitation of the opposite hemisphere as this latency was longer, and
more likely due to activation of uncrossed polysynaptic corticospinal fibres from the same motor cortex (Ziemann et al.,
1999). However, those studies stimulated 2 cm laterally from
the midline where the potential for concurrent stimulation may
be minimal. Furthermore, studies to date have used surface electrodes to record responses from the abdominal muscles. As
surface electrodes can record myoelectric activity from multiple abdominal muscle layers due to cross-talk, this makes it
difficult to be certain that the responses evoked from different
stimulation sites involve the same specific abdominal muscles.
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
Therefore, whether stimulation of sites medial or lateral to the
recommended site (2 cm lateral to midline) could induce concurrent excitation of both motor cortices to evoke responses in
the abdominal muscles, recorded with intramuscular electrodes,
has not been systematically investigated.
The primary aim of this study was to examine the potential for stimulation of both motor cortices during TMS to elicit
responses in the abdominal muscles at various distances from
the midline. If motor evoked potentials (MEP) contralateral and
ipsilateral to the side of stimulation have similar latencies, this
could indicate simultaneous stimulation of the faster crossed
corticospinal pathways from direct excitation of both motor cortices. However, if the latency of MEPs on the contralateral side
of stimulation is faster than those evoked ipsilateral to the stimulated hemisphere, this would suggest that both responses are
evoked from the same motor cortex.
2. Materials and methods
2.1. Participants
Eleven right-handed healthy individuals were recruited (4
male, 7 female; age: 23 ± 3 (mean ± S.D.) years; height:
170 ± 10 cm; weight: 67 ± 16 kg). Subjects were excluded if
they had a history or family history of epilepsy, any major
circulatory, orthopaedic, neurological or respiratory conditions,
recent or current pregnancies, previous surgery to the abdomen
or back, or if they had undertaken any form of abdominal
exercises in the preceding 12 months. Hand dominance was
determined by the Edinburgh Handedness Inventory (Oldfield,
1971). The study conformed to the Declaration of Helsinki and
was approved by the Institutional Medical Research Ethics Committee.
2.2. Electromyography
Electromyographic activity (EMG) from the deep abdominal muscle, transversus abdominis (TrA), was recorded on both
sides using intramuscular fine-wire electrodes (Teflon-coated
stainless steel wire, 75 ␮m with 1 mm of Teflon removed and tips
bent back ∼1 and ∼2 mm to form hooks). Wires were threaded
into a hypodermic needle and inserted with ultrasound guidance
(Hodges and Richardson, 1997). EMG data were preamplified
2000 times, band-pass filtered between 30 and 1000 Hz and sampled at 2000 Hz using a Power1401 Data Acquisition System and
Spike2 software (CED, UK).
2.3. Procedures
Subjects were seated comfortably in a reclined chair with
hips flexed to ∼70◦ , knees flexed to ∼45◦ , and both arms
well-supported. Magnetic stimulation of the motor cortex was
performed using a single-pulse monophasic MagStim 2002 stimulator (MagStim Company, UK). A 7 cm figure-of-eight coil
was used with cross-over position above respective scalp sites
and the coil handle positioned at ∼45◦ from the mid-sagittal
plane such that the induced current flowed in an anteromedial
133
direction (Sakai et al., 1997). At the midline, stimulation of the
motor cortex was performed using two coil orientations: one
with induced current flowing towards the left hemisphere, and
the other with induced currents towards the right hemisphere.
MEPs from stimulation at the midline were averaged across
both coil orientations. The figure-of-eight coil provides better
focality of stimulation compared to the standard circular coil
(Brasil-Neto et al., 1992; Cohen et al., 1990), and has been
shown to evoke consistent responses from the abdominal muscles during submaximal voluntary contractions (Tunstill et al.,
2001). However, pilot trials showed that ipsilateral responses
during voluntary contractions and contralateral responses from
relaxed TrA were often difficult to evoke with a figure-of-eight
coil, even at maximum stimulator output. Thus, a double-cone
coil (MagStim Company, UK) was also used as it produces a
stronger magnetic field and can evoke consistent ipsilateral and
contralateral responses in the trunk and lower limb muscles at
rest and during voluntary contractions. This coil was positioned
with the coil handle perpendicular to the scalp site and induced
current flowing in an anterior direction.
To determine targets for voluntary contraction of TrA, EMG
activity was recorded during maximum voluntary contractions
(MVC). Subjects performed a forced expiratory manoeuvre with
verbal encouragement (Ninane et al., 1992). No pain or discomfort was reported by any subject during MVC. Three repetitions
were completed with 1 min rest between each. The highest rootmean-square (RMS) EMG amplitude over a 1 s interval was
recorded (RMSmax ). The target for voluntary contraction of TrA
was set at 10% RMSmax , and visual feedback was provided to the
subject via a computer screen. Animal and human experiments
suggest that normal respiratory drive can modulate excitability
of the abdominal motoneurons (Gill and Kuno, 1963; Hodges
et al., 1997; Sears, 1964) and thus would be expected to influence the amplitude of MEPs (Lissens et al., 1995). To control
for modulation in respiratory drive, subjects kept their glottis
opened and ceased breathing at the end of normal expiration
prior to stimulation. This was monitored online by measurement of rib cage displacement using a pressure cuff strapped to
the chest (Hodges et al., 1997).
A tight-fitting elastic cap was worn and the vertex was located
using the International 10/20 system (Jasper, 1958). The optimal
location for responses contralateral to the stimulated cortex (i.e.
scalp site which induced the largest abdominal response) was
determined using the figure-of-eight coil set at suprathreshold
intensity (∼70–100% maximum stimulator output) during 10%
voluntary contraction of TrA. In most subjects, this was found to
be 2 cm anterior and lateral to the vertex and is consistent with
the location used by previous studies for the obliquus internus
and externus abdominis muscles, although the exact location of
this position had not been systematically evaluated (Fujiwara et
al., 2001; Strutton et al., 2004).
At the optimal location, the following motor thresholds (MT)
were identified for the TrA contralateral to the stimulated cortex:
active MT using the figure-of-eight coil (cMT8 ); active MT using
the double-cone coil (cMT2 ); resting MT using the double-cone
coil (RMT2 ). In addition, the active MT for responses ipsilateral
to the stimulated hemisphere using the double-cone coil was
134
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
also identified (iMT2 ). To ensure the responses ipsilateral to
the stimulated motor cortex were representative of uncrossed
corticospinal projections, the latency of MEPs, which should be
∼2–5 ms slower than contralateral responses (Fujiwara et al.,
2001; Strutton et al., 2004; Tunstill et al., 2001), was verified
visually. All MTs were defined as the minimum intensity that
elicited five consecutive abdominal MEPs which were clearly
discernible from background EMG activity (Mills and Nithi,
1997).
To determine the potential for concurrent stimulation of both
motor cortices, TMS was delivered at 1 cm intervals along a line
2 cm anterior to the vertex and parallel with the coronal plane
from the midline to 5 cm lateral. Ten stimuli were delivered ∼5 s
apart at each site during 10% voluntary contraction (120% cMT8
using the figure-of-eight coil) and when TrA was relaxed (120%
cMT2 using the double-cone coil). For the figure-of-eight coil, if
120% of cMT8 exceeded the maximum stimulator output of the
TMS, the stimulus intensity was set to the maximum stimulator
output. For the double cone coil, the maximum intensity used
in this experiment was 90% of the maximum stimulator output.
This limit was set because 100% maximum stimulator output
was poorly tolerated by our naı̈ve subject group who were not
experienced with this type of procedure. Thus, if 120% of cMT2
exceeded 90% of maximum stimulator output of the TMS with
the double cone coil, the stimulus intensity was set to 90% of
the maximum stimulator output. The procedures were repeated
for each hemisphere.
TrA) using repeated-measures analysis of variance (ANOVA).
Due to differences in coil properties between the figure-of-eight
and double-cone coils, the MTs identified using the figure of
eight coil (cMT8 ) were not compared directly with MTs from
the double-cone coil (as this analysis would simply indicate differences in coil properties and not differences in excitability
of corticospinal projections to the abdominal responses). Thus
cMT8 were compared between the left and right TrA using a
Student’s t-test. To evaluate the difference in MEPs with stimulation at different scalp sites, the onset and amplitude of MEPs
were compared between Muscles (left and right TrA), Positions
(scalp sites), and Activities (resting versus contraction) using
repeated-measures ANOVA. Previous studies showed that the
onset and duration of silent periods can vary between scalp
sites (Wassermann et al., 1993). As silent periods were not
induced consistently over all scalps sites, the onset and duration of contralateral and ipsilateral silent periods were unable
to be compared across scalp sites. Nevertheless, the onset and
duration of the silent period evoked at the optimal location for
the muscle contralateral and ipsilateral to the stimulated cortex
during trials that involved voluntary contractions were compared
using a Student’s t-test. Post hoc testing was performed using
Duncan’s multiple range test. Significance was set at p < 0.05.
2.4. Data analysis
Fig. 1 shows group data of each MT for the left and right
TrA. MTs could be identified over at least one motor cortex
in all subjects at the optimal site. In five subjects, contralateral
Data were analysed using Matlab 7 (The Mathwork, USA).
Individual trials were full-wave rectified, averaged across trials for each scalp site, and the average background EMG was
subtracted (55–5 ms prior to stimulation). For trials in the resting condition, the onset and offset of MEPs relative to the time
of the stimulus were visually identified. For trials that involved
voluntary contractions, the onset and offset of MEPs (onset of
silent period), and the duration of the silent period were visually identified. To minimise bias during visual identification of
temporal parameters, data were presented in random order without reference to the identity of the muscle or site of magnetic
stimulation. As recordings were made using relatively selective
intramuscular electrodes, the peak-to-peak amplitude is variable due to inter-trial differences in the motor units excited by
the descending volley. Thus, the amplitude of TrA MEPs from
onset to offset was measured using RMS EMG. The RMS EMG
was normalised to the peak response for each muscle across all
scalp sites. This allowed comparison of the amplitude of MEPs
between different scalp sites for the TrA muscle on the left and
right side.
2.5. Statistical analysis
Statistical analysis was performed using Statistica 7 (Statsoft,
USA). The MTs at the optimal location using the double-cone
coil were compared between stimulus paradigms (Condition:
cMT2 , iMT2 and RMT) and between Muscles (left and right
3. Results
3.1. Motor threshold (MT)
Fig. 1. Motor thresholds (MT) for left and right transversus abdominis (TrA).
Values are expressed as percentage of maximum stimulator output. The lowest
MT was found for contralateral TrA during voluntary contraction using the
double-cone coil (cMT2 ). No significant difference was found for each MT
between the left and right TrA, although individual data shows variations in MT
between sides. (cMT8 : active MT for contralateral responses using figure-ofeight coil; iMT2 : active MT for ipsilateral responses using double-cone coil;
RMT2 : resting MT for contralateral responses using double-cone coil; *p < 0.05
for comparisons between MT to stimulation using the double-cone coil).
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
responses could not be evoked over one hemisphere using the
figure-of-eight coil, even at maximum stimulator output, and
in four subjects, ipsilateral responses could not be evoked over
one side using the double-cone coil, even at 90% maximum
stimulator output.
For the double-cone coil, MTs at the optimal location were
significantly different between stimulus paradigms (main effect:
condition, p < 0.001). The lowest MT was observed for active
MT of the contralateral responses (cMT2 ; post hoc, p < 0.001).
The resting MT (RMT2 ) was significantly lower than the active
MT for ipsilateral responses (iMT2 ; p = 0.021). No differences
in MTs were found between the left and right TrA for any
condition using the double-cone coil (main effect for muscle,
p = 0.84) or for cMT8 identified using the figure-of-eight coil
(p = 0.99). Observation of individual data showed greater differences in iMT2 between the left and right side compared to
MTs for contralateral responses, although the hemisphere with
greater MT was not consistently found on the dominant or nondominant side. The absolute difference of each MT between the
left and right TrA in each individual was calculated and comparisons showed that the difference in iMT2 between sides was
greater than the difference in MTs between sides for contralateral
responses (active and relaxed, p < 0.019; Fig. 2).
3.2. Motor evoked potential (MEP) across scalp sites
Fig. 3 shows the onset and amplitude of TrA MEPs at rest
(Fig. 3A) and during 10% MVC (Fig. 3B). The onset of TrA
MEPs was shorter during voluntary contraction than TrA MEPs
when the muscle was relaxed (main effect: activity, p < 0.001).
When stimulation was applied at the vertex and 1 cm lateral to the
midline, onsets of MEPs were similar between left and right TrA
(interaction—position × muscle: p = 0.001; post hoc: p > 0.14
135
Fig. 2. Absolute difference in transcranial magnetic stimulation (TMS) motor
threshold (MT) between the left and right transversus abdominis (TrA). Difference in MT was the greatest for ipsilateral responses in TrA (iMT2 ). (cMT8 :
active MT for contralateral responses using figure-of-eight coil; cMT2 : active
MT for contralateral responses using double-cone coil; RMT2 : resting MT for
contralateral responses using double-cone coil; *p < 0.05).
for all comparisons of MEPs (at midline and 1 cm) between
left and right TrA), and equivalent to the latency of contralateral MEPs at sites further lateral from the midline (post hoc:
p > 0.09 for all comparisons of MEPs evoked at and 1 cm lateral
to the midline with those evoked at sites 2–5 cm lateral). This
suggests that stimulation at the midline or 1 cm lateral causes
activation of contralateral corticospinal pathways, bilaterally.
However, the onsets of contralateral MEPs were significantly
shorter than ipsilateral MEPs when stimulation occurred at sites
at least 2 cm lateral to the vertex and suggests no excitation
Fig. 3. Amplitude and onset of motor evoked potential (MEP) for the left and right transversus abdominis (TrA) at various scalp sites lateral from the midline (Mid)
at rest (A) and during 10% voluntary contraction (B). Amplitude is expressed as root-mean-square (RMS) EMG of MEP, normalised to the peak response. When
stimulation evoked bilateral responses, the amplitude of contralateral MEPs was significantly larger than ipsilateral MEPs except at midline. In addition, stimulation
at sites 2–5 cm lateral to midline evoked contralateral MEPs with significantly shorter onset latency than that for ipsilateral MEPs. This difference was not observed
when stimulation was delivered at midline or 1 cm lateral to midline (*p < 0.05).
136
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
Fig. 4. Rectified average motor evoked potentials (MEPs) from a subject during TMS over the midline (A) and at 2 cm lateral (B) at rest (right (R) transversus
abdominis (TrA) EMG calibration: 0.2 mV; left (L) TrA EMG calibration: 0.1 mV). Corticospinal pathways activated by TMS are also diagrammatically represented.
When TMS was delivered at midline, the latency of TrA MEPs (arrow) was similar between the left and right sides. This suggests activation of crossed corticospinal
pathways from both motor cortices (represented as solid and dotted lines on the right hand diagram). When stimulation was delivered at 2 cm lateral to the midline
over the left hemisphere, the latency of the MEP on the right side was faster than that on the left side. This suggests activation of crossed and uncrossed corticospinal
pathways from the left motor cortex (solid and dashed lines, respectively, on the right hand diagram).
of fast contralateral pathways from the opposite motor cortex
(all p < 0.045). Representative data from a single subject are
presented in Fig. 4.
The amplitude of TrA MEPs was significantly greater during
voluntary contraction compared to MEPs at rest (main effect for
activity, p = 0.0036). Furthermore, for stimulation of scalp sites
that evoked bilateral abdominal responses (except at the vertex
(p = 0.33)), the amplitude of contralateral MEPs was larger than
MEPs on the ipsilateral side (interaction—position × muscle:
p < 0.001; post hoc: p < 0.046). However as the MT for responses
on the ipsilateral side were higher than those on the contralateral side, it is not possible to directly compare the amplitude of
MEPs between sides due to optimisation of TMS procedures for
contralateral responses.
As stimulation across scalp sites was optimised to contralateral MT, ipsilateral MEPs were not consistently evoked in all
subjects. Inspection of individual data showed that a silent period
on the ipsilateral side was present in some trials in the absence of
observable excitation. This was most commonly observed when
TMS was delivered at sites 3–5 cm lateral to the midline. Analysis for responses evoked at the optimal location showed that the
duration of silent period was longer in the muscles contralateral to the stimulated hemisphere (73.5 ± 28.2 ms) compared to
those ipsilateral to the stimulated hemisphere (37.0 ± 16.8 ms;
p = 0.018). However, this is again likely to be due to optimisation of magnetic stimulation to the contralateral MT. No
differences in onset of silent period were found at the optimal location between responses contralateral (39.7 ± 9.4 ms)
and ipsilateral to the stimulated motor cortex (38.2 ± 11.8.9 ms;
p = 0.71).
4. Discussion
This study showed that spread of current to excite the corticospinal pathways in the opposite motor cortex is possible
when TMS is delivered medially near the vertex. However,
magnetic stimulation of scalp sites at least 2 cm lateral to
the vertex evokes responses in the contralateral abdominal
muscle that were faster than those in the ipsilateral muscle.
This is consistent with activation of crossed and uncrossed
corticospinal pathways from the same motor cortex without
concurrent excitation of fast corticospinal projections from the
opposite motor cortex. These results demonstrate that selective activation of the abdominal muscles over one motor cortex
is best achieved if TMS is applied at least 2 cm lateral to
the vertex. This can be highlighted through analysis of timing of MEPs, which can provide insight into the origin of
responses.
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
4.1. Concurrent stimulation of corticospinal projections to
the abdominal muscles from both motor cortices
This is the first study to systematically demonstrate that the
largest MEPs of an abdominal muscle, TrA, can be evoked 2 cm
lateral to the vertex and that these responses can be recorded
with intramuscular EMG electrodes. Stimulation delivered at
least 2 cm lateral to midline induced MEPs bilaterally in TrA
with distinct differences in latency. The onsets of contralateral
responses were found to be ∼3–4 ms shorter than MEPs on
the ipsilateral side. This difference in latency is consistent with
previous studies that used surface EMG recordings of superficial abdominal muscles (Fujiwara et al., 2001; Strutton et
al., 2004; Tunstill et al., 2001). The size of the difference in
MEP latencies between contralateral and ipsilateral responses
also agrees with data for other axial muscles (e.g., pectoralis
major (Quartarone et al., 1999)) and limb muscles (Chen et
al., 2003; Colebatch et al., 1990; Wassermann et al., 1991;
Ziemann et al., 1999). These findings suggest stimulation of
scalp sites at least 2 cm lateral to the midline excites corticospinal pathways from the underlying motor cortex. That is,
the faster crossed corticospinal tract (Gandevia and Plassman,
1988) and the slower uncrossed polysynaptic corticospinal tracts
(Chen et al., 2003; Ziemann et al., 1999), without concurrent
stimulation of contralateral projections from the opposite motor
cortex.
When TMS was delivered near the midline, there was no
difference in onset between MEPs on the right and left sides.
Although it is likely that slower uncrossed corticospinal pathways are also excited and contribute to the later part of the
recorded MEP, similarities in latencies between MEPs on both
sides is consistent with excitation of faster crossed corticospinal
pathways from both motor cortices. These findings are consistent
with an earlier study that evoked responses in the rectus abdominis muscles using TMS over the midline and also showed no
difference in latency of MEPs between sides (Carr et al., 1994).
Although these authors argued that these responses originated
from one motor cortex as stimulation of scalp sites further lateral
to the midline also elicited bilateral responses, a lack of difference in the latency of abdominal responses between sides in that
study would suggest that concurrent stimulation of both motor
cortices had occurred.
Interestingly, the average latency for ipsilateral responses
evoked at 1 cm lateral to the midline fell between the latency
for responses evoked from crossed and uncrossed corticospinal
pathways. Observation of individual data showed that in
some subjects, TMS delivered at 1 cm lateral to the midline
induced faster contralateral MEPs and slower ipsilateral MEPs
suggesting activation of crossed and uncrossed corticospinal
projections, whereas other subjects showed similar latencies
of MEPs between sides when stimulation was delivered 1 cm
lateral to the midline. This finding highlights that it is possible to excite motor cortex on one side at 1 cm lateral to
the midline in some subjects, but not in others. In addition,
abdominal MEPs were smaller in amplitude and slower in
latency when stimulation was delivered at rest compared to
trials that involved voluntary contractions. This is consistent
137
with facilitation of evoked responses reported in earlier studies
(Hess et al., 1986, 1987; Kischka et al., 1993), and is argued
to occur at least in part to a greater probability of firing of
motoneurons due to increased excitability (Thompson et al.,
1991).
As the duration of silent period depends upon stimulus intensity (Triggs et al., 1993) and our procedures were optimised for
contralateral MEPs, differences observed in duration of silent
period between muscles contralateral and ipsilateral to the stimulated cortex could be explained by higher MT for ipsilateral
responses. Differences in duration of silent period could also be
attributed to difference in underlying mechanisms. Silent period
evoked for muscles on the contralateral side of stimulation are
argued to occur through spinal (Fuhr et al., 1991) and cortical
mechanisms (Davey et al., 1994). In contrast, ipsilateral silent
periods are argued to be mediated at least partly by transcallosal
pathways that project to and inhibit the opposite hemisphere
(Ferbert et al., 1992; Rothwell et al., 1991). Furthermore, as
the MT for inhibition of muscle activity is less than that for
excitation, it could be possible that silent periods in the ipsilateral muscle may have resulted from current spread to the
opposite hemisphere, which was sufficient to inhibit, but not
excite the ipsilateral abdominal muscles. Further investigations
are needed.
4.2. Greater hemispheric asymmetry in motor threshold
(MT) for ipsilateral motor evoked potentials (MEPs)
An interesting finding was that absolute differences in MT
between the left and right TrA was greater for longer latency
responses elicited by stimulation of the ipsilateral motor cortex
compared to the faster crossed pathways from the contralateral
motor cortex. This greater asymmetry in MT for ipsilateral corticospinal projections is unlikely to be related to variations in
motor unit thresholds between sides, as this would also affected
contralateral responses. Asymmetry in ipsilateral MT is consistent with previous studies in other abdominal (Strutton et al.,
2004) and axial muscles (Hamdy et al., 1996; MacKinnon et
al., 2004; Quartarone et al., 1999). As all our subjects were
right-handed, the variability of the side with the highest threshold confirms that this asymmetry is unlikely to be related to
handedness (MacKinnon et al., 2004).
Differences in excitability of ipsilateral projections between
sides could reflect an asymmetry in motor cortical control of
the abdominal muscles. It is unlikely that this asymmetry is
related to use-dependent plasticity with one abdominal muscle activated more than the muscle on the opposite side, as
this should also be associated with asymmetry in excitability
of contralateral corticospinal projections. Instead, these findings could be interpreted to suggest that activation of the
abdominal muscles is controlled through descending contralateral and ipsilateral corticospinal projections from predominantly
one hemisphere. Control of the trunk muscles in this manner
could be argued to be advantageous for midline muscles as
it may simplify coordination of muscles that are commonly
activated bilaterally during function. This hypothesis is supported by previous studies that showed greater synchronisation
138
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
of motor unit firing between bilateral axial muscles (Carr et
al., 1994; Marsden et al., 1999). Motor unit synchronisation
has been argued to be mediated by common drive to separate
motoneuron pools, and this could be simplified by inputs from
a single hemisphere. Bilateral drive to the trunk muscles could
also explain the observation that patients with unilateral stroke
often have deficits in activation of the trunk muscles on both
sides (Dickstein et al., 1999, 2004). However, this observation is complicated by compensatory increases in excitability
of descending corticospinal projections from the unaffected
motor cortex (Fujiwara et al., 2001). Further work is needed
to examine the functional relevance of asymmetrical ipsilateral
corticospinal projections.
The results suggest that contralateral and ipsilateral corticospinal projections to the abdominal muscles can be
discriminated when stimulation is applied at sites at least 2 cm
lateral to the midline. TMS delivered in this manner minimises
concurrent excitation of corticospinal projections from the motor
cortex contralateral to the side of stimulation.
Acknowledgments
Financial support was provided by the National Health and
Medical Research Council of Australia and the Physiotherapy
Research Foundation.
References
Brasil-Neto JP, Cohen LG, Panizza M, Nilsson J, Roth BJ, Hallett M. Optimal
focal transcranial magnetic activation of the human motor cortex: effects of
coil orientation, shape of the induced current pulse and stimulus intensity. J
Clin Neurophysiol 1992;9:132–6.
Carr LJ, Harrison LM, Stephens JA. Evidence for bilateral innervation of certain
homologous motoneurone pools in man. J Physiol (Lond) 1994;475:217–
27.
Chen R, Yung D, Li J-Y. Organization of ipsilateral excitatory and inhibitory
pathways in the human motor cortex. J Neurophysiol 2003;89:1256–
64.
Cohen LG, Roth BJ, Nilsson J, Dang N, Panizza M, Bandinelli S, et al. Effects
of coil design on delivery of focal magnetic stimulation. Technical considerations. Electroencephalogr Clin Neurophysiol 1990;75:350–7.
Colebatch JG, Rothwell JC, Day BL, Thompson PD, Marsden CD. Cortical outflow to proximal arm muscles in man. Brain 1990;113:1843–
56.
Davey NJ, Romaiguère P, Maskill DW, Ellaway PH. Supression of voluntary
motor activity revealed using transcranial magnetic stimulation of the motor
cortex in man. J Physiol (Lond) 1994;447:223–35.
Dickstein R, Heffes Y, Laufer Y, Ben-Haim Z. Activation of selected trunk
muscles during symmetric functional activities in poststroke hemiparetic
and hemiplegic patients. J Neurol Neurosurg Psychiatry 1999;66:218–21.
Dickstein R, Shefi S, Marcovitz E, Villa Y. Electromyographic activity of voluntarily activated trunk flexor and extensor muscles in post-stroke hemiparetic
subjects. Clin Neurophysiol 2004;115:790–6.
Ferbert A, Priori A, Rothwell JC, Colebatch J, Marsden CD. Transcallosal
inhibition of the human motor cortex. J Physiol (Lond) 1992;453:525–
46.
Fuhr P, Agostino R, Hallett M. Spinal motor neuron excitability during the
silent period after cortical stimulation. Electroencephalogr Clin Neurophysiol 1991;81:257–62.
Fujiwara T, Sonoda S, Okajima Y, Chino N. The relationships between trunk
function and the findings of transcranial magnetic stimulation among patients
with stroke. J Rehab Med 2001;33:249–55.
Gandevia SC, Plassman BL. Responses in human intercostal and truncal muscles to motor cortical and spinal stimulation. Respir Physiol 1988;73:325–
38.
Gill PK, Kuno M. Excitatory and inhibitory actions on phrenic motoneurones.
J Physiol (Lond) 1963;168:274–89.
Hamdy S, Aziz Q, Rothwell JC, Singh KD, Barlow J, Hughes DG, et al. The
cortical topography of human swallowing musculature in health and disease.
Nat Med 1996;2:1217–24.
Hess CW, Mills KR, Murray NM. Magnetic stimulation of the human brain:
facilitation of motor responses by voluntary contraction of ipsilateral and
contralateral muscles with additional observations on an amputee. Neurosci
Lett 1986;71:235–40.
Hess CW, Mills KR, Murray NM. Responses in small hand muscles from magnetic stimulation of the human brain. J Physiol (Lond) 1987;388:397–419.
Hodges PW, Gandevia SC, Richardson CA. Contraction of specific abdominal
muscles in postural tasks are affected by respiratory maneuvers. J Appl
Physiol 1997;83:753–60.
Hodges PW, Richardson CA. Feedforward contraction of transversus abdominis is not influenced by the direction of arm movement. Exp Brain Res
1997;114:362–70.
Jasper HH. The ten–twenty electrode system of the international federation.
Electroencephalogr Clin Neurophysiol 1958;10:370–5.
Kischka U, Fajfr R, Fellenberg T, Hess CW. Facilitation of motor evoked potentials from magnetic brain stimulation in man: a comparative study of different
target muscles. J Clin Neurophysiol 1993;10:505–12.
Lissens MA, De Muynck MC, Decleir AM, Vanderstraeten GG. Motor evoked
potentials of the abdominal muscles elicited through magnetic transcranial
brain stimulation. Muscle Nerve 1995;18:1353–4.
MacKinnon CD, Quartarone A, Rothwell JC. Inter-hemispheric asymmetry of
ipsilateral corticofugal projections to proximal muscles in humans. Exp
Brain Res 2004;157:225–33.
Marsden JF, Farmer SF, Halliday DM, Rosenberg JR, Brown P. The unilateral
and bilateral control of motor unit pairs in the first dorsal interosseous and
paraspinal muscles in man. J Physiol (Lond) 1999;521:553–64.
Mills KR, Nithi KA. Corticomotor threshold to magnetic stimulation: normal
values and repeatability. Muscle Nerve 1997;20:570–6.
Ninane V, Rypens F, Yernault JC, De Troyer A. Abdominal muscle use during
breathing in patients with chronic airflow obstruction. Am Rev Respir Dis
1992;146:16–21.
Oldfield RC. The assessment and analysis of handedness: the Edinburgh Inventory. Neuropsychologia 1971;9:97–113.
Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studies by electrical stimulation. Brain 1937;60:389–
443.
Plassman BL, Gandevia SC. Comparison of human motor cortical projections
to abdominal muscles and intrinsic muscles of the hand. Exp Brain Res
1989;78:301–8.
Quartarone A, MacKinnon C, Rothwell JC. Ipsilateral EMG responses in pectoralis major muscle evoked by transcranial magnetic stimulation over the
motor cortex. J Physiol (Lond) 1999;520P:74P.
Rothwell JC, Colebatch J, Britton TC, Priori A, Thompson PD, Day BL, et al.
Physiological studies in a patient with mirror movements and agenesis of
the corpus callosum. J Physiol (Lond) 1991;438:34P.
Sakai K, Ugawa Y, Terao Y, Hanajima R, Furubayashi T, Kanazawa I. Preferential activation of different I waves by transcranial magnetic stimulation with
a figure-of-eight-shaped coil. Exp Brain Res 1997;113:24–32.
Sears TA. The slow potentials of thoracic respiratory motorneurones and their
relation to breathing. J Physiol (Lond) 1964;175:404–24.
Strutton PH, Beith I, Theodorou S, Catley M, McGregor AH, Davey NJ. Corticospinal activation of internal oblique muscles has a strong ipsilateral
component and can be lateralized in man. Exp Brain Res 2004;158:474–9.
Thompson PD, Day BL, Rothwell JC, Dressler D, Maertens de Noordhout A,
Marsden CD. Further observations on the facilitation of muscle responses
to cortical stimulation by voluntary contraction. Electroencephalogr Clin
Neurophysiol 1991;81:397–402.
Triggs WJ, Cros D, Macdonell RAL, Chiappa KH, Fang J, Day BJ. Cortical and
spinal motor excitability during the transcranial magnetic stimulation silent
period in humans. Brain Res 1993;628:39–48.
H. Tsao et al. / Journal of Neuroscience Methods 171 (2008) 132–139
Tunstill SA, Wynn-Davies AC, Nowicky AV, McGregor AH, Davey N. Corticosinal facilitation studied during voluntary contraction of human abdominal
muscles. Exp Physiol 2001;86:131–6.
Wassermann EM, Fuhr P, Cohen LG, Hallett M. Effects of transcranial magnetic
stimulation on ipsilateral muscles. Neurology 1991;41:1795.
Wassermann EM, Pascual-Leone A, Valls-Sole J, Toro C, Cohen LG, Hallett M.
Topography of the inhibitory and excitatory responses to transcranial mag-
139
netic stimulation in a hand muscle. Electroencephalogr Clin Neurophysiol
Evok Potent 1993;89:424–33.
Ziemann U, Ishii K, Borgheresi A, Yaseen Z, Battaglia F, Hallett M, et al.
Dissociation of the pathways mediating ipsilateral and contralateral motorevoked potentials in human hand and arm muscles. J Physiol (Lond)
1999;518:895–906.