This article has been
cited by other articles in PMC.
Abstract
In
brain injury rehabilitation, passive movement exercises are frequently
used to maintain or improve mobility and range of motion. They can also
induce beneficial and sustained neuroplastic changes. Neuroimaging
studies have revealed that passive movements without motor commands
activate not only the primary somatosensory cortex but also the primary
motor cortex, supplementary motor area, and posterior parietal cortex as
well as the secondary somatosensory cortex (S2) in healthy subjects.
Repetitive passive movement has also been reported to induce increases
or decreases in cortical excitability. In this review, we focused on the
following: cortical activity following passive movement; cortical
excitability during passive movement; and changes in cortical
excitability after repetitive passive movement.
Keywords: Passive movement, Cortical excitability, Magnetoencephalography, Transcranial magnetic stimulation
Various cortical imaging techniques measure brain
activity, including functional magnetic resonance imaging (fMRI),
positron emission tomography (PET), near-infrared spectroscopy (NIRS),
electroencephalography (EEG), magnetoencephalography (MEG), and
transcranial magnetic stimulation (TMS). Because EEG and MEG have
excellent temporal resolution compared to fMRI, PET, and NIRS, they have
been used to analyze the temporal aspects of cortical sensorimotor
information processing; MEG fields, in particular, provide a direct
reflection of the primary neural current, giving excellent spatial and
temporal resolution.
TMS is a noninvasive brain
stimulation technique that can be used to determine corticospinal
excitability by measuring the muscular response to stimulation (the
motor-evoked potential, MEP). Intracortical inhibition (ICI) can be
examined by the paired-TMS paradigm, by presenting a subthreshold
conditioning stimulus followed by a suprathreshold test stimulus. With a
short inter-stimulus interval (ISI) of 1-4 ms, the test responses are
inhibited, known as short interval intracortical inhibition (SICI)1).
This inhibition depends on local gamma-aminobutyric acid A (GABAa)
receptor-mediated cortical inhibition and so is used as an indicator of
GABAa-mediated circuit activity in the primary motor cortex2-6).
Depending on the ISI between the nerve stimulus and TMS pulse, the
peripheral nerve stimulation delivered prior to the TMS pulse either
inhibits the MEP or facilitates it7,8).
MEPs are suppressed by electrical stimulation with an ISI of
approximately 20 ms, which is the first cortical component of
somatosensory evoked potential following peripheral nerve stimulation;
this is referred to as short latency afferent inhibition (SAI)7,9-12).
Conversely, MEPs are facilitated by electrical stimulation with an ISI
of about 50-100 ms between the electrical stimulation and TMS pulse,
known as afferent facilitation (AF)9,12-15). Pharmacological studies have demonstrated that cholinergic and GABAergic systems both play a role in SAI16,17).
There
have been several studies of cortical activities or cortical
excitability following voluntary movement, passive movement, peripheral
nerve stimulation, mechanical stimulation, motor-point stimulation,
water immersion, and noninvasive transcranial electrical brain
stimulation. These have used fMRI, PET, NIRS, EEG, MEG, and TMS and have
involved SICI, SAI, and AF techniques. In this review, we focus on the
cortical activity and cortical excitability that arise from passive
movement.
Cortical Activation Following Passive Finger Movement
Numerous
studies have measured brain activity following passive movement using
fMRI and PET, revealing that passive movements without motor commands
activate not only the primary somatosensory cortex (S1), but also the
primary motor area (M1), supplementary motor area (SMA), posterior
parietal cortex (PPC), and bilateral secondary somatosensory areas (S2)18-23).
However, the M1 activation in response to passive movement was not
observed in patients with severe distal sensory neuropathy, suggesting
that peripheral somatosensory afferent activation contributes to M1
activation19).
However, unlike MEG, PET, and fMRI do not have sufficient temporal
resolution to elucidate the time course of activity in these cortical
areas.
Some studies have used MEG systems to investigate the somatosensory evoked fields (SEFs) that accompany passive movement24-28). For example, Xiang et al.28)
identified four SEF components, with peak latencies of 20, 46, 70, and
119 ms after the onset of passive finger movement. Several researchers
have reported that the large SEF component after passive movement was of
a long duration, with two peaks from 30 to 100 ms after the onset24-26). The equivalent current dipoles (ECDs) of these two components were located in area 3b24), area 425), and areas 3b and 426,28).
However, many MEG studies have shown no evidence of activity in the
SMA, PPC, or S2 following passive movements. In a previous study, we
recorded SEFs following active and passive finger movement with a
multiple dipole analysis system to examine the detailed time course of
cortical activity and source localizations29).
Consistent with earlier studies, the two peaks of the MEG response
associated with passive movement were recorded between 30 and 100 ms
after the onset of movement. Figure
shows the resulting isocontour maps over the left hemisphere at 34 ms,
89 ms, and 121 ms, and over the right hemisphere at 140 ms, after active
and passive movements. The earliest and second components (PM1 and PM2)
had peaks at approximately 36 and 86 ms after passive movement onset.
The ECD of PM1 was estimated to be in area 4. And the ECDs of PM2 were
estimated to be in area 4, SMA, and the PPC over the hemisphere
contralateral to the movement, and in S2 of both hemispheres (Figure ).
The peak latency of each source activity was obtained in the range
54-109 ms in the SMA, 64-114 ms in the PPC, and 84-184 ms in the S2
(Figure ).
Isocontour
maps over the left hemisphere at 34 ms, 89 ms, 121 ms, and over the
right hemisphere at 140 ms, after active (a) and passive (b) movement in
a representative subject. Red areas indicate magnetic flux exiting the
head and blue areas flux entering the head. (Onishi et al., 201329))
Equivalent
current dipoles (ECDs) following passive movement overlapped on the
inflated brain of a representative subject. In this subject, the ECDs
were estimated at the primary sensorimotor area (dipole 1),
supplementary motor area (SMA, dipole 2), posterior parietal cortex
(PPC, dipole 3), and contralateral secondary somatosensory cortex (cS2,
dipole 4). (Onishi et al., 201329))
Time
courses of source activity and the locations of sources using Brain
Electrical Source Analysis. a) Time course for each source activity in
all subjects. b) Time course of the averaged source activity of each
source. c) Schematic presentation of the locations of all the dipoles
following passive movement. Area 4 (n = 13); area 6 (supplementary motor
area, SMA, n = 12); posterior parietal cortex (PPC, n = 7);
contralateral secondary somatosensory cortex (cS2, n = 7); ipsilateral
secondary somatosensory cortex (iS2, n = 7). (Onishi et al., 201329))
Cortical Excitability During passive Movements
Several
studies have reported consistent MEP changes in response to TMS at
specific intervals following peripheral nerve electrical stimulation.
For example, SAI was observed when the ISI between median nerve
electrical stimulation and TMS was set at 20-50 ms7,9-12), whereas AF was observed with an ISI of 50-100 ms9,12-15).
It is possible that passive movement could affect M1 excitability by
activating afferent inputs according to the direction of movement.
Unlike electrical peripheral stimulation paired with TMS, there have
been no studies on SAI or AF during passive movement, so we conducted a
study to elucidate the effects of joint angle and passive movement
direction on M1 excitability30).
This showed that M1 excitability increased 125 ms after the start of
passive movements, suggesting that AF was evoked by somatosensory input
with passive movements30).
We also investigated whether M1 excitability depended on the ISI
between the passive movement and the TMS pulse and whether modulation of
M1 excitability depended on the movement velocity (40°/s and 160°/s)
and joint angle of the passive movement31). Hence, the index finger was passively moved in the adduction direction (Figure )
followed by TMS pulses in the midrange of the metacarpophalangeal joint
with ISIs of 30, 60, 90, 120, 150, 180, and 210 ms. The results showed
that MEPs were significantly facilitated at 90, 120, and 150 ms without
F-wave changes (Figure ), and that MEP facilitation depended on the movement velocity (Figure ) and joint angle (Figure ).
In addition, MEP significantly decreased at 30 ms of ISI only under
160°/s velocity. Thus, these experiments demonstrated that the
facilitation of M1 excitability by somatosensory inputs induced by
passive movements depended on the ISIs, the passive movement velocity,
and the joint angle.
The
passive movement control apparatus. a) The main device controls the
velocity and range of the movement. b) c) The secondary device produces
repetitive passive movements of the right index finger. (Sasaki, et al.,
201749))
The
change in the amplitude of motor-evoked potentials (MEPs) with the
inter-stimulus interval (ISI). The data show mean ± standard error MEP
amplitudes (n = 20). Passive movements of the index finger for the
adduction direction were performed at a velocity of 80°/s. Transcranial
magnetic stimulation (TMS) pulses followed the passive movements with
ISIs of 30, 60, 90, 120, 150, 180, and 210 ms. The TMSs were performed
for the left M1, and the MEPs were recorded when the index finger was at
the intermediate position. *p < 0.05 compared with baseline.
(Sasaki, et al., 201731))
The
effect of movement velocity on motor-evoked potential (MEP) amplitudes.
Passive movements of the index finger in the adduction direction were
performed at velocities of 40°/s and 160°/s. Inter-stimulus intervals
(ISIs) were set at 30, 90, and 150 ms, and the MEPs were recorded with
the finger in the intermediate position. Black and white circles
indicate the mean ± standard error MEP amplitudes for all subjects (n =
15) for each ISI at velocities of 40°/s and 160°/s, respectively. The
MEPs differed significantly between 40°/s and 160°/s for each ISI. At
40°/s, there was a significant increase in MEP only at an ISI of 90 ms;
but at 160°/s, the MEP was significantly lower than baseline at 30 ms
and significantly higher at 90 and 120 ms. *p < 0.05 compared with
baseline. † p < 0.05 for the difference between the two velocities.
(Sasaki, et al., 201731))
The
effect of joint angle on motor-evoked potential (MEP) amplitudes.
Passive movements of the index finger in the adduction direction was
performed at a velocity of 80°/s. Inter-stimulus intervals (ISIs) were
set at 30, 90, and 150 ms, and MEPs were recorded with the finger in the
10° abduction or 10° adduction position (shortening vs. extension of
the first dorsal interosseous muscle). Black and white circles show the
mean ± standard error MEP amplitudes for all subjects (n = 15) in the
10° abduction position and 10° adduction, respectively, for each ISI.
The MEPs were significantly different between the 10° abduction and 10°
adduction positions at 90 and 150 ms. For 10° abduction, the MEP was
significantly higher than baseline only at 90 ms; for 10° adduction, the
MEP was significantly higher at 90 and 150 ms. *p < 0.05 compared
with baseline. † p < 0.05 for the difference between the two joint
angles. (Sasaki, et al., 201731))
Cortical Excitability after Repetitive Passive Movements
In
the field of rehabilitation, repetitive voluntary movements or passive
movements are widely used to enhance muscle strength, to improve range
of motion, and to promote motor learning or motor function in patients,
such as for those who have suffered a stroke. Corticospinal excitability
temporarily declines after voluntary exercise, a phenomenon referred to
as post-exercise cortical depression (PED)32-37). A number of studies have reported that PED persisted for 20-30 min after exhaustive exercise at maximum voluntary contraction35,37,38), and that it can be induced by light repetitive voluntary movement39-43).
In our own study, we showed that PED was induced by index finger
abduction movement at 10% maximum voluntary contraction at a rate of 0.5
Hz for 10 min41), and Teo et al.42)
observed PED that persisted for 8 min after 10 s of non-exhaustive
finger flexion-extension movement. These observations demonstrate that
both exhaustive and non-exhaustive movement can induce PED.
During the period of PED, the F-wave amplitude remains stable, indicating that spinal excitability does not change35,44), whereas inhibitory cortical circuit transmission increases, as shown by increased SICI42,43).
It is therefore thought that PED results from inhibition of the
excitability of M1. In addition, PED fluctuates according to the
frequency of voluntary movement and the strength and type of muscle
contraction39,42,45).
Potential mechanisms for PED include cortical depression, decreased
neurotransmitter levels, the decreased excitability of intracortical
glutamatergic networks, and the increased excitability of inhibitory
GABAergic networks.
There have been inconsistent results for how M1 excitability changes after passive exercise, with studies reporting decreased41), increased46), and unchanged47) excitability. Miyaguchi et al.41)
reported that repetitive passive movement (RPM) of the index finger for
10 min at 0.5 Hz reduced M1 excitability, whereas Mace et al.46)
reported that RPM of the wrist for 60 min at an average frequency of
1.0 Hz increased M1 excitability, and Lotze et al. and McDonnell et al.
observed no changes in MEPs after RPM for 30 min47,48). Table
summarizes six reports of studies that investigated MEP changes
associated with passive movements. The differences in M1 excitability
between studies may have been influenced by differences in various
stimuli, such as the duration of movement, speed of movement, presence
or absence of a duty cycle of repeated movement and rest, and the degree
of active attention given to the movement by the participant.
Table 1.
Summary of studies investigating changes in motor-evoked potentials (MEP) after repetitive passive movements.
| duration (min) | number of movements | velosity (degree/sec) | joint of movement | range of movement | duty cycle | attention | MEP |
|---|
| Notes:
"Index" refers to index finger. Duty cycle + and - refer to the
presence or absence, respectively, of a duty cycle including rest
periods. Attention + and - refer to attention being given or not given,
respectively, to the joint being moved. The arrows ↑, ↓, and → indicate
increased, decreased, or unchanged MEPs, respectively. |
| Mace46) | 60 | 1800 | 120 | wrist | -45⇔45 | + | + | ↑ |
| Lotze47) | 30 | 300 | 309 | wrist | 0⇔55 | + | + | → |
| Miyaguchi41) | 10 | 300 | 20 | index | 0⇔20 | - | - | ↓ |
| Sasaki49) | 10 | 300 | 20 | index | 0⇔20 | - | - | ↓ |
| 10 | 600 | 40 | index | 0⇔20 | - | - | ↓ |
| 10 | 1800 | 120 | index | 0⇔20 | - | - | ↓ |
| 10 | 3000 | 200 | index | 0⇔20 | - | - | ↓ |
| Otsuka53) | 10 | 300 | 15 | index | -15⇔0 | - | - | ↓ |
| 10 | 300 | 15 | index | 0⇔15 | - | - | ↓ |
| 10 | 150 | 15 | index | -15⇔15 | - | - | ↓ |
| 10 | 300 | 15 | index | 15⇔30 | - | - | ↓ |
| Tsuiki56) | 10 | 600 | 40 | index | 0⇔20 | - | - | ↓ |
| 10 | 240 | 40 | index | 0⇔20 | + | - | → |
| 10 | 600 | 100 | index | 0⇔20 | + | - | ↓ |
To
clarify the factors that influence M1 excitability, we first
investigated the effect of passive movement speed on PED after 10 min of
RPM49).
We applied RPMs of different frequencies to examine whether movement
frequency contributed to the modulation of M1 excitability. The right
index finger was passively abducted and adducted for 10 min at 0.5, 1.0,
3.0, and 5.0 Hz. We confirmed movement frequency and joint angle during
RPM for all subjects using a electrogoniometer attached to the
metacarpophalangeal joint of right index finger (Figure ).
Background EMG was monitored from the right FDI muscle during RPM to
confirm passive movement (i.e., no EMG activity) (Figure ).
RPMs at 0.5 Hz and 1.0 Hz both resulted in MEPs decreased relative to
baseline for 2 min, whereas 5.0-Hz RPM reduced MEPs for 15 min; however,
3.0-Hz RPM resulted in no change in MEPs (Figure ).
No F-wave changes were observed following any RPM intervention. We then
used the paired-pulse TMS technique to investigate whether RPM
modulated the cortical inhibitory circuit. We measured SICI before and
after 1.0, 3.0, and 5.0-Hz RPM using paired-pulse TMS with an ISI of 3
ms. Both 1.0-Hz and 5.0-Hz RPM resulted in an increase in SICI compared
with baseline. These results suggest that M1 excitability decreases
after RPM in a manner that depends on the movement frequency, possibly
through frequency-dependent enhancement of the cortical inhibitory
circuit in M1. However, those experiments were not able to clarify why
the MEP depression and SICI increase were not observed specifically at
3.0-Hz RPM. We therefore investigated whether RPM affected primary
somatosensory cortex (S1) excitability50).
Somatosensory evoked potentials were recorded after 1.0, 3.0, and
5.0-Hz RPM for 10 min. Only the 3.0-Hz RPM resulted in a decrease in the
P45 component of the somatosensory evoked potentials. Because theta
burst stimulation over S1 suppresses S1 excitability and facilitates M1
excitability51,52),
it is possible that the depression in S1 excitability after the 3.0-Hz
RPM may enhance M1 excitability through the nerve fibers from S1 to M1,
thereby resulting in PED being disturbed by the 3.0-Hz RPM.
Kinematic
measurements during repetitive passive movement of the right index
finger. (A) Joint angle during 0.5 Hz-RPM, (B) Joint angle during 1.0
Hz-RPM, (C) Joint angle during 3.0 Hz-RPM, (D) Joint angle during 5.0
Hz-RPM, (E) EMGs of the FDI muscle during 5.0 Hz-RPM. (Sasaki, et al.,
201749))
The
effect of the frequency of rapid passive movement (RPM) on MEP
amplitudes. Time course of change in mean ± standard error MEP
amplitudes for all subjects (n = 15) following 0.5, 1.0, 3.0, and 5.0-Hz
RPM. * p < 0.05 compared with the pre value. (Sasaki, et al., 201749))
We next examined the influence on PED of the range of passive movement (the extension amplitude of the muscle)53).
The index finger was passively moved from 15° abduction to 15°
adduction, 15° abduction to 0°, 0° to 15° adduction, and 15° adduction
to 30° adduction, with each movement at 15°/s for 10 min. MEPs and
F-waves were measured before and after each RPM. The amplitude of MEPs
significantly decreased after all the RPMs, but the F-wave amplitude
remained stable. These results suggest that the range of passive
movement does not markedly influence the magnitude of PED.
A
previous study reported that corticospinal excitability increased after
passive movements for 60 min with a duty cycle that included 5-8 s of
rest after every 10 movements46).
In another study that used peripheral electric stimulation,
intermittent stimulation with a duty cycle of repeated stimulation and
rest at an intensity above the motor threshold resulted in significantly
increased corticospinal excitability54).
It has also been shown that corticospinal excitability significantly
decreased with continuous theta burst stimulation but increased when the
theta burst stimulation was intermittent55).
These findings suggest that continuous and intermittent interventions
with duty cycles of repeated stimulus and rest may have different
effects on corticospinal excitability. We therefore examined the effect
on M1 excitability of the presence or absence of a duty cycle for the
RPM56).
Repetitive passive abduction-adduction movements of the right index
finger from a neutral position to 20° of abduction were performed for 10
min under three conditions: 1, movement velocity of 40°/s and
continuous RPM (600 movements); 2, movement velocity of 40°/s and
intermittent RPM (240 movements), with a configured duty cycle of 4 s
on/6 s off; and 3, movement velocity of 100°/s and intermittent RPM (600
movements), with a configured duty cycle of 4 s on/6 s off. M1
excitability was significantly reduced under conditions 1 and 3 (Figure ).
These results suggest that changes in corticospinal excitability do not
depend on the presence or absence of the duty cycle but are influenced
by the number of movements.
Motor-evoked
potential (MEP) amplitudes before and after repetitive passive
movements (RPMs) under three conditions. a) Continuous RPM (600
movements); the MEP amplitude decreased significantly at Post-0 and
Post-5 compared with Pre. b) Intermittent passive movement (240
movements); there were no significant differences in MEP amplitude
between before and after the passive movements. c) Intermittent passive
movement (600 movements); the MEP amplitude decreased significantly at
Post-0 and Post-5 compared with Pre. ** p < 0.01 compared with the
pre value. * p < 0.05 compared with the pre value.
Attention
is closely related to cortical excitability. For example, during paired
associative stimulation interventions, M1 excitability increased
significantly when attention was directed to the stimulation side, but
there was no change when focusing on the other hand57).
It has also been reported that SICI decreases and M1 excitability
increased when attention was paid to the target hand during finger
movements58), repetitive TMS59), or vibration stimulation60),
but with no change in M1 excitability when there was no attention on
the hand. These findings suggested that attention to the stimulated side
during an intervention diminishes the activity of the suppressive
circuits in the cortex, thereby increasing corticospinal excitability.
We therefore investigated the influence on M1 excitability of paying
attention to passive movement and found that, when attention was paid to
the moving finger during passive movement, corticospinal excitability
increased, whereas corticospinal excitability did not change under
conditions where no attention was directed to the passive finger
movements (under review).
Conclusion
In
this review, we summarized cortical activity and excitability
associated with passive movement, focusing in particular on cortical
excitability after RPM. Corticospinal excitability is thought to be
influenced by various factors, such as the duration and velocity of
movements, the presence or absence of a duty cycle of repeated movement
and rest, and whether active attention was directed at the movement. We
therefore performed several experiments involving RPMs, which showed
that, whether or not there was a duty cycle of repeated movement and
rest, RPM resulted in a temporary decrease in cortical excitability when
no attention was paid to the passive movement, but an increase in
cortical excitability when attention was directed at the movement.
