Clin Neurophysiol. 2007 (Feb); 118 (2): 391–402
Heidi Haavik Taylor, PhD, BSc, Bernadette Murphy, PhD, DC
Human Neurophysiology and Rehabilitation Laboratory,
Department of Sport and Exercise Science,
Tamaki Campus, University of Auckland,
Private Bag 92019,
261 Morrin Road,
Glen Innes, Auckland, New Zealand.
OBJECTIVE: To study the immediate sensorimotor neurophysiological effects of cervical spine manipulation using somatosensory evoked potentials (SEPs).
METHODS: Twelve subjects with a history of reoccurring neck stiffness and/or neck pain, but no acute symptoms at the time of the study were invited to participate in the study. An additional twelve subjects participated in a passive head movement control experiment. Spinal (N11, N13) brainstem (P14) and cortical (N20, N30) SEPs to median nerve stimulation were recorded before and for 30min after a single session of cervical spine manipulation, or passive head movement.
RESULTS: There was a significant decrease in the amplitude of parietal N20 and frontal N30 SEP components following the single session of cervical spine manipulation compared to pre-manipulation baseline values. These changes lasted on average 20min following the manipulation intervention. No changes were observed in the passive head movement control condition.
CONCLUSIONS: Spinal manipulation of dysfunctional cervical joints can lead to transient cortical plastic changes, as demonstrated by attenuation of cortical somatosensory evoked responses.
SIGNIFICANCE: This study suggests that cervical spine manipulation may alter cortical somatosensory processing and sensorimotor integration. These findings may help to elucidate the mechanisms responsible for the effective relief of pain and restoration of functional ability documented following spinal manipulation treatment.
Keywords: Cervical spine manipulation; Human; Somatosensory evoked potentials; Brain plasticity; Somatosensory system; Sensorimotor integration
From the FULL TEXT Article:
Spinal manipulation is a commonly used conservative
treatment for neck, back, and pelvic pain. The effectiveness
of spinal manipulation in the treatment of acute and chronic
low back and neck pain has been well established by outcome-
based research (for review, see Hurwitz et al., 1996
and; Vernon, 1996). However, the mechanism(s) responsible
for the effective relief of pain and restoration of functional
ability after spinal manipulation are not well
understood, as there is limited evidence to date regarding
the neurophysiological effects of spinal manipulation. The
evidence to date indicates that spinal manipulation can
lead to alterations in reflex excitability (Herzog et al.,
1999; Murphy et al., 1995; Symons et al., 2000), altered
sensory processing (Zhu et al., 2000, 1993) and altered
motor excitability (Herzog et al., 1999; Dishman et al.,
2002; Suter et al., 2000).
Spinal manipulation is used therapeutically by a number
of health professionals, including physical medicine specialists,
physiotherapists, osteopaths and chiropractors.
The different professions have different terminology for
the ‘‘entity’’ or ‘‘manipulable lesion’’ that they manipulate.
This manipulable lesion may be called ‘‘vertebral (spinal)
lesion’’ by physical medical specialists or physiotherapists,
‘‘somatic dysfunction’’ or ‘‘spinal lesion’’ by osteopaths,
and ‘‘vertebral subluxation’’ or ‘‘spinal fixation’’ by
chiropractors (Leach, 1986). Joint dysfunction as discussed
in the literature ranges from experimentally induced joint
effusion (Shakespeare et al., 1985), to pathological joint
disease such as osteoarthritis (O’Connor et al., 1993), as
well as the more subtle functional alterations that are commonly
treated by manipulative therapists (Suter et al.,
1999, 2000). For the purposes of this paper, the ‘‘manipulable
lesion’’ will be referred to as an area of spinal
There is a growing body of evidence suggesting that the
presence of spinal dysfunction of various kinds has an
effect on central neural processing. For example, several
authors have suggested spinal dysfunction may lead to
altered afferent input to the CNS (Bolton and Holland,
1996, 1998; Murphy et al., 1995; Zhu et al., 1993, 2000).
Altering afferent input to the CNS is well known to lead
to plastic changes in the way that it responds to any subsequent
input (Brasil-Neto et al., 1993; Byl et al., 1997; Hallett
et al., 1999; Pascual-Leone and Torres, 1993). Neural
plastic changes take place both following increased (Byl
et al., 1997; Pascual-Leone and Torres, 1993) and
decreased (Brasil-Neto et al., 1993; Hallett et al., 1999; Ziemann
et al., 1998) afferent input.
Altered afferent input from joints can lead to both inhibition
and facilitation of neural input to related muscles.
Numerous studies of painful joints have shown arthrogenous
muscle inhibition (Hides et al., 1994; McPartland
et al., 1997; Stokes and Young, 1984). However, even painless
experimentally induced joint dysfunction (joint effusion)
has been shown to inhibit surrounding muscles
(Shakespeare et al., 1985). This altered motor control was
also shown to persist even after aspiration of the joint effusion
(Shakespeare et al., 1985). In the early 1980s, Steinmetz
et al. demonstrated that relatively short (15–30 min)
episodes of moderately intense afferent input to the spinal
reflex pathways of rats causes increases in neural excitability
that persists for several hours (Steinmetz et al., 1982,
1985). Once these facilitated areas are established, there
may be no need for ongoing afferent input to maintain
the altered output patterns. Since these early experiments,
numerous studies have shown rapid central plastic changes
after injuries and altered sensory input from the body (for
review, see Wall et al., 2002). This can explain the findings
of Shakespeare et al. (1985) of altered motor control persisting
even after aspiration of the joint effusion. This process
provides a potential explanation for altered neural
processing as a result of joint dysfunction, and a rationale
for the effects of spinal manipulation on neural processing
that have been described in the literature.
Given that spinal dysfunction would alter the balance of
afferent input to the CNS we propose that this altered afferent
input may over time lead to potential maladaptive neural
plastic changes in the CNS. We further propose that
spinal manipulation can effect this. By recording SEPs
and monitoring the peripheral nerve afferent volley, it is
possible to determine where in the somatosensory pathway
changes induced by spinal manipulation may be occurring.
Twelve subjects (five women and seven men), aged
20–53 (mean age 29.9), participated in the spinal manipulation
study. An additional twelve age-matched subjects
(4 males and 8 females), aged 21–35 (mean age
27.1 years), participated in the passive head movement
control study. The subjects were allocated into either
group in a pseudo-randomized order. It was decided
in advance that the first 12 volunteers that fit the inclusion/
exclusion criteria for the study would become the
manipulation group as this would enable the next 12
subjects to be age matched, if needed. However, both
groups were of similar ages, so no additional age-matching
was necessary. All 24 subjects agreed to have their
cervical spines manipulated and/or their head moved
by the researcher, and no subject knew which group
they were taking part in prior to their experimental session
taking place. To be included subjects could not
have a history of neurological disease.
The subjects were
required to have a history of reoccurring neck pain or
stiffness (e.g., repeatedly present during the performance
of certain tasks such as work or study). However, at the
time of the experiment all subjects were required to be
pain free. This was done in order to assess the potential
effects of joint manipulation delivered to dysfunctional
joints alone without the presence of acute pain, as the
presence of pain alone is known to induced a significant
reduction of the post-central N20–P25 complex and a
significant increase of the N18 wave (Rossi et al.,
2003). Table 1 contains the experimental subjects’
details, including their neck complaint history and
known past neck (and/or head) trauma. Informed consent
was obtained and the local ethical committee
approved the study.
Somatosensory evoked potentials
All SEP recording electrodes (7 mm Ag/AgCl Hydrospot
™ disposable adhesive electrodes from Physiometrix)
were placed according to the International Federation of
Clinical Neurophysiologists (IFCN) recommendations
(Nuwer et al., 1994). Recording electrodes were placed
on the ipsilateral Erb’s point, over the C6 spinous process
(Cv6), and 2 cm posterior to contralateral central
and frontal scalp cites C3/4 and F3/4, which will be
referred to as Cc', and Fc', respectively. All recording
electrodes were referenced to the ipsilateral earlobe.
The C6 spinous electrode was also referenced to the
anterior neck (tracheal cartilage). The Erb’s point electrode
and the central Cc' electrode were also referenced
to the contralateral shoulder, as SEP components originating
from subcortical regions are best recorded with
a non-cephalic reference (Ulas et al., 1999). A ground
electrode was attached to Fz. Stimuli consisted of electrical square wave pulses of 1 ms duration delivered
through Grass gold cup 7 mm electrodes (impedance
< 5 kΩ). The stimulating electrodes (cathode distal)
were placed over the median nerve at the wrist of the
dominant arm. Stimuli were delivered at 1.25 times
motor threshold using a constant current stimulation at
a rate of 1.98 Hz, a rate that does not lead to SEP attenuation
(Fujii et al., 1994; GarciaLarrea et al., 1992).
Sweep length was 55 ms (5 ms pre-stimulus and 50 ms
post-stimulus) and filtering bandwidth was 3–1000 Hz
(6 dB octave roll-off). A total of 500 sweeps were averaged
using a purpose written LabView 5.1® program and
the averaged waveform was displayed in an analysis panel
from which the waveforms of interest were measured
for amplitude and latency. The amplitude of the individual
SEP components was measured from their peak to
the preceding or succeeding trough according to the
IFCN guidelines (Nuwer et al., 1994). According to the
guidelines the N11 peak amplitude was measured to
the preceding positive trough and the N13 to the succeeding
positive trough (Nuwer et al., 1994). The latencies
were recorded at their maximal peak of each
component. We identified and analyzed the following
SEP components: the peripheral N9, the spinal N11
and N13, the far-field P14–N18 complex (which was analyzed
from both the frontal and parietal recording sites),
the parietal N20 (P14–N20 and N20–P27 complexes),
and the frontal N30 (P22–N30 complex).
The high velocity type of manipulation was chosen
specifically since previous research (Herzog et al., 1995)
has shown that reflex EMG activation observed following
manipulation only occurred following high-velocity
low amplitude manipulations (as compared to lower
velocity mobilizations), and would therefore be more likely
to alter afferent input to the CNS and lead to measurable SEP
Spinal manipulation intervention
This intervention consisted of spinal manipulation of
the subjects’ dysfunctional cervical joints, which was
determined by a registered chiropractor. The clinical evidence
of joint dysfunction includes tenderness to palpation
of the relevant joints, restricted intersegmental
range of motion, palpable asymmetric intervertebral muscle
tension, abnormal or blocked joint play and end-feel
of a joint, and sensorimotor changes in the upper
extremity (Fryer et al., 2004; Hestboek and Leboeuf-
Yde, 2000). The most reliable spinal-dysfunction-indicator
is tenderness with palpation of the dysfunctional joint
(Hubka and Phelan, 1994; Jull et al., 1988). Cervical
range of motion (Rheault et al., 1992; Youdas et al.,
1992) has also been shown to have good inter- and
intra-examiner reliability. For the purpose of this study
spinal dysfunction was therefore defined as the presence
of both restricted intersegmental range of motion and
tenderness to palpation of the joint at at least one cervical
spine segment. This was detected in the following
manner. The examiner, a registered chiropractor with
at least seven years of clinical experience, would passively
move the subjects’ head, while palpating and stabilizing
over the zygapophyseal joints. For each spinal segment
the head would be gently and passively moved from neutral
position to the maximal range of lateral flexion in
the coronal plane, to both the left and the right. If this
movement appeared restricted, the examiner would apply
gentle pressure to the joint, while watching for signs of
discomfort from the subject. The examiner would also
ask the subject if the pressure to the joint elicited pain.
Spinal segments that were deemed both restricted in
lateral flexion range of motion and elicited pain on
palpation were defined for the purpose of this study to
The spinal manipulations carried out in this study were
high velocity, low amplitude thrusts to the spine held in lateral
flexion, with slight rotation and slight extension. This is
a standard manipulative technique used by manipulative
physicians, physiotherapists, and chiropractors. The
mechanical properties of this type ofCNS perturbation have
been investigated, and although the actual force applied to
the subjects’ spine depends on the therapist, the patient,
and the spinal location of treatment, the general shape of
the force-time history of spinal manipulation is very consistent
(Hessell et al., 1990), and the duration of the thrust is
always less than 200 ms (for review, see Herzog, 1996
Passive head movement intervention
The passive head movement intervention was carried
out by the same chiropractor who had pre-checked the
subjects for spinal dysfunction and who performed the
spinal manipulations for the spinal manipulation experiment.
The passive head movement intervention involved
the subjects’ head being passively laterally flexed, and
slightly extended and rotated to a position that the chiropractor
would normally manipulate that person’s cervical
spine, and then return the subjects head back to
neutral position. This was repeated to both the left and
the right. However, the experimenter was particularly
careful not to put pressure on any individual cervical
segment. Loading a joint, as is done prior to spinal
manipulation, has been shown to alter paraspinal proprioceptive
firing in anesthetized cats (Pickar and Wheeler,
2001), and was therefore carefully avoided by ending
the movement prior to end-range-of-motion when
passively moving the subjects’ heads. No spinal manipulation
was performed during any passive head movement
experiment. The passive head movement experiment was
not intended to act as a sham manipulation but to act as
a physiological control for possible changes occurring
due to the cutaneous, muscular or vestibular input that
would occur with the type of passive head movement
involved in preparing a subject/patient for a cervical
All the subjects’ cervical spines were first checked by a
registered chiropractor to determine if and where their
spines would be manipulated. If the subjects were judged
to have cervical spine dysfunction the relevant information
(including detailed medical history) was then obtained. All
subjects were also screened for evidence of vertebral artery
ischemia, with their head in a position of extension, lateral
flexion and rotation, which are neck positions shown to
have the greatest mechanical stress to the contra-lateral
vertebral artery (Arnold et al., 2004). Subjects were also
screened for other contraindications for cervical manipulation,
such as recent history of trauma, known conditions
such as inflammatory or infectious arthropathies, or bone
Three pre-intervention (baseline) SEP trials were then
recorded. Following this, the subjects would either have
their heads passively moved (as described in the passive
head movement intervention section) or the subject’s cervical
spine was manipulated at the predetermined levels (as
described in the spinal manipulation intervention section).
The passive head movements and manipulations were carried
out with the subject still seated in the slightly reclined
La-Z-boy™ chair to minimize any disturbance to the
recording electrodes. The spinal segments manipulated
for each subject in the manipulation group in this study
are shown in Table 2. Immediately following either intervention
the three post-intervention SEP trials were recorded.
The first post-intervention SEP trial (trial 0–10 min)
was recorded within the first 10 min post-intervention.
The second post-intervention SEP trial (trial 10–20 min)
was recorded within 10–20 min post-intervention and the
third post-intervention SEP trial (trial 20–30 min) was
recorded within 20–30 min post-intervention
Prior to any SEP peak analysis the data files were coded
by an independent person to reduce any bias during SEP
peak amplitude and latency analysis. The first baseline SEP
trial was used to familiarise the subjects with the electrical
stimulation needed to elicit SEPs and was therefore discarded
for all subjects. The following two pre-intervention
SEP trials (trials 2 and 3) of both latencies and amplitudes
were examined with the non-parametric Wilcoxon Signed
Ranks test (as the SEP amplitude values did not follow a
Gaussian distribution) prior to combining them into one
set to ensure that there were no statistically significant differences.
Once it was determined that there were no statistically
significant differences, the two pre-intervention trials
were collapsed down into pre-intervention mean values.
The pre-intervention means and post-intervention data (trials
0–10 min, 10–20 min, and 20–30 min) were then decoded,
grouped according to intervention and transferred into
the SPSS statistical package for statistical analysis. Note
that the frontal P14–N18 and the parietal P14–N20 were
analysed post hoc at the suggestion of reviewers. Results
of those SEP components should therefore be interpreted
As the SEP amplitude values did not follow a Gaussian
distribution non-parametric analysis was carried out for
our statistical evaluation. Non-parametric analysis or variance
for repeated measures (Friedman test) was applied to
latency and amplitude values for each SEP peak. Whenever
a significant effect was found (p < 0.05), planned comparisons
(Wilcoxon signed rank test) were executed. The
planned comparisons that were made were between the
pre-intervention averaged results and the subsequent
post-intervention results. The level of significance was set
at p < 0.05. Planned comparisons were chosen instead of
post hoc analysis to minimize Type 1 error, without reducing
the sensitivity to relevant effects (Perneger, 1998). This
is also in accordance with previous SEP research (Rossi
et al., 2003; Rossi et al., 2002).
Table 1 shows the spinal manipulation subjects’ details,
including age, gender, history of cervical pain or tension,
and any known neck and/or head trauma. Table 2 shows
the spinal segments manipulated for each subject in this
Passive head movement intervention
There were no statistically significant differences
between the pre-intervention trials prior to the passive head
movement intervention. There were no significant changes
observed in either latency or amplitude of any of the SEP
peaks following passive head movement (see Fig. 1 and
Table 3). Fig. 1 depicts the percentage of change in the
averaged normalized SEP peak amplitudes ± SE before
(pre-intervention mean), and after (trials 0–10 min, 10–
20 min and 20–30 min) the single session of passive head
movements (the upper bar graphs A) and cervical spine
manipulation (the lower bar graphs B).
Spinal manipulation intervention
There were no statistically significant differences between
the pre-intervention trials prior to the spinal manipulation
intervention. Pre-intervention representational traces of
the various SEP components of a single subject are shown
in Fig. 2, with both pre-intervention trials to show reproducibility
of SEP recordings. There were no statistically significant
changes for the latency of any peak in the spinal
manipulation condition. The peripheral N9, spinal N11
and N13, and brainstem P14–N18 (measured from both
frontal and parietal recording sites) SEP peaks did not show
any significant amplitude changes following cervical spine
manipulation (see Fig. 1). However, the two cortical peaks,
parietal N20 and frontal N30, significantly decreased in
amplitude following the cervical spine manipulations (Parietal
P14–N20 Friedman non-parametric ANOVA
p = 0.025, parietal N20–P27 Friedman non-parametric
ANOVA p = 0.006 and frontal P22–N30 Friedman nonparametric
ANOVA p = 0.009), particularly during the first
20 min. The significant changes are marked with a star in
Fig. 1 and Table 3. There was variability between the subjects,
however the trend of the changes were similar for all
subjects. The parietal P14–N20, the parietal N20–P27 and
the frontal P22–N30 peak amplitudes decreased for every
single subject during the first post-manipulation trial. By
the second post-manipulation trial only one of the subjects’
parietal P14–N20 peak amplitudes returned to baseline levels
and only two were at baseline values for the third postmanipulation
trial. For parietal N20–P27 only two of the
subjects’ peak amplitudes returned to baseline levels by
the second post-manipulations trial and both remained at
baseline values for the third post-manipulation trial. On
the other hand, four of the subjects’ frontal P22–N30 peak
amplitudes had returned to baseline values by the second
post-manipulation trial, and by the third post-manipulation
trial half the subjects’ frontal P22–N30 peak amplitudes had
returned to baseline values. Table 3 contains the raw averaged
SEP data for both the spinal manipulation group
and the passive head movement group.
The greatest amplitude attenuation occurred with the
frontal P22–N30 SEP complex. It was on average over
30% attenuated during the first 10 min following the cervical
spinal manipulation (trial 0–10 min), then on average
15% attenuated during the next 10 min (trial 10–20 min)
(see Fig. 1). These changes were significant with the
planned comparisons (Wilcoxon signed ranks test) (trial
0–10 min; p = 0.002, and trial 10–20 min; p = 0.019).
Fig. 3 depicts the attenuation of the frontal P22–N30
SEP component for an individual subject. The parietal
P14–N20 complex remained significantly attenuated with
the planned comparisons (Wilcoxon signed ranks test) during
all three post-manipulation trials, 0–10 min (p = 0.012),
10–20 min (p = 0.025), and 20–30 min (p = 0.036) (see
Fig. 1). The parietal N20–P27 complex also remained significantly
attenuated with the planned comparisons (Wilcoxon
signed ranks test) during all three post-manipulation
trials, 0–10 min (p = 0.021), 10–20 min (p = 0.010), and
20–30 min (p = 0.010) (see Fig. 1). The parietal P14–N20
amplitude was attenuated by 12–18% of the pre-intervention
amplitude. The parietal N20–P27 amplitude was only
attenuated by 8–11% of the pre-intervention amplitude.
Figs. 4 and 5 depict spinal, parietal and frontal representational
traces of the various SEP components of a single
subject prior to and after both the spinal manipulation
and passive head movement interventions, respectively.
The horizontal dashed lines represent the onset, peak or
subsequent trough of the various SEP components for
the pre-intervention condition to demonstrate clearly the
attenuation of both cortical SEP components (parietal
N20 and frontal N30) following spinal manipulation. Note
no changes occurred to any of the SEP components for this
subject following the passive head movement control
The major finding in this study was that a single session
of spinal manipulation of dysfunctional joints resulted in
attenuated cortical (parietal N20 and frontal N30) evoked
responses. On average the frontal N30 changes persisted
for 20 min post-manipulation before returning to baseline
levels. The parietal N20 changes persisted for at least
30 min, i.e. during all three post-manipulation recordings.
To our knowledge, no prior study has shown persistent
changes in neither somatosensory processing nor sensorimotor
integration following spinal manipulation.
Evidence for cortical neural plastic changes following spinal manipulation
As the peripheral N9 peak, representing the afferent volley
in the brachial plexus (Cracco and Cracco, 1976; Desmedt
and Cheron, 1981; Jones, 1977), was unaltered the
changes observed in our study most likely reflect central
changes. However, the exact mechanisms of these central
changes can only be hypothesized. These plastic changes
lasted for the entire 30 min of recordings for the parietal
N20 SEP component. The frontal N30 SEP peak changes
lasted on average 20 min post-manipulation and then
returned to baseline values. Some individual subjects demonstrated
changes that persisted during all three post-manipulation
trials (for example see Fig. 3 for lasting frontal
N30 peak changes). Other subjects demonstrated changes
in only the first (0–10 min) post-manipulation trial.
Although no brainstem SEP peak amplitude changes
were observed in this study, the design of the study does
limit the ability to exclude the possibility that sub-cortical
changes did occur. It is generally agreed that although
500 sweeps may be sufficient to record reliable peripheral
Erb’s and cortical SEP potentials, far-field potentials such
as subcortical P14–N18 do generally require a higher number
of averaged sweeps (F. Mauguiere, 1999; Nuwer et al.,
1994). The possibility for sub-cortical SEP changes following
spinal manipulation does therefore need further
The parietal N20 SEP peak changes
The parietal N20 SEP component, generated in the primary
somatosensory cortex (S1) (Desmedt and Cheron,
1980; F. Mauguiere, 1999 ; Nuwer et al., 1994) was significantly
decreased during all post-manipulation blocks, both
when measured from P14 to N20 and when measured from
N20 to N27. This suggests that S1 processing was reduced
post manipulation. One possible contribution to the
reduced parietal N20 could come from enhanced active
inhibition. For example, thalamocortical afferents monosynaptically
activate an intracortical feed-forward inhibition
to area 3b pyramidal cells supposed to generate
parietal N20 (Desmedt and Cheron, 1980; F. Mauguiere,
1999; Nuwer et al., 1994) via the summation of excitatory
postsynaptic potentials. The cervical spine manipulation(s)
could have altered the afferent information originating
from the cervical spine (from joints, muscles, etc), in turn
altering the way that the 3b pyramidal cells respond to
any subsequent afferent input such as the median nerve
stimulation in our study.
The passive head movement SEP experiment demonstrated
that no significant changes occurred following a
simple movement of the subject’s head. Our results are
therefore not simply due to altered input from vestibular,
muscle or cutaneous afferents as a result of the chiropractor’s
touch or due to the actual movement of the subjects
head. This therefore suggests that the results in this study
are specific to the delivery of the high-velocity, low-amplitude
thrust to the dysfunctional joints. The passive head
movement experiment was to control for the potential neural
changes due to the afferent traffic resulting from touch
and head movement alone. It was not intended to be a
It is however also possible that subject arousal levels
differed between the two groups and this may be a confounding
variable in this study. It is for example known
that sleep can prolong the latency and alter the morphology
of the parietal N20 component of median nerve
SEPs in healthy subjects (Emerson et al., 1988). The subjects
in the manipulation group may have been anxious
about the manipulations and therefore more alert than
the subjects in the passive head movement group. As it
is not possible to perform a sham adjustment, this is
an inherent problem with this type of intervention that
is difficult to overcome. It is the authors’ opinion that
it is unlikely that the subjects in the passive head movement
control condition were merely more relaxed
throughout the study for several reasons. First of all
there were no changes observed in any SEP peak latency
in this study, and latency changes are known to occur
with alterations in arousal levels (Emerson et al., 1988).
Furthermore, the subjects’ frontal N30 SEP component
returned to baseline by the third post-manipulation trial.
It would not make sense that the subjects were anxious
prior the manipulations, then all relaxed instantly following
the manipulations, and then became anxious again
towards the end of the study.
The frontal P22–N30 SEP peak changes
The greatest amplitude change was observed with the
frontal N30 component of the SEP peaks. Although some
authors suggest this peak is also generated in the post-central
cortical regions (i.e. S1) (Allison et al., 1989b, 1991,
1989a), most evidence suggests that this peak is related to
a complex cortical and subcortical loop linking the basal
ganglia, thalamus, pre-motor areas, and primary motor
cortex (Kanovsky´ et al., 2003; Mauguiere et al., 1983; Rossini
et al., 1987, 1989; Waberski et al., 1999). The frontal
N30 peak is therefore thought to reflect sensorimotor integration
(Rossi et al., 2003). The attenuated frontal N30
SEP peak observed in our study therefore suggests that
there may be a decrease of activity in these cortical and
subcortical loops linking the basal ganglia, thalamus, pre-motor areas and primary motor cortex resulting from median
nerve stimulation, immediately following spinal manipulation,
lasting on average 20 min. Again this may be due to
altered afferent input following spinal manipulation.
Previous research has shown that the frontal N30 component
has independent cortical generators with a separate
thalamo–cortical input (Balzamo et al., 2004; Mauguiere
et al., 1983). Intracortical human recordings have also
shown afferent information following median nerve stimulation
project directly to the primary motor cortex (Balzamo
et al., 2004), as has been documented previously in
primates (Strick and Preston, 1982; Tanji and Wise,
1981). The frontal N30 component is therefore subject to
more complex inputs than those flowing on from the parietal
N20 generator alone. That the frontal N30 amplitude
attenuation was on average 20% greater than the parietal
N20 amplitude attenuation could be a reflection of the
additional inputs from the separate thalamo–cortical
The afferent mediators of the central neural effects of spinal manipulation
The current study supports previous work that suggest
muscle afferents (probably Ia) are the most likely mediators
of the central neural effects of spinal manipulation (Bolton
and Holland, 1996, 1998; Murphy et al., 1995; Pickar and
Wheeler, 2001; Zhu et al., 2000, 1993). As mentioned, the
N30 peak was the most dramatically attenuated SEP peak
in this study. Previous work has demonstrated that the
frontal N30 peak is the most vulnerable SEP component
to vibration interference (Hoshiyama and Kakigi, 2000).
Vibration of relaxed muscles, as used in Hoshiyama and
Kakigi’s study, 2000, predominantly stimulates the group
Ia fibres (from muscles spindles) (Roll et al., 1989). Group
Ib and II fibres are not sensitive to this type of stimulation
(Roll et al., 1989). This study demonstrates that the frontal
N30 component is highly sensitive to changing levels of Ia
afferent input (Hoshiyama and Kakigi, 2000). The significant
attenuation of the frontal N30 SEP component
observed in the current study thus suggests that spinal
manipulation may alter Ia afferent processing. This is consistent
with previous animal studies that have demonstrated
that displacement of vertebrae is signaled to the central
nervous system by afferent nerves arising from deep intervertebral
muscles (Bolton and Holland, 1996, 1998; Pickar
and Wheeler, 2001). In particular, both the velocity and relative
position of the vertebral displacement appeared to be
encoded by afferent nerve activity from intervertebral muscles
(Bolton and Holland, 1996, 1998).
If Ia afferents are the mediators of the central neural
effects documented following spinal manipulation of dysfunctional
joints, the persistent attenuation of the N20
component, seen in our study, may reflect a reduction of
activity at S1 following spinal manipulation. It is possible
that joint dysfunction leads to a bombardment of the
CNS with Ia afferent signaling from the surrounding intervertebral
muscles. Bolton and Holland’s (1996, 1998)
observations of increased Ia activity from the paravertebral
musculature surrounding surgically fixated vertebrae in
cats support this theory. Furthermore, Pickar and Wheeler
(2001) have demonstrated in anesthetized cats that both
spindle cells and golgi tendon organs can respond to
high-velocity, short-duration loads, i.e. loads with a
force-time profile similar to that of a load delivered during
spinal manipulation. Therefore, if spinal manipulation
reduces excessive signaling from the involved intervertebral
muscles this altered afferent input to the CNS may
change the way it responds to any subsequent input, such
as the electrical stimulation from the median nerve in this
Implications for investigations of neural plasticity and spinal manipulation
Episodes of acute pain, such as following an injury, may
initially induce plastic changes in the sensorimotor system
(for review, see Wall et al., 2002). Pain alone, without deafferentation,
has been shown to induce increased SEP peak
amplitudes (Tinazzi et al., 2000, 2004) and increased
somatosensory evoked magnetic fields (Soros et al.,
2001). As the sensorimotor disturbances are known to persist
beyond the acute episode of pain (Jull et al., 2002; Sterling
et al., 2003), and thought to play a defining role in the
clinical picture and chronicity of different chronic neck
pain conditions (Michaelson et al., 2003), then the reduced
cortical SEP peak amplitudes observed in the current study
following spinal manipulation may reflect a normalization
of such injury/pain-induced central plastic changes, which
may reflect one mechanism for the improvement of functional
ability reported following spinal manipulation.
The observations in the present study suggest that spinal
manipulation of dysfunctional joints may modify transmission
in neuronal circuitries not only at a spinal level as indicated
by previous research (Herzog et al., 1999; Murphy
et al., 1995; Symons et al., 2000), but at a cortical level,
and possibly also deeper brain structures such as the basal
ganglia. Further studies are needed to elucidate the role
and mechanisms of these cortical changes, and their relationship
to a patient’s clinical presentation.
The head author was initially supported with a fellowship
from the Foundation for Chiropractic Education
and Research, and later by a Bright Futures Top Achievers
Allison T, McCarthy G, Wood CC, Darcey TM, Spencer DD, Williamson
Human cortical potentials evoked by stimulation of the median nerve. II. Cytoarchitectonic areas generating short-latency activity.
J Neurophysiol 1989b;62(3):694–710.
Allison T, McCarthy G, Wood CC, Jones SJ.
Potentials evoked in human and monkey cerebral cortex by stimulation of the median nerve. A review of scalp and intracranial recordings.
Brain 1991;114(Pt 6):2465–503.
Allison T, McCarthy G, Wood CC, Williamson PD, Spencer DD.
Human cortical potentials evoked by stimulation of the median nerve. II. Cytoarchitectonic areas generating long-latency activity.
J Neurophysiol 1989a;62(3):11–722.
Arnold C, Bourassa R, Langer T, Stoneham G.
Doppler studies evaluating the effect of a physical therapy screening protocol on vertebral artery blood flow.
Man Ther 2004;9(1):13–21.
Balzamo E, Marquis P, Chauvel P, Regis J.
Short-latency components of evoked potentials to median nerve stimulation recorded by intracerebral electrodes in the human pre- and postcentral areas.
Clin Neurophysiol 2004;115(7):1616–23.
Bolton PS, Holland CT.
Afferent signaling of vertebral displacement in the neck of the cat.
Soc Neurosci Abstr 1996;22:1802.
Bolton PS, Holland CT.
An in vivo method for studying afferent fibre activity from cervical paravertebral tissue during vertebral motion in anaesthetised cats.
J Neurosci Methods 1998;85(2):211–8.
Brasil-Neto JP, Valls-Sole J, Pascual-Leone A, Cammarota A, Amassian
VE, Cracco R, et al.
Rapid modulation of human cortical motor outputs following ischaemic nerve block.
Brain 1993;116(Pt 3):511–25.
Byl NN, Merzenich MM, Cheung S, Bedenbaugh P, Nagarajan SS,
A primate model for studying focal dystonia and
repetitive strain injury: effects on the primary somatosensory cortex.
Phys Ther 1997;77(3):269–84.
Cracco RQ, Cracco JB.
Somatosensory evoked potential in man: far field potentials.
Electroencephalogr Clin Neurophysiol 1976;41(5):460–6.
Desmedt JE, Cheron G.
Central somatosensory conduction in man: neural generators and interpeak latencies of the far-field components recorded from neck and right or left scalp and earlobes.
Electroencephalogr Clin Neurophysiol 1980;50(5–6):382–403.
Desmedt JE, Cheron G.
Non-cephalic reference recording of early somatosensory potentials to finger stimulation in adult or aging
normal man: differentiation of widespread N18 and contralateral N20 from the prerolandic P22 and N30 components.
Electroencephalogr Clin Neurophysiol 1981;52(6):553–70.
Dishman JD, Ball KA, Burke J.
Central motor excitability changes after spinal manipulation: a transcranial magnetic stimulation study.
J Manipulative Physiol Ther 2002;25(1):1–9.
Emerson RG, Sgro JA, Pedley TA, Hauser WA.
State-dependent changes in the N20 component of the median nerve somatosensory evoked potential.
Fryer G, Morris T, Gibbons P.
Paraspinal Muscles and Intervertebral Dysfunction: Part One.
J Manipulative Physiol Ther 2004;27(4):267–74.
Fujii M, Yamada T, Aihara M, Kokubun Y, Noguchi Y, Matsubara M,
The effects of stimulus rates upon median, ulnar and radial nerve somatosensory evoked potentials.
Electroencephalogr Clin Neurophysiol 1994;92(6):518–26.
GarciaLarrea L, Bastuji H, Mauguiere F.
Unmasking of cortical SEP components by changes in stimulus rate: a topographic study.
Electroencephalogr Clin Neurophysiol 1992;84(1):71–83.
Hallett M, Chen R, Ziemann U, Cohen LG.
Reorganization in motor cortex in amputees and in normal volunteers after ischemic limb deafferentation.
Electroencephalogr Clin Neurophysiol Suppl 1999;51:183–7.
Mechanical, Physiologic, and Neuromuscular Considerations of Chiropractic treatment.
In: Lawrence DJ, Cassidy JD, McGregor M, Meeker WC, Vernon. HT, editors.
Advances in Chiropractic, vol 3.
New York: Mosby-Year Book; 1996. p. 269–85.
Herzog W, Conway PJ, Zhang YT, Gail J, Guimaraes ACS.
Reflex responses associated with manipulative treatments on the thoracic spine: a pilot study.
J Manipulative Physiol Ther 1995;18:233–4.
Herzog W, Scheele D, Conway PJ.
Electromyographic responses of back and limb muscles associated with spinal manipulative therapy.
Hessell BW, Herzog W, Conway PJ, McEwen MC.
Experimental measurement of the force exerted during spinal manipulation using the Thompson technique.
J Manipulative Physiol Ther 1990;13(8):448–53.
Hestboek L, Leboeuf-Yde C.
Are chiropractic tests for the lumbo-pelvic spine reliable and valid? A systematic critical literature review.
J Manipulative Physiol Ther 2000;23(4):258–75.
Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH.
Evidence of lumbar multifidus muscle wasting ipsilateral to symptoms in patients with acute/subacute low back pain.
Hoshiyama M, Kakigi R.
Vibratory stimulation of proximal muscles does not affect cortical components of somatosensory evoked potential following distal nerve stimulation.
Clin Neurophysiol 2000;111(9):1607–10.
Hubka MJ, Phelan SP.
Interexaminer reliability of palpation for cervical spine tenderness.
J Manipulative Physiol Ther 1994;17(9):591–5.
Hurwitz EL, Aker PD, Adams AH, Meeker WC, Shekelle PG.
Manipulation and Mobilization of the Cervical Spine:
A Systematic Review of the Literature
Spine (Phila Pa 1976) 1996 (Aug 1); 21 (15): 1746–1760
Short latency potentials recorded from the neck and scalp following median nerve stimulation in man.
Electroencephalogr Clin Neurophysiol 1977;43(6):853–63.
Jull G, Bogduk N, Marsland A.
The Accuracy of Manual Diagnosis for Cervical Zygapophysial Joint Pain Syndromes
Med J Aust. 1988 (Mar 7); 148 (5): 233–236
Jull G, Trott P, Potter H, Zito G, Niere K, Shirley D, et al.
A Randomized Controlled Trial of Exercise and Manipulative Therapy
for Cervicogenic Headache
Spine (Phila Pa 1976) 2002 (Sep 1); 27 (17): 1835—1843
Kanovsky´ P, Bare M, Rektor I.
The selective gating of the N30 cortical component of the somatosensory evoked potentials of median nerve is
different in the mesial and dorsolateral frontal cortex: evidence from intracerebral recordings.
Clin Neurophysiol 2003;114(6):981–91.
Manipulation terminology in the chiropractic, osteopathic, and medical literature.
In: The Chiropractic Theories: A Synopsis of Scientific Research.
Baltimore: Williams and Wilkins; 1986, 15-2.
Somatosensory evoked potentials: normal responses, abnormal waveforms and clinical applications in neurological diseases.
In: Niedermeyer E, editor.
Electroencephalography: basic principles, clinical applications, and related fields.
Baltimore: Williams and Wilkins; 1999.
Mauguiere F, Desmedt JE, Courjon J.
Astereognosis and dissociated loss of frontal or parietal components of somatosensory evoked potentials in hemispheric lesions. Detailed correlations with clinical signs and computerized tomographic scanning.
Brain 1983;106(Pt 2):271–311.
McPartland JM, Brodeur RR, Hallgren RC.
Chronic neck pain, standing balance, and suboccipital muscle atrophy – a pilot study.
J Manipulative Physiol Ther 1997;20(1):24–9.
Michaelson P, Michaelson M, Jaric S, Latash ML, Sjolander P,
Vertical Posture and Head Stability in Patients with Chronic Neck Pain.
J Rehab Med 2003;35(5):229–35.
Murphy BA, Dawson NJ, Slack JR.
Sacroiliac joint manipulation decreases the H-reflex.
Electromyogr Clin Neurophysiol 1995;35(2):87–94.
Nuwer MR, Aminoff M, Desmedt J, Eisen AA, Goodin D, Matsuoka S,
IFCN recommended standards for short latency somatosensory evoked potentials. Report of an IFCN committee. International Federation of Clinical Neurophysiology.
Electroencephalogr Clin Neurophysiol 1994;91(1):6–11.
O’Connor BL, Visco DM, Brandt KD, Albrecht M, O’Connor AB.
Sensory nerves only temporarily protect the unstable canine knee joint from osteoarthritis. Evidence that sensory nerves reprogram the central nervous system after cruciate ligament transection.
Arthritis Rheum 1993;36(8):1154–63.
Pascual-Leone A, Torres F.
Plasticity of the sensorimotor cortex representation of the reading finger in Braille readers.
Brain 1993;116(Pt 1):39–52.
What’s wrong with Bonferroni adjustments.
Pickar JG, Wheeler JD.
Response of muscle proprioceptors to spinal manipulative-like loads in the anesthetized cat.
J Manipulative Physiol Ther 2001;24(1):2–11.
Rheault W, Albright B, Byers C, Franta M, Johnson A, Skowronek BS,
Intertester reliability of the cervical range of motion device.
J Orthop Sports Phys Ther 1992;15(3):147–50.
Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages induced
by tendon vibration in man: a microneurographic study.
Exp Brain Res 1989;76(1):213–22.
Rossi S, della Volpe R, Ginanneschi F, Ulivelli M, Bartalini S, Spidalieri
R, et al.
Early somatosensory processing during tonic muscle pain in humans: relation to loss of proprioception and motor ‘defensive’ strategies.
Clin Neurophysiol 2003;114(7):1351–8.
Rossi S, Tecchio F, Pasqualetti P, Ulivelli M, Pizzella V, Romani GL,
Somatosensory processing during movment observation in humans.
Clin Neurophysiol 2002;113:16–24.
Rossini PM, Babiloni F, Bernardi G, Cecchi L, Johnson PB, Malentacca
A, et al.
Abnormalities of short-latency somatosensory evoked potentials in parkinsonian patients.
Electroencephalogr Clinical Neurophysiol 1989;74(4):277–89.
Rossini PM, Gigli GL, Marciani MG, Zarola F, Caramia M.
Noninvasive evaluation of input-output characteristics of sensorimotor cerebral areas in healthy humans.
Electroencephalogr Clin Neurophysiol 1987;68(2):88–100.
Shakespeare DT, Stokes M, Sherman KP, Young A.
Reflex inhibition of the quadriceps after meniscectomy: lack of association with pain.
Clin Physiol 1985;5(2):137–44.
Soros P, Knecht S, Bantel C, Imai T, Wusten R, Pantev C, et al.
Functional reorganization of the human primary somatosensory cortex after acute pain demonstrated by magnetoencephalography.
Neurosci Lett 2001;298(3):195–8.
Steinmetz JE, Beggs AL, Molea D, Patterson MM.
Long-term retention of a peripherally induced flexor reflex alteration in rats.
Brain Res 1985;327:312–5.
Steinmetz JE, Lupica CR, Secard CM, Patterson MM.
Effects of manipulating stimulation intensity and duration on fixation of a peripherally induced spinal reflex alteration in rats.
Physiol Behav 1982;29:1039–44.
Sterling M, Jull G, Vicenzino B, Kenardy J, Darnell R.
Development of motor system dysfunction following whiplash injury.
Stokes M, Young A.
The contribution of reflex inhibition to arthrogenous muscle weakness.
Clin Sci 1984;67(1):7–14.
Strick PL, Preston JB.
Two representations of the hand in area 4 of a primate.II.Somatosensory input organization.
J Neurophysiol 1982;48:150–9.
Suter E, McMorland G, Herzog W, Bray R.
Decrease in quadriceps inhibition after sacroiliac joint manipulation in patients with anterior knee pain.
J Manipulative Physiol Ther 1999;22(3):149–53.
Suter E, McMorland G, Herzog W, Bray R.
Conservative lower back treatment reduces inhibition in knee-extensor muscles: a randomized controlled trial.
J Manipulative Physiol Ther 2000;23(2):76–80.
Symons BP, Herzog W, Leonard T, Nguyen H.
Reflex responses associated with activator treatment.
J Manipulative Physiol Ther 2000;23(3):155–9.
Tanji J, Wise SP.
Submodality distribution in sensorimotor cortex of the unanesthetized monkey.
J Neurophysiol 1981;45(3):467–81.
Tinazzi M, Fiaschi A, Rosso T, Faccioli F, Grosslercher J, Aglioti SM.
Neuroplastic Changes Related to Pain Occur at Multiple Levels of the Human Somatosensory System: A Somatosensory-Evoked Potentials Study in Patients with Cervical Radicular Pain.
J Neurosci 2000;20(24):9277–83.
Tinazzi M, Valeriani M, Moretto G, Rosso T, Nicolato A, Fiaschi A,
Plastic interactions between hand and face cortical representations in patients with trigeminal neuralgia: a somatosensory-evoked potentials study.
Ulas UH, Odabasi Z, Ozdag F, Eroglu E, Vural O.
Median nerve somatosensory evoked potentials: recording with cephalic and noncephalic references.
Electromyogr Clin Neurophysiol 1999;39(8):473–7.
Spinal manipulation as a valid treatment for low back pain.
Del Med J 1996;68(3):175–8.
Waberski TD, Buchner H, Perkuhn M, Gobbele R, Wagner M, Kucker
W, et al.
N30 and the effect of explorative finger movements: a model of the contribution of the motor cortex to early somatosensory potentials.
Clin Neurophysiol 1999;110(9):1589–600.
Wall JT, Xu J, Wang X.
Human brain plasticity: an emerging view of the multiple substrates and mechanisms that cause cortical changes and related sensory dysfunctions after injuries of sensory inputs from the body.
Brain Res Rev 2002;39(2-3):181–215.
Youdas JW, Garrett TR, Suman VJ, Bogard CL, Hallman HO, Carey JR.
Normal range of motion of the cervical spine: an initial goniometric study.
Phys Ther 1992;72(11):770–80.
Zhu Y, Haldeman S, Hsieh CY, Wu P, Starr A.
Do cerebral potentials to magnetic stimulation of paraspinal muscles reflect changes in palpable muscle spasm, low back pain, and activity scores?[comment].
J Manipulative Physiol Ther 2000;23(7):458–64.
Zhu Y, Haldeman S, Starr A, Seffinger MA, Su SH.
Paraspinal muscle evoked cerebral potentials in patients with unilateral low back pain.
Ziemann U, Hallett M, Cohen LG.
Mechanisms of deafferentationinduced plasticity in human motor cortex.
J Neurosci 1998;18(17):7000–7.
Return to the CHIROPRACTIC SUBLUXATION Page
Return to the CHIROPRACTIC AND CHRONIC NECK PAIN Page