ORIGINAL ARTICLE

Brain structural changes in patients with chronic myofascial

pain
D.M. Niddam1,2, S.-H. Lee3,4, Y.-T. Su5, R.-C. Chan3,4
1 Brain Research Center, National Yang-Ming University, Taipei, Taiwan
2 Institute of Brain Science, School of Medicine, National Yang-Ming University, Taipei, Taiwan
3 Department of Physical Medicine and Rehabilitation, National Yang-Ming University, Taipei, Taiwan
4 Department of Physical Medicine and Rehabilitation, Taipei Veterans General Hospital, Taipei, Taiwan
5 Department of Physical Medicine and Rehabilitation, Far Eastern Memorial Hospital, New Taipei City, Taiwan

Correspondence Abstract
David M. Niddam
E-mail: niddam@ym.edu.tw Background: Myofascial trigger points (MTrPs) are a highly prevalent
source of musculoskeletal pain. Prolonged ongoing nociceptive input
Disclosures from MTrPs may lead to maladaptive changes in the central nervous
This work was supported by grants from the system. It remains, however, unknown whether pain from MTrPs is
National Science Council (NSC 99-2314-B-
associated with brain atrophy. In addition, stress, which may contribute
010-010-MY3) and the Ministry of Science
to the formation of MTrPs, is also known to affect brain structures. Here,
and Technology, Aim for the Top University
Plan. we address whether structural brain changes occur in patients with
chronic pain originating from MTrPs and whether such changes are
Conflict of interest related to pain or stress.
The authors declare no conflicts of interest. Methods: Voxel-based morphometry was used to compare grey-matter
(GM) volumes in 21 chronic pain patients, with MTrPs in the bilateral
Accepted for publication upper trapezius muscles, with 21 healthy controls. Hyperalgesia was
30 May 2016
assessed by pressure pain thresholds, and stress was assessed by cortisol
doi:10.1002/ejp.911
levels and anxiety questionnaires.
Results: Patients exhibited normal stress levels but lowered pain
thresholds. GM atrophy was found in dorsal and ventral prefrontal
regions in patients. The GM density of the right dorsolateral prefrontal
cortex correlated with pain thresholds in patients, i.e. the more atrophy,
the lower pain threshold. GM atrophy was also found in the anterior
hippocampus, but the atrophy was neither related to pain nor stress.
Conclusions: Patients with chronic myofascial pain exhibit GM atrophy
in regions involved in top-down pain modulation and in processing of
negative affect. The relationship between the dorsolateral prefrontal
cortex and pain thresholds suggests the presence of pain disinhibition.
No evidence was found for the involvement of stress. It remains unclear
whether the observed atrophy contributes to the development of the
chronic pain state or is caused by the ongoing nociceptive input.
Significance: Chronic myofascial pain, caused by myofascial trigger
points, is associated with localized brain atrophy in areas involved in
pain processing and modulation, among others. These findings extend
previous knowledge about peripheral and spinal changes to the
supraspinal level.

© 2016 European Pain Federation - EFICâ Eur J Pain  (2016) – 1

Brain structural changes in myofascial pain
1. Introduction
Myofascial trigger points (MTrP) are characterized by
abnormal focal contractures within the muscle that
appear in taut bands of hardened muscle. MTrPs
may be associated with various degrees of sensory
and motor dysfunctions, including spontaneous pain,
referred pain and hyperalgesia. Trigger points are
classified as being active or latent depending on the
presence or absence of spontaneous pain, respec-
tively, when the muscle is at rest (Shah et al.,
2015). Prolonged ongoing nociceptive input, such as
from active MTrPs, can lead to central sensitization,
a state where functional and structural changes
occur throughout the central nervous system (Latre-
moliere and Woolf, 2009; Mense, 2010; Shah et al.,
2015). Central sensitization is associated with allody-
nia and hyperalgesia and is a key factor in the devel-
opment and maintenance of chronic pain
(Latremoliere and Woolf, 2009). Hyperalgesia from
MTrPs has previously been shown to result in
enhanced activity in brain regions involved in the
processing of sensory-discrimination and affect (Nid-
dam et al., 2008), similar to findings in other chronic
pain conditions (Gracely et al., 2002; Verne et al.,
2003; Maihofner et al., 2005). Although such find-
ings are indicative of central changes, peripheral and
spinal contributions to the altered brain processing
cannot be ruled out, as nociceptive input is transmit-
ted through both. Chronic pain conditions are also
associated with grey-matter atrophy in brain regions
responsible for pain perception and modulation
(Cauda et al., 2014). Grey-matter atrophy may, in
part, be responsible for maladaptive changes under-
lying the chronification process. It is, however,
unknown whether myofascial pain arising from
active MTrPs is associated with structural brain
changes as well.
Emotional and physical stress has been hypothe-
sized to contribute to the formation and perpetua-
tion process of MTrPs through both peripheral and
central mechanisms (Niddam, 2009). Peripherally,
prolonged periods of stress may result in a persistent
increased muscle tone and muscle injury (Treaster
et al., 2006). Centrally, stress may target the brain
by inducing maladaptive functional and structural
changes resulting in exacerbation of the pain (McE-
wen, 2007). Indications of elevated baseline levels of
cortisol, a biological marker of stress, exist in patients
with myofascial pain, especially of temporomandibu-
lar origin (Yoshihara et al., 2005; Robinson et al.,
2006; Nadendla et al., 2014). In addition, we previ-
ously showed that the hippocampus, a key region in
2 Eur J Pain  (2016) –
D. M. Niddam et al.
the control of the stress response (McEwen, 2007),
exhibits abnormal hypoactivity in response to painful
stimulation of MTrPs (Niddam et al., 2008). The hip-
pocampus is targeted directly by cortisol and is con-
sidered to be structurally highly sensitive to stress
(McEwen, 2007). Taken together, active MTrPs may
be associated with stress-related compromised struc-
tural integrity of the hippocampus.
To address whether active MTrPs are related to
structural brain changes and if such changes are
related to stress or pain, hippocampal volumetry and
whole-brain voxel-based morphometry (VBM) were
performed in chronic myofascial pain patients with
active MTrPs in the bilateral upper trapezius muscles
and in matched healthy controls. We hypothesized
that active MTrPs were associated with (1) reduced
hippocampal grey-matter volumes; (2) grey-matter
atrophy in pain and stress-related brain regions and
(3) elevated stress levels that would be related to
structural brain changes. Basal serum cortisol levels
and psychological profiles were obtained to assess
physiological and emotional stress levels.
2. Materials and methods
2.1 Study population
Twenty-one patients with myofascial pain emanating
from active trigger points in the bilateral upper
trapezius muscles were recruited from the outpatient
clinic of Department of Physical Medicine and Reha-
bilitation, Taipei Veterans General Hospital. Patients
were evaluated by a physician experienced in diag-
nosing MTrPs. The inclusion criteria for the patients
were (1) palpable taut bands and/or hardened nod-
ules in the bilateral upper trapezius muscles; (2)
spontaneous myofascial pain emanating from a well-
localized area within the taut band/nodules; (3)
exacerbation of the pain upon palpation of the taut
band/nodules; (4) presence of spontaneous myofas-
cial pain for more than 6 month, in order to increase
the likelihood of structural brain changes; (5) no
previous treatment of the myofascial pain arising
from the taut band/nodules; (6) no intake of preven-
tive medications, such as beta-blockers, antidepres-
sants, anticonvulsants or calcium channel blockers,
within 6 month prior to scanning; and (7) no intake
of over-the-counter pain relievers, such as nons-
teroidal anti-inflammatory drugs, 1 week prior to
scanning. Twenty-one healthy volunteers with no
taut bands or nodules in the upper trapezius muscles
were enrolled in the control group (CTL). Apart from
© 2016 European Pain Federation - EFICâ
D. M. Niddam et al.
the diagnosis of the MTrPs in the patient group, all
participants were normal in physical and neurologi-
cal examinations. All study participants received
financial compensation to cover time off from work
and transportation costs. The study was approved by
the Institutional Review Board of Taipei Veterans
General Hospital and was conducted in accordance
with the Declaration of Helsinki (sixth edition). Prior
to entering the study, all participants gave their
informed consents to participate.
2.2 Cortisol measurement
For all participants, the study was performed in the
morning between 8 and 10 am when cortisol levels
are expected to be the highest. Blood samples were
collected at the beginning of the experiment and
were centrifuged immediately after sampling. Blood
serum samples were stored at 70 °C until the time
of analysis. Cortisol assays were performed at Taipei
Veterans General Hospital.
2.3 Psychological and psychophysical
assessments
The overall spontaneous myofascial pain in patients
was assessed by the McGill pain questionnaire
(MPQ). To further evaluate the general physical dys-
function of the shoulder, the Disabilities of the Arm,
Shoulder and Hand (DASH) questionnaire was
administered to all participants. The psychological
state of all the participants was assessed by the pain
catastrophizing scale, the Spielberger state-trait anxi-
ety inventory and the Beck depression inventory. All
questionnaires were filled out immediately before
scanning. Pressure algometry (Baselineâ Dolorime-
ters, NY, USA) was used to assess the subjective pain
thresholds of the upper left and right trapezius mus-
cles prior to scanning. Pressure was applied in four
series (the first was discarded) to each of the MTrPs
in patients and at equivalent sites in healthy con-
trols. Participants were instructed to immediately
indicate when the algometer induced pain. Pressure
algometry was performed by the same person
throughout the experiment.
2.4 Imaging acquisition
This study was part of a protocol in which structural,
functional and spectroscopic magnetic resonance
imaging data was acquired on a 3T MR system (Trio;
SIEMENS Medical Solutions, Erlangen, Germany)
with a 32-channel head coil array. The participants’
heads were placed in the scanner after being
© 2016 European Pain Federation - EFICâ
Brain structural changes in myofascial pain
immobilized with pads. The structural scan consisted
of a high-resolution 3D MPRAGE (Magnetization Pre-
pared Rapid Acquisition Gradient Echo) sequence
with the following parameters: repetition time,
2530 ms; echo time, 3.03 ms; flip angle, 7°; field of
view, 224 9 256 9 192; and voxel size, 1 9 1 9
1 mm.
2.5 Hippocampal volumetric analysis
Hippocampal grey-matter volumes were calculated
using the HV toolbox (Jung Diagnostics GmbH, Ham-
burg, Germany; http://www.jung-diagnostics.de/eng/
tools.php) integrated into the statistical parametrical
mapping toolbox (SPM8; Wellcome Trust Centre for
Neuroimaging, London, http://www.fil.ion.ucl.ac.uk/
spm). In short, the individual anatomical images were
segmented into grey-matter (GM), white-matter
(WM) and cerebrospinal fluid (CSF) images using the
New Segment toolbox in SPM8. The hippocampal GM
volumes were then extracted from the GM image by
summing over all voxel intensities within predefined
masks covering the left and right hippocampi (Suppa
et al., 2015a,b). The total intracranial volume (TIV)
was calculated by summing up all voxel intensities
within GM, WM and CSF images. Since the anatomi-
cal template and tissue probability maps used in the
HV toolbox are generated from an elderly population,
these were replaced by the ICBM152 high-resolution
template and tissue probability maps (http://www.
bic.mni.mcgill.ca/ServicesAtlases/ICBM152NLin2009).
To further investigate volumetric changes of the ante-
rior and posterior portions of the hippocampi, the
masks provided were split into halves along the ante-
rior–posterior direction using MRIcron (http://
www.mccauslandcenter.sc.edu/mricro/mricron/).
2.6 Voxel-based morphometry
A VBM analysis was performed on the segmented
GM images using SPM8. Following the procedures
from Tu et al. (2010), the DARTEL toolbox was first
used to create an optimized GM template in standard
space. Registration parameters and deformation maps
generated during this procedure were applied for
spatial warping of the individual GM images. Spatial
warping was performed with the GM volume pre-
served. This step ensures that GM volumes are com-
pared instead of GM concentrations. Finally, the GM
images were spatially smoothed with a 4-mm3 (full
width at half maximum) Gaussian kernel.
Statistical analysis of the GM images was per-
formed using age and TIV as covariates of no
Eur J Pain  (2016) – 3
Brain structural changes in myofascial pain D. M. Niddam et al.

interest. The rationale for this is to remove the results were considered significant when passing
unwanted contributions of GM decreases with age p < 0.05.
and GM scaling with the size of the brain. Voxels
with a GM concentration <0.2 were excluded from
3. Results
the analysis to avoid possible boundary effects
between tissue types and to preserve homogeneity.
3.1 Demographics and clinical characteristics
The analysis proceeded in three steps. First, regional
between-group GM changes within the hippocampi Demographics and the pain profile of the patients
were investigated by means of a two-sample t-test. are provided in Table 1. The two groups of partici-
Voxels passing a threshold of p < 0.05 corrected for pants were gender matched and did not differ signifi-
family-wise errors within the volume of the bilateral cantly with respect to age. Anxiety, depression or
hippocampi were considered significant. The small pain catastrophizing scores did also not differ
voxel-wise volume correction was obtained by between the two groups and were within the normal
applying a mask covering the bilateral hippocampi range. Basal serum cortisol levels (17 controls and
using the WFU Pickatlas (Maldjian et al., 2003). Sec- 20 patients) and TIV did also not differ significantly
ond, to further investigate possible between-group between the two groups. As evident from the MPQ
differences, an explorative whole-brain analysis was scores, patients were characterized by spontaneous
performed using an uncorrected threshold of pain emanating from the upper trapezius muscles.
p < 0.005 combined with a minimum cluster extent The spontaneous pain had been present for more
of 50 voxels. Finally, to further address stress- and
pain-related changes, regression analyses were per-
Table 1 Demographics of healthy controls and patients (mean  SD).
formed within the two subject groups, separately,
using individual anxiety scores, morning cortisol Controls Patients
(N = 21) (N = 21) p
levels, MPQ scores and the mean PPT of left and
right sides. Clusters overlapping spatially with clus- Age (years) 38.0  7.7 42.1  12.7 0.207a
ters from the between-group analysis were identified Gender (F/M) 16/5 16/5 1.000b
by applying an uncorrected voxel threshold of MPQ total pain na 21.7  13.0 –
score (0–78)
p < 0.005 in conjunction with an inclusive mask
MPQ present na 2.8  1.2 –
consisting of significant GM changes from the
pain index (0–5)
between-group analysis. Only clusters with a mini- Pressure pain
mum overlap of 10 voxels were considered. For each threshold (kg)
of these clusters, the entire spatial extent was then Left 5.81  2.43 3.95  1.66 0.016a
extracted by repeating the analysis without the Right 6.05  2.19 4.25  1.58 0.012a
masking procedure using an uncorrected voxel Pain duration (N)
0.5–2 years na 6 –
threshold of p < 0.005. Coordinates of significant
2–5 years na 6 –
clusters were transformed into Talairach space using
5–10 years na 2 –
the mni2tal transform (http://imaging.mrc-cbu.ca- >10 years na 7 –
m.ac.uk/imaging/MniTalairach), and the correspond- DASH (0–100) 3.6  3.8 24.0  10.8 <0.001a
ing anatomical structures were labelled using the Pain 14.25  6.6 19.1  13.1 0.186a
Talairach Client (http://www.talairach.org/client. catastrophizing
html). (0–52)
BDI (0–63) 5.4  5.5 8.8  7.9 0.115a
STAI trait (20–80) 41.8  10.2 41.5  11.7 0.941a
2.7 Statistics STAI state (20–80) 38.0  7.7 43.2  12.3 0.172a
Between-group differences in age, cortisol levels, Cortisol (lg/dL) 10.9  4.8 11.8  3.9 0.585a
Total intracranial 1353.3  147.6 1287.3  104.0 0.102a
PPT, TIV and scores from the questionnaires were
volume (mL)
assessed by two-tailed two-sample t-tests, and gender
differences were examined with a chi-square test. Score ranges for each questionnaire is given in brackets. F, females;
Hippocampal volumes (whole volume, anterior and M, males; na, not available; N, number of subjects; MPQ, McGill pain
questionnaire; BDI, Beck depression inventory; DASH, disabilities of
posterior portions) were compared between the two
the arm, shoulder and hand questionnaire; STAI, State-trait anxiety
subject groups using analysis of covariance with age inventory.
and TIV as covariates. All of these analyses were per- a
Two-sample t-test.
formed using SPSS 20 (SPSS Inc., USA), and the b
Chi-square test.

4 Eur J Pain  (2016) – © 2016 European Pain Federation - EFICâ

D. M. Niddam et al. Brain structural changes in myofascial pain

than 2 years in the majority of the patients. Pressure
algometry further revealed a significantly lowered
pain threshold in patients for both the left and right
side indicating the presence of muscular hyperalgesia
at the trigger point sites. In addition, patients had
significantly elevated DASH scores compared with
healthy controls, showing that the state of the
patients was associated with some disability. Finally,
DASH scores were positively correlated with the total
MPQ scores (Pearson correlation; r = 0.645,
p = 0.002), but not the mean PPT ratings of left and
right sides (r = 0.072, p = 0.755), suggesting the dis-
ability to be related to the presence of the sponta-
neous pain rather than the hyperalgesia. No
relationship was found between any of the pain
parameters and cortisol levels or anxiety scores.
3.2 Hippocampal volumetry
Mean hippocampal volumes and adjusted volumes
are listed in Supplementary Table 1. A difference
between patients and healthy controls in overall hip-
pocampal GM volume was found neither in the left
nor in the right hemisphere. When the hippocampal
volumes were partitioned into anterior and posterior
portions, a trend towards reduced volume in the left
anterior hippocampus of patients (F = 2.91;
p = 0.096) was found. However, repeating the analy-
sis with age and TIV as covariates resulted in no sig-
nificant difference (F = 0.304; p = 0.585). None of
the other between-group comparisons resulted in
significant findings. To test the effect of spontaneous
and evoked pain on hippocampal GM volumes, cor-
relations with total MPQ scores and mean PPT levels
were examined in the patient group. However, a sig-
nificant correlation was not found between any of
the measures. To further test the effect of stress on
hippocampal GM volumes, correlations with cortisol
levels and anxiety scores were examined. Borderline
significant correlations with serum cortisol were
found in patients for left (r = 0.443, p = 0.050) and
right (r = 0.410, p = 0.073) posterior hippocampi.
However, using second-order partial correlations to
account for age and TIV resulted in no significant
findings for either side (left: r = 0.222, p = 0.376;
right: r = 0.278, p = 0.351). Significant correlation
with serum cortisol was also not found for any of
the hippocampal volume measures in healthy con-
trols as was the case for both subject groups for cor-
relations with anxiety scores.
3.3 Voxel-based morphometry
In accord with our a priori defined hypothesis,
patients with active MTrPs demonstrated a signifi-
cant reduction in GM volume in the left hippocam-
pus, when applying a small volume correction
(Table 2, Fig. 1). Reduced GM volume was also
found in the right hippocampus ([x, y, z] = [32, 8,
20], Zmax = 3.07, extent = 14 voxels) albeit with-
out passing the extent threshold. In the explorative
whole-brain analysis, additional reductions in GM

Table 2 Significant grey-matter volume reductions in patients compared with healthy controls.

Anatomical area Brodmann area Cluster size Pcor Zmax Puncor Talairach coordinates

Hypothesis-driven analysis
L Parahippocampal G Hipp 77 0.022 4.25 <0.001 24 13 11
Explorative whole-brain analysis
R fusiform G 20 419 0.016 5.21 <0.001 47 28 23
R inferior temporal G 20 0.976 3.83 <0.001 53 10 27
R superior frontal G 10 54 0.377 4.41 <0.001 21 65 12
L Parahippocampal G Hipp 102 0.58 4.25 <0.001 24 13 11
L claustrum – 141 0.95 3.9 <0.001 29 13 6
L fusiform G 20 211 0.997 3.7 <0.001 44 27 24
R middle frontal G 8 56 0.998 3.67 <0.001 30 30 41
R middle temporal G 21 146 1 3.53 <0.001 64 39 8
R inferior frontal G 9 51 1 3.41 <0.001 39 10 24
L superior frontal G 10 64 1 3.36 <0.001 31 60 10
L middle temporal G 21 55 1 3.3 <0.001 68 28 13
L middle temporal G 39 59 1 3.24 0.001 55 53 15
R claustrum – 72 1 3.09 0.001 30 11 7

Total intracranial volume and age were used as covariates of no interest. Statistical maps were thresholded at an uncorrected voxel level of
p < 0.005.
Hipp, hippocampus; L, left; R, right; Pcor, voxel-level p-values corrected for family-wise errors; Puncor, uncorrected voxel-level p-values; Zmax, peak
voxel Z-value.

© 2016 European Pain Federation - EFICâ Eur J Pain  (2016) – 5

Brain structural changes in myofascial pain
volumes were found in the bilateral superior frontal
gyri, right ventrolateral and dorsolateral prefrontal
cortex, bilateral claustra extending into the anterior
insula, and bilateral middle and inferior temporal
gyri (Table 2). Increased GM volumes in the patient
group were not found in any regions of the brain.
We further explored the contribution of pain- and
stress-related factors to the observed between-group
GM changes (Table 3). The regression analysis with
total MPQ scores in patients revealed a positive rela-
tionship with GM volume in the left inferior tempo-
ral and fusiform gyri (Fig. 2A) and a negative
relationship with GM volume in the left anterior
insula/claustrum (Fig. 2B). The two clusters in the
left inferior temporal cortex were mainly located
adjacent to the cluster from the between-group com-
parison and only overlapped along the edges. For
the mean PPT, a positive relationship was found
with the right middle frontal gyrus in patients
(Fig. 2C) and with the left inferior temporal gyrus in
healthy controls (Fig. 2E). Further analyses revealed
a positive relationship between serum cortisol levels
and GM volume of patients in the right inferior tem-
poral gyrus (Fig. 2D) and in the left hippocampus
Figure 1 Regional grey-matter volume reductions in chronic myofascial pain patients. Reductions were found in the bilateral inferior temporal gyri
(ITG)/fusiform gyri, the left hippocampus (Hipp), bilateral middle temporal gyri (MTG), bilateral superior frontal gyri (SFG), bilateral anterior insula
(aIns)/claustrum, and the right inferior frontal (IFG) and middle frontal (MFG) gyri. The results are superimposed on the SPM T1 template, and the
colour bar refers to Z-values for grey-matter reductions.
6 Eur J Pain  (2016) –
D. M. Niddam et al.
(Fig. 3). The former was located adjacent to the clus-
ter from the between-group analysis and only over-
lapped along the edges. In healthy controls, serum
cortisol levels exhibited a positive relationship with
left hippocampal GM volume (Fig. 3). Although not
exhibiting a significant between-group difference, it
is noteworthy that a positive relationship with corti-
sol was also observed in the right hippocampus
([x, y, z] = [30 8 20], Zmax = 3.14, extent = 13
voxels) in healthy controls but not in patients
(Fig. 3). Finally, a positive relationship with state
anxiety scores was found for right fusiform/inferior
temporal GM volume in healthy controls (Fig. 2F).
Regression with trait anxiety scores did not yield any
significant findings in any of the subject groups.
4 Discussion
The central mechanisms of chronic pain arising from
MTrPs are incompletely understood. Questions
remain regarding whether structural brain changes
occur and whether such changes are pain or stress
related. We addressed these questions by first exam-
ining changes in GM volumes of the hippocampus, a
© 2016 European Pain Federation - EFICâ
D. M. Niddam et al. Brain structural changes in myofascial pain

Table 3 Correlations of grey-matter volumes in patients and healthy controls. Total intracranial volume and age were used as covariates of no
interest.

Anatomical area Brodmann area Cluster size Zmax Puncor Talairach coordinates

Patients: total MPQ

L inferior temporal G 20 62 3.22 0.001 38 14 34
L fusiform G 20 52 3.02 0.001 45 30 21
L inferior frontal G 47 212 3.67 <0.001 36 32 4
L claustrum – 3.05 0.001 27 25 3
Patients: mean PPT
R middle frontal G 8 14 3.14 0.001 32 30 40
Controls: mean PPT
L inferior temporal G 20 117 3.93 <0.001 42 19 32
Patients: cortisol
R inferior temporal G 20 73 3.87 <0.001 42 10 31
L Parahippocampal G Hipp 10 3.18 0.001 35 16 12
Controls: cortisol
L parahippocampal G Hipp 11 3.12 0.001 26 12 12
Controls: state anxiety
R Fusiform G 20 57 3.82 <0.001 42 11 25

Statistical maps were thresholded at an uncorrected voxel level of p < 0.005. Negative Z-values denote negative correlations. Cluster size refers to
the full extent of the cluster and not to the size of the overlap with clusters from the between-group analysis.

A B C D

E F

Figure 2 Regional grey-matter (GM) volume correlations in patients and in healthy controls. In the patient group, correlations were found for (A
and B) total MPQ scores in the left inferior temporal (ITG) and inferior frontal gyri (IFG), (C) mean PPT levels in the right middle frontal gyrus (MFG)
and (D) cortisol levels in the right ITG. In the control group, correlations were found for (E) mean PPT levels in the left ITG and for (F) state anxiety
scores in the right ITG. Negative and positive X coordinates refer to sagittal slices in the left and right hemispheres, respectively. Positive correla-
tions are marked in red and negative correlations in blue.

key region in shaping the stress response, and then changes and pain- and stress-related parameters. The
by performing an explorative whole-brain search for main findings of the study were as follows: (1)
additional regional changes. We, furthermore, exam- Chronic myofascial pain was associated with subtle
ined the relationship between regions exhibiting GM GM atrophy in the left anterior hippocampus. (2)

© 2016 European Pain Federation - EFICâ Eur J Pain  (2016) – 7

Brain structural changes in myofascial pain
Figure 3 Grey-matter volumes positively correlating with cortisol levels in chronic myofascial pain patients (red) and in healthy controls (blue).
Pink colour denotes spatial overlap between the two subject groups. Slices to the right are located more anterior than slices to the left. For illus-
trative purpose, the maps have been thresholded at an uncorrected voxel threshold of p < 0.05. Note: the cluster in the right hemisphere is con-
sidered nonsignificant because it was not located within a significant cluster from the between-group analysis.
GM volumes of the left hippocampus correlated posi-
tively with serum cortisol levels in both subject
groups. (3) Additional GM atrophy was found in
prefrontal and temporal regions. GM volumes of sev-
eral of these loci correlated with pain-related param-
eters. (4) The patient cohort exhibited normal
physiological and emotional stress levels but lowered
pain thresholds and increased disability scores. It is
noteworthy that the average level of spontaneous
pain in our patient cohort was of a magnitude com-
parable with that reported for other severe pain con-
ditions, such as chronic low back pain and
neuropathic pain (Ruoff et al., 2003; Novak and
Katz, 2010).
Left hippocampal GM atrophy was observed in
patients within a regionally specific portion of the
anterior hippocampus. Evidence suggests the hip-
pocampus to be functionally divided along the sep-
totemporal axis with regulation of anxiety and stress
responses localized to the anterior portion (Banner-
man et al., 2004). This region of the hippocampus is,
furthermore, known to be directly targeted by stress
mediators such as cortisol (Wang et al., 2013).
Nonetheless, the observed atrophy found in the
patient group is most like not driven by stress-related
factors as cortisol levels and anxiety scores were
within normal range and did not differ between
patients and controls. Also, GM volumes exhibited a
positive, and not a negative, correlation with serum
cortisol levels in both subject groups. Thus, our find-
ings should be considered within the context of nor-
mal stress levels at which cortisol exerts a trophic
effect. This is congruent with observations of
enhanced neuronal growth in response to slight ele-
vations of circulating glucocorticoids (Schoenfeld and
Gould, 2013). In this context, it is important to
8 Eur J Pain  (2016) –
D. M. Niddam et al.
mention that the regression models accounted for
the effect of age on GM volume, i.e. decreased vol-
ume with increased age. However, since cortisol
levels are known to increase with age (Sudheimer
et al., 2014), our model may also have accounted for
this effect. Other factors not accounted for may have
contributed to the observed hippocampal atrophy.
For example, proinflammatory cytokines have been
associated with suppressed neurogenesis, increased
levels of apoptosis and reduced hippocampal volume
(Sudheimer et al., 2014; Chesnokova et al., 2016).
Elevated cytokine levels have been observed in the
vicinity of active MTrPs as well as at noninvolved
distant sites (Shah et al., 2008) and may act on the
brain via primary afferent pathways or the circula-
tion (Chesnokova et al., 2016). However, indications
of elevated serum levels have not been found in
patients with chronic pain originating from MTrPs
(King et al., 1997; Deitos et al., 2015). Disease dura-
tion is another possible factor which previously has
been related to GM changes in chronic pain condi-
tions (Tu et al., 2010; Jensen et al., 2013). As pain
history was based on recalled self report in our
study, it was not accurate enough to use in the
regression analysis. It is, thus, not possible to discern
the influence of disease duration on the observed
GM changes in this study.
Grey-matter atrophy was also observed in several
regions known to be involved in top-down pain
modulation, e.g. the ventrolateral and dorsolateral
prefrontal cortices. In particular, the right dorsolat-
eral prefrontal cortex has been implicated in top-
down inhibition of pain and has been shown to be
affected in several pain conditions (Lorenz et al.,
2003; Apkarian et al., 2004; Tu et al., 2010). The
notion of pain disinhibition is supported by our
© 2016 European Pain Federation - EFICâ
D. M. Niddam et al.
finding of a positive correlation with the mean PPT
in patients, i.e. more severe atrophy was associated
with a more severely reduced pain threshold. Such a
response may arise as a consequence of ongoing
nociceptive input and may be considered part of the
maladaptive chronification process. Disinhibition by
the dorsolateral prefrontal cortex may lead to
changes in other key brain circuits involved in pain
processing. In this study, GM atrophy was also
observed bilaterally in regions covering the inferior
frontal cortex, the anterior insula and the claustrum.
These regions have been implicated in pain affect,
including pain anticipation and pain salience (Dow-
nar et al., 2003; Paulus and Stein, 2006; Lin et al.,
2013). All three regions have previously been shown
to be engaged in response to painful stimulation of
active MTrPs (Niddam et al., 2007). The GM volume
of these regions in the left hemisphere was found to
correlate negatively with the total MPQ score, sug-
gesting a compensatory response to the ongoing and
highly salient pain assault.
Additional atrophic GM changes were found in
lateral temporal brain regions not typically associated
with pain processing. Most pronounced were the
reduced GM volumes of the bilateral inferior tempo-
ral cortices. These regions are considered association
areas responsible for the integration of visual stimuli
and are influenced by brain circuits involved in cog-
nitive and affective processing (Greicius et al., 2003;
Uddin et al., 2009; Su et al., 2015). The inferior tem-
poral cortices are, furthermore, considered part of
the default mode network, a network known to be
disengaged by sensory stimuli and attention
demanding tasks (Greicius et al., 2003; Uddin et al.,
2009). Atrophic changes in inferior temporal regions
have not only been observed in other pain condi-
tions (Cauda et al., 2014) but also in clinically
depressed patients (Grieve et al., 2013) and in post-
traumatic stress disorder (Kroes et al., 2011). Inter-
estingly, left inferior temporal GM volumes were
found to increase with increasing pain thresholds,
and right inferior temporal GM volumes were found
to increase with increasing state anxiety scores in
healthy controls but not in myofascial pain patients.
This may suggest the absence of trophic, perhaps
neuroprotective, mechanisms in the patient group
that are usually present within normal ranges of
pain thresholds and anxiety levels. In patients, two
additional clusters in the left inferior temporal cortex
were found to correlate positively with total MPQ
scores, and one cluster in the right inferior temporal
cortex was found to correlate positively with cortisol
levels. These clusters, however, were mainly located
© 2016 European Pain Federation - EFICâ
Brain structural changes in myofascial pain
adjacent to the clusters from the between-group
analysis. It is possible that the absence of the afore-
mentioned trophic mechanisms in one area in the
patient group have been replaced by expansion of
adjacent areas, e.g. in a manner depending on the
level of ongoing pain.
Some limitations of this study need to be
addressed. Although VBM can be used to detect
changes in gross neuroanatomy, the corresponding
microscopic mechanisms are not known. GM atro-
phy can be ascribed not only to neuronal apoptosis
and spine/synapse degeneration but also to changes
in cell size, microglia and interstitial fluid (May
et al., 2007; Schweinhardt et al., 2008; Rodriguez-
Raecke et al., 2009). In the same vein, it is not possi-
ble to discern any functional implications based on
VBM alone. In addition, our cross-sectional study
design does not allow to disentangle whether the
observed atrophy is reactive or causative to pain.
Evidence from longitudinal studies suggests that
structural changes, at least in part, arise as a conse-
quence of pain and/or pain-related factors, as some
changes persist in the absence of pain, e.g. in the
periovulatory phase of primary dysmenorrhoea
patients (Tu et al., 2010) and in postsurgery
osteoarthritis patients experiencing pain relief
(Rodriguez-Raecke et al., 2013). The notion of corti-
cal plasticity-induced pain is mainly supported by
findings in patients with phantom limb pain (Flor
et al., 1995). Despite the limited interpretation of
the present results, it is, nonetheless, possible to con-
clude that objective brain structural changes exist in
patients with chronic myofascial pain. Methodologi-
cally, an additional explorative analysis was per-
formed in order to generate novel hypotheses about
central changes in patients with chronic myofascial
pain. A liberal statistical threshold was applied in this
analysis and loci were reported that did not pass a
correction for multiple comparisons. Thus, despite
applying a minimum cluster extent, false-positive
voxels and clusters may still have been included.
However, it can be reasoned that clusters represent-
ing stochastic noise would be expected not to sys-
tematically increase in size at lowered thresholds.
Indeed, when applying a lowered voxel threshold
(p < 0.01), we found all the listed clusters to grow
substantially, i.e. between 63% and 167%. This sug-
gests that these findings should not be considered as
noise but rather that our study was not sufficiently
powered. Nonetheless, our explorative imaging
results should be considered tentative and would
gain in strength by being reproduced. Finally, we did
not confirm our hypothesis about pain- or stress-
Eur J Pain  (2016) – 9
Brain structural changes in myofascial pain
related hippocampal changes in patients with chronic
pain arising from MTrPs in the upper trapezius mus-
cles. In addition, the patient population in our study
did not exhibit exaggerated physiological or psycho-
logical stress levels. Since our initial hypothesis
mainly was based on findings from myofascial pain of
temporomandibular origin, it is likely that stress fac-
tors have a different role in temporomandibular pain
compared to pain originating from the trapezius mus-
cles. It can be speculated that emotional stress is more
prominent in the former (Lorduy et al., 2013), while
physical stress is more prominent in the latter (Treas-
ter et al., 2006) or that stress is more important in the
formation phase of the MTrPs. Further studies are
needed to address this issue.
Author contributions
DMN and RCC contributed to the study design. SHL, YTS
and RCC contributed to the recruitment and diagnosis of
patients. DMN, SHL and YTS contributed to the execution
of the experiment. DMN contributed to the data analysis.
All authors contributed to the paper writing.
References
Apkarian, A.V., Sosa, Y., Sonty, S., Levy, R.M., Harden, R.N., Parrish,
T.B., Gitelman, D.R. (2004). Chronic back pain is associated with
decreased prefrontal and thalamic gray matter density. J Neurosci 24,
10410–10415.
Bannerman, D.M., Rawlins, J.N., McHugh, S.B., Deacon, R.M., Yee,
B.K., Bast, T., Zhang, W.N., Pothuizen, H.H., Feldon, J. (2004).
Regional dissociations within the hippocampus–memory and anxiety.
Neurosci Biobehav Rev 28, 273–283.
Cauda, F., Palermo, S., Costa, T., Torta, R., Duca, S., Vercelli, U.,
Geminiani, G., Torta, D.M. (2014). Gray matter alterations in chronic
pain: a network-oriented meta-analytic approach. Neuroimage Clin 4,
676–686.
Chesnokova, V., Pechnick, R.N., Wawrowsky, K. (2016). Chronic
peripheral inflammation, hippocampal neurogenesis, and behavior.
Brain Behav Immun pii, S0889–1591(16)30017-4. doi: 10.1016/
j.bbi.2016.01.017.
Deitos, A., Dussan-Sarria, J.A., Souza, A., Medeiros, L., Tarrago Mda, G.,
Sehn, F., Chassot, M., Zanette, S., Schwertner, A., Fregni, F., Torres,
I.L., Caumo, W. (2015). Clinical value of serum neuroplasticity
mediators in identifying the central sensitivity syndrome in patients
with chronic pain with and without structural pathology. Clin J Pain 31,
959–967.
Downar, J., Mikulis, D.J., Davis, K.D. (2003). Neural correlates of the
prolonged salience of painful stimulation. NeuroImage 20, 1540–
1551.
Flor, H., Elbert, T., Knecht, S., Wienbruch, C., Pantev, C., Birbaumer, N.,
Larbig, W., Taub, E. (1995). Phantom-limb pain as a perceptual
correlate of cortical reorganization following arm amputation. Nature
375, 482–484.
Gracely, R.H., Petzke, F., Wolf, J.M., Clauw, D.J. (2002). Functional
magnetic resonance imaging evidence of augmented pain processing
in fibromyalgia. Arthritis Rheum 46, 1333–1343.
Greicius, M.D., Krasnow, B., Reiss, A.L., Menon, V. (2003). Functional
connectivity in the resting brain: a network analysis of the default
mode hypothesis. Proc Natl Acad Sci USA 100, 253–258.
10 Eur J Pain  (2016) –
D. M. Niddam et al.
Grieve, S.M., Korgaonkar, M.S., Koslow, S.H., Gordon, E., Williams,
L.M. (2013). Widespread reductions in gray matter volume in
depression. Neuroimage Clin 3, 332–339.
Jensen, K.B., Srinivasan, P., Spaeth, R., Tan, Y., Kosek, E., Petzke, F.,
Carville, S., Fransson, P., Marcus, H., Williams, S.C., Choy, E., Vitton,
O., Gracely, R., Ingvar, M., Kong, J. (2013). Overlapping structural
and functional brain changes in patients with long-term exposure to
fibromyalgia pain. Arthritis Rheum 65, 3293–3303.
King, S., Fisher, B., Brunham, R., Maclean, I. (1997). Tumor necrosis
factor, interleukin-1b, and tissue oxygen levels in myofascial pain
and fibromyalgia syndromes. J Musculoskelet Pain 5, 53–66.
Kroes, M.C., Rugg, M.D., Whalley, M.G., Brewin, C.R. (2011).
Structural brain abnormalities common to posttraumatic stress
disorder and depression. J Psychiatry Neurosci 36, 256–265.
Latremoliere, A., Woolf, C.J. (2009). Central sensitization: a generator
of pain hypersensitivity by central neural plasticity. J Pain 10, 895–
926.
Lin, C.S., Hsieh, J.C., Yeh, T.C., Lee, S.Y., Niddam, D.M. (2013).
Functional dissociation within insular cortex: the effect of pre-
stimulus anxiety on pain. Brain Res 1493, 40–47.
Lorduy, K.M., Liegey-Dougall, A., Haggard, R., Sanders, C.N., Gatchel,
R.J. (2013). The prevalence of comorbid symptoms of central
sensitization syndrome among three different groups of
temporomandibular disorder patients. Pain Pract 13, 604–613.
Lorenz, J., Minoshima, S., Casey, K.L. (2003). Keeping pain out of
mind: the role of the dorsolateral prefrontal cortex in pain
modulation. Brain 126, 1079–1091.
Maihofner, C., Forster, C., Birklein, F., Neundorfer, B., Handwerker,
H.O. (2005). Brain processing during mechanical hyperalgesia in
complex regional pain syndrome: a functional MRI study. Pain 114,
93–103.
Maldjian, J.A., Laurienti, P.J., Kraft, R.A., Burdette, J.H. (2003). An
automated method for neuroanatomic and cytoarchitectonic atlas-
based interrogation of fMRI data sets. NeuroImage 19, 1233–1239.
May, A., Hajak, G., Ganssbauer, S., Steffens, T., Langguth, B.,
Kleinjung, T., Eichhammer, P. (2007). Structural brain alterations
following 5 days of intervention: dynamic aspects of neuroplasticity.
Cereb Cortex 17, 205–210.
McEwen, B.S. (2007). Physiology and neurobiology of stress and
adaptation: central role of the brain. Physiol Rev 87, 873–904.
Mense, S. (2010). How do muscle lesions such as latent and active
trigger points influence central nociceptive neurons? J. Musculoskelet
Pain 18, 348–353.
Nadendla, L.K., Meduri, V., Paramkusam, G., Pachava, K.R. (2014).
Evaluation of salivary cortisol and anxiety levels in myofascial pain
dysfunction syndrome. Korean J Pain 27, 30–34.
Niddam, D.M. (2009). Brain manifestation and modulation of pain
from myofascial trigger points. Curr Pain Headache Rep 13, 370–375.
Niddam, D.M., Chan, R.C., Lee, S.H., Yeh, T.C., Hsieh, J.C. (2007).
Central modulation of pain evoked from myofascial trigger point. Clin
J Pain 23, 440–448.
Niddam, D.M., Chan, R.C., Lee, S.H., Yeh, T.C., Hsieh, J.C. (2008).
Central representation of hyperalgesia from myofascial trigger point.
NeuroImage 39, 1299–1306.
Novak, C.B., Katz, J. (2010). Neuropathic pain in patients with upper-
extremity nerve injury. Physiother Can 62, 190–201.
Paulus, M.P., Stein, M.B. (2006). An insular view of anxiety. Biol
Psychiatry 60, 383–387.
Robinson, R.C., Garofalo, J.P., Gatchel, R.J. (2006). Decreases in
cortisol variability between treated and untreated jaw pain patients. J
Appl Biobehav Res 11, 179–188.
Rodriguez-Raecke, R., Niemeier, A., Ihle, K., Ruether, W., May, A.
(2009). Brain gray matter decrease in chronic pain is the
consequence and not the cause of pain. J Neurosci 29, 13746–13750.
Rodriguez-Raecke, R., Niemeier, A., Ihle, K., Ruether, W., May, A.
(2013). Structural brain changes in chronic pain reflect probably
neither damage nor atrophy. PLoS ONE 8, e54475.
Ruoff, G.E., Rosenthal, N., Jordan, D., Karim, R., Kamin, M., Protocol,
C.-S.G. (2003). Tramadol/acetaminophen combination tablets for the
treatment of chronic lower back pain: a multicenter, randomized,
© 2016 European Pain Federation - EFICâ
D. M. Niddam et al.
double-blind, placebo-controlled outpatient study. Clin Ther 25, 1123–
1141.
Schoenfeld, T.J., Gould, E. (2013). Differential effects of stress and
glucocorticoids on adult neurogenesis. Curr Top Behav Neurosci 15,
139–164.
Schweinhardt, P., Kuchinad, A., Pukall, C.F., Bushnell, M.C. (2008).
Increased gray matter density in young women with chronic vulvar
pain. Pain 140, 411–419.
Shah, J.P., Danoff, J.V., Desai, M.J., Parikh, S., Nakamura, L.Y.,
Phillips, T.M., Gerber, L.H. (2008). Biochemicals associated with
pain and inflammation are elevated in sites near to and remote
from active myofascial trigger points. Arch Phys Med Rehabil 89, 16–
23.
Shah, J.P., Thaker, N., Heimur, J., Aredo, J.V., Sikdar, S., Gerber, L.
(2015). Myofascial trigger points then and now: a historical and
scientific perspective. PM R 7, 746–761.
Su, Q., Yao, D., Jiang, M., Liu, F., Jiang, J., Xu, C., Dai, Y., Yu, M.,
Long, L., Li, H., Liu, J., Zhang, Z., Zhang, J., Xiao, C., Guo, W.
(2015). Increased functional connectivity strength of right inferior
temporal gyrus in first-episode, drug-naive somatization disorder.
Aust N Z J Psychiatry 49, 74–81.
Sudheimer, K.D., O’Hara, R., Spiegel, D., Powers, B., Kraemer, H.C.,
Neri, E., Weiner, M., Hardan, A., Hallmayer, J., Dhabhar, F.S.
(2014). Cortisol, cytokines, and hippocampal volume interactions in
the elderly. Front Aging Neurosci 6, 153.
Suppa, P., Anker, U., Spies, L., Bopp, I., Ruegger-Frey, B., Klaghofer,
R., Gocke, C., Hampel, H., Beck, S., Buchert, R. (2015a). Fully
automated atlas-based hippocampal volumetry for detection of
Alzheimer’s disease in a memory clinic setting. J Alzheimers Dis 44,
183–193.
Suppa, P., Hampel, H., Spies, L., Fiebach, J.B., Dubois, B. and Buchert,
R. & Alzheimer’s Disease Neuroimaging, I. (2015b) Fully automated
atlas-based hippocampus volumetry for clinical routine: validation in
© 2016 European Pain Federation - EFICâ
Brain structural changes in myofascial pain
subjects with mild cognitive impairment from the ADNI Cohort. J
Alzheimers Dis, 46, 199–209.
Treaster, D., Marras, W.S., Burr, D., Sheedy, J.E., Hart, D. (2006).
Myofascial trigger point development from visual and postural
stressors during computer work. J Electromyogr Kinesiol 16, 115–124.
Tu, C.H., Niddam, D.M., Chao, H.T., Chen, L.F., Chen, Y.S., Wu, Y.T.,
Yeh, T.C., Lirng, J.F., Hsieh, J.C. (2010). Brain morphological
changes associated with cyclic menstrual pain. Pain 150, 462–468.
Uddin, L.Q., Kelly, A.M., Biswal, B.B., Castellanos, F.X., Milham, M.P.
(2009). Functional connectivity of default mode network
components: correlation, anticorrelation, and causality. Hum Brain
Mapp 30, 625–637.
Verne, G.N., Himes, N.C., Robinson, M.E., Gopinath, K.S., Briggs, R.W.,
Crosson, B., Price, D.D. (2003). Central representation of visceral and
cutaneous hypersensitivity in the irritable bowel syndrome. Pain 103,
99–110.
Wang, Q., Van Heerikhuize, J., Aronica, E., Kawata, M., Seress, L.,
Joels, M., Swaab, D.F., Lucassen, P.J. (2013). Glucocorticoid receptor
protein expression in human hippocampus; stability with age.
Neurobiol Aging 34, 1662–1673.
Yoshihara, T., Shigeta, K., Hasegawa, H., Ishitani, N., Masumoto, Y.,
Yamasaki, Y. (2005). Neuroendocrine responses to psychological
stress in patients with myofascial pain. J Orofac Pain 19, 202–208.
Supporting Information
Additional Supporting Information may be found online in
the supporting information tab for this article:
Table S1. Volumetry of the whole hippocampus (HV) and
of the anterior (aHV) and posterior (pHV) portions.
Eur J Pain  (2016) – 11