Download Atypical age-related cortical thinning in episodic

Original Article
Atypical age-related cortical thinning
in episodic migraine
Cephalalgia
0(0) 1–10
! International Headache Society 2014
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0333102414531157
cep.sagepub.com
Catherine D Chong1, David W Dodick1, Bradley L Schlaggar2
and Todd J Schwedt1
Abstract
Background: Prior studies demonstrate reduced cortical thickness and volume in migraineurs. However, the effect of age
on cortical thickness has not been assessed in migraineurs. In this study we investigated whether the process of aging on
cortical thickness affects migraineurs differently compared to age-matched healthy controls, i.e. whether aging exacerbates cortical thinning in migraineurs.
Methods: Cortical thickness was estimated using a general linear model vertex-by-vertex approach for 32 healthy controls (mean age ¼ 35.3 years; SD ¼ 11.6) and 27 episodic migraine patients (mean age ¼ 33.6 years; SD ¼ 12.3). Results
were modeled using a main effect analysis to estimate the effect of age on cortical thickness for each group separately,
and an age-by-group analysis to estimate differences in age-related cortical thinning between migraine patients and
normal controls.
Results: Although migraineurs and normal controls both have expected age-related thinning in many regions along the
cortical mantle, migraineurs have age-related thinning of regions that do not thin in healthy controls, including: bilateral
postcentral, right fusiform, and right temporal pole areas. Cortical thinning of these regions is more prominent with
advancing age.
Conclusion: Results suggest that migraine is associated with atypical cortical aging, suggesting that the migraine disease
process interacts with aging to affect cortical integrity.
Keywords
Neuroimaging, migraine, age-related cortical thinning, pain, cortical thickness, aging
Date received: 31 December 2013; revised: 6 February 2014; 19 February 2014; accepted: 17 March 2014
Introduction
Migraine is a debilitating disorder that consists of
recurrent headache, hypersensitivities to lights, sounds
and odors, nausea and vomiting. Approximately 12%
of people experience migraine each year in North
America, with women being three times more likely to
develop migraine compared to men (1–5). Migraine
usually begins during the late childhood years to early
twenties and typically persists for at least several decades (5). Although the clinical manifestations of
migraine are well described, its pathophysiology
remains poorly understood. However, a growing
number of migraine imaging studies demonstrate atypical structure and function of a complex network of
pain processing areas, commonly termed the ‘‘pain
matrix,’’ including sensory-discriminative, cognitive,
and affective regions of pain processing (6–10).
Anatomical studies using voxel-based morphometry
or vertex-wise analysis methods have reported a
reduction of cortical volume and thickness for migraine
subjects in areas involved in pain processing (8,11,12),
and several studies have hypothesized that recurrent
activation of areas associated with pain would eventually lead to this cortical decline. As such, the argument
has been made that migraine could be a progressive
disorder, i.e. that repeated migraine attacks over an
individual’s lifetime could induce a negative and
compounding effect on gray matter (GM) health
(10). Although several studies have documented a
1
2
Mayo Clinic, Phoenix, AZ, USA
Washington University School of Medicine, St. Louis, MO, USA
Corresponding author:
Todd J Schwedt, Mayo Clinic, 5777 East Mayo Blvd., Phoenix, AZ 85054,
USA.
Email: [email protected]
2
correlation between cortical atrophy and migraine frequency and duration (10,13,14), more recently published studies have found the opposite, i.e. the
absence of an interaction between migraine frequency
and brain structure decline (8,15,16). Another prominent factor that negatively affects cortical integrity is age.
With advancing age, cortical thickness decreases
(17–19). Several studies have assessed the effect of specific disease processes on the aging brain (20–25) and
reported accelerated or atypical age-related decline
(atrophy) in pain disorders (22,23,25), autism (24),
schizophrenia (21), and Alzheimer’s disease (20) relative to healthy, normally aging controls. As such, it is
possible that the aging brain is more vulnerable to disease states than the younger brain (26). However, the
effect of age on regional thickness patterns has not been
directly assessed in migraine.
Here, we examined the effects of aging in migraine
patients and control subjects and explore the age-bygroup interactions using a vertex-wise general linear
model (GLM) to estimate cortical thickness along the
entire cortical mantle. We hypothesized that the
migraine disease process interacts with brain aging,
resulting in abnormal cortical aging patterns of the
adult migraine brain.
Methods
Subjects
This study evaluated cortical thickness in 32 healthy
control subjects (25 females, seven males; mean age
35.3 years; age range 20–60 years; SD 11.6) and 27
episodic migraine subjects (22 females, five males;
mean age 33.6 years; age range 18–64 years; SD 12.3).
Although 32 episodic migraine subjects were originally
evaluated, four subjects had to be excluded because
they were unable to tolerate the magnetic resonance
(MR) scanning session and a single subject had to be
excluded because of a brain abnormality identified on
T1 imaging, thus leaving 27 episodic migraineurs in the
final analysis.
All subjects had given informed written consent
prior to participation in accordance with the institutional review board guidelines of Washington
University Medical Center. All migraine subjects met
International Classification of Headache Disorders II
(ICHD-II) diagnostic criteria for episodic migraine
(27). Migraine subjects were not taking migraine
prophylactic medications and were not overusing
migraine abortive medications according to ICHD-II
criteria. All migraine subjects reported having migraine
symptoms for at least three years. Nine of the 27
migraine patients reported visual auras, while the
rest reported no migraine-related auras. Magnetic
Cephalalgia 0(0)
resonance imaging (MRI) was conducted when
migraine patients were headache free for at least 48
hours. Control subjects were healthy community-dwelling volunteers without history of migraine or headache
disorders other than infrequent tension-type headache
(<3 headache days/month). Twenty-four control subjects reported at least one tension-type headache a
month, and eight subjects denied experiencing tension-type headaches. Control subjects were similar in
age (each migraine subject was within four years of
age relative to a healthy control) with proportionate
male/female ratios. All participants completed the
Beck Depression Inventory (BDI-II) and the state/
trait anxiety inventory (STAI, Form Y-1 and Form
Y-2) to assess symptoms of depression and anxiety
(28,29). Subjects were screened for history of head
trauma, psychiatric disorders, chronic or acute pain disorders (other than migraine for patients) and on occurrence excluded from this study. All subjects had normal
T1 and T2 scans, including the absence of white matter
(WM) T2 hyperintensities.
MRI acquisition parameters
All images were obtained on a Siemens MAGNETOM
Trio 3 T scanner (Erlangen, Germany) with total imaging matrix (TIM) technology using a 12-channel head
matrix coil and Food and Drug Administration (FDA)approved sequences. Structural scans included a highresolution three-dimensional (3D) T1-weighted sagittal
magnetization-prepared rapid gradient echo (MPRAGE) series (echo time (TE) ¼ 3.16 ms, repetition
time (TR) ¼ 2.4 s, inversion time (TI) ¼ 1000 ms,
flip angle ¼ 8 degrees, 176 continuous slices with
1 1 1 mm voxels, 256-mm field of view (FOV),
acquisition matrix 256 256) and T2-weighted images
in an axial plane TE ¼ 88 ms, TR ¼ 6280 ms, flip
angle ¼ 120 degrees, 1 1 4 mm voxels, 256-mm
FOV, 36 slices, acquisition matrix 256 256.
Cortical reconstruction and segmentation
T1 MP-RAGE sequence image processing was performed using the FreeSurfer image analysis suite (version 5.3), which is documented and freely available for
download online (http://surfer.nmr.mgh.harvard.edu/).
To avoid postprocessing irregularities between workstations, all image postprocessing was conducted
using a single Mac workstation running OS X Lion
10.7.5 software. The methodology for this procedure is described in detail in prior papers (30–41). In
short, processing includes skull stripping (41),
automated Talairach transformation, segmentation of
subcortical WM and GM structures (34,38), intensity
normalization (42), tessellation of brain boundaries,
3
Chong et al.
automated topology correction (41), and surface
deformation (30–32).
For statistical interpretation, we used surface-based
group analysis tools within FreeSurfer (QDEC) to
investigate cortical thickness differences between
migraine and control subjects. Cortical thickness was
interpreted as a distance measured from the boundary
of the GM and WM to the boundary of the GM/cerebral spinal fluid at each vertex along the brain surface
(32). Cortical thickness maps were smoothed using a
smoothing kernel of 15 mm full width at half maximum
(FWHM).
FreeSurfer output of cortical thickness was manually
inspected to ensure that GM and WM boundaries were
correctly delineated before subject data were included
for further analysis. This is a time-consuming but
important procedure that requires up to one hour per
subject, but it validates that none of the subjects
included in this study had significant errors in estimating cortical thickness throughout the brain.
Statistical analysis
Demographic group characteristics and estimates for
total intracranial volume (ICV), and total GM and
WM volume were assessed using independent sample t
tests (two tailed). Gender differences were assessed using
a chi-square test. First, for migraine subjects and healthy
controls, regional cortical thickness-to-age interactions
were analyzed within FreeSurfer (QDEC) using a GLM
whole-brain vertex-wise analysis. This was conducted
for each group and each hemisphere separately.
Statistical significance levels were set at p < 0.01, uncorrected. Next, an age-by-group interaction analysis was
conducted in order to estimate whether cortical thinning
was more strongly related to age for one group versus
another. In order to ascertain that depression (BDI-II),
anxiety (STAI), and migraine burden (calculated as
years with migraine headache frequency) were not
contributors to group differences they were entered as
covariates and statistically controlled for within the
GLM model. To control for multiple comparisons, statistical significance levels were cluster-corrected for both
hemispheres using a Monte Carlo simulation (43) of
p < 0.025. Lastly, to visually graph regional differences
in the age-related thinning slopes of migraine subjects
and healthy controls, mean cortical thickness of selected
regions surviving the p < 0.025 cluster correction was
estimated. These regions were exported into SPSS 21.0
(SPSS Inc, Chicago, IL) for further analysis. To test
whether the mean thickness-to-age correlation slopes
between both groups differed significantly from each
other, we used a Web utility (http://quantpsy.org) that
converts the correlation coefficients into a z score using
Fisher’s r-to-z transformation (44) and a formula
described by Cohen and Cohen (45) (formula 2.8.5,
page 54) that estimates significance levels of correlation
coefficients taking into account the sample sizes of both
subject groups.
Results
Subject demographics and migraine characteristics are
shown in Table 1. There were no significant differences
on age (p ¼ 0.59), sex (p ¼ 0.75), depression scores
(p ¼ 0.17) or anxiety scores (STATE: p ¼ .66; TRAIT:
p ¼ 0.50) between groups. There were no significant differences between migraine and control subjects for total
intracranial volume (migraineurs: 1377 cm3; healthy
controls: 1464.2 cm3; p ¼ 0.16), total GM volume
(migraineurs: 600.1 cm3; healthy controls: 616.6 cm3;
p ¼ 0.38), or total WM volume (migraineurs: 428 cm3;
healthy controls: 451.9 cm3; p ¼ 0.11).
Additionally, in order to estimate overall cortical
thickness differences between groups, a GLM analysis
was conducted using age as a covariate of no interest.
Following Monte Carlo cluster correction for both hemispheres (p < 0.025), results indicated no surviving clusters in the left or right hemisphere, thus indicating no
significant overall cortical thickness group differences.
Main effect analysis of age-related thinning
First, we computed the effect of age on cortical thickness for each subject group separately (Figure 1).
Table 1. Subject demographics and clinical characteristics of
migraineurs and healthy controls.
Means (SD)
Age
Gender (f/m)
BDI
State
Trait
Migraine onset
(age in years)
Disease duration (years)
Headache frequency
(days per month)
Healthy
controls
n ¼ 32
Migraineurs
n ¼ 27
p value
35.3 (11.6)
25/7
4.6 (4.9)
25.6 (10.3)
29.2 (9.3)
n/a
33.6 (12.3)
22/5
2.6 (4.8)
25.1 (9.3)
31.2(10.5)
17.6 (8.4)
0.59
0.75
0.17
0.66
0.50
n/a
n/a
n/a
16 (9.2)
6.4(3.0)
n/a
n/a
Migraine onset refers to the age (in years) when migraine symptoms first
started. Disease duration was calculated as the number of years lived
with migraine. Headache frequency is indicated in days per month.
f: female; m: male; BDI: Beck Depression Inventory; SD: standard deviation; n/a: not applicable.
4
Healthy controls and episodic migraineurs had widespread bilateral cortical thinning with advancing age
over frontal and temporal regions. There were no significant areas of age-related thickening in the migraine
group. In the control group, a small cluster of left lateral occipital thickening was found.
Age-by-group interaction
There were significant group differences in the age-bycortical thickness correlations (Figure 2). The episodic
migraine group had progressive age-related cortical
thinning compared to control subjects in a variety of
bilateral regions, although slightly more pronounced
in the right hemisphere. Cluster-corrected regions of
age-related thinning included those within the bilateral
postcentral, right fusiform and right temporal pole.
See Table 2 for a complete list of these cortical
regions. Slopes of cortical thinning according to age
were calculated for mean cortical thickness for selected
regions surviving cluster correction. Regions surviving
cluster correction showed significant differences of cortical thinning in migraine patients and healthy controls. Scatterplots in Figure 3 demonstrate the agegroup interaction of mean cortical thickness (calculated across both hemispheres), right fusiform, left
postcentral, and right temporal pole cortical thickness
Cephalalgia 0(0)
with age. Migraineurs show significant age-related
thinning in the left postcentral (r ¼ 0.624;
p ¼ 0.001),
right
temporal
pole
(r ¼ 0.631;
p < 0.0001), and right fusiform (r ¼ 0.569, p ¼ 0.002)
region. In contrast, healthy control subjects indicate
no significant thinning in the left postcentral area
(r ¼ 0.070; p ¼ 0.702), mild, but nonsignificant thickening in the right temporal pole (r ¼ 0.348; p ¼ 0.051),
and significant thickening by age in the right fusiform
(r ¼ 0.410, p ¼ 0.020). Aging slopes for the left postcentral (p ¼ 0.0037), right temporal pole (p < 0.0001),
and the right fusiform (p < 0.0001) are significantly
different between migraineurs and healthy controls.
Both groups indicate significant age-related mean cortical thinning across both hemispheres (migraineurs:
r ¼ 0.785, p < 0.0001; healthy controls: r ¼ 0.411,
p ¼ 0.019). Aging slopes are significantly different
between groups (p ¼ 0.024), with migraineurs showing
steeper overall cortical thinning compared to healthy
controls. Overall, our results indicate that migraine
patients have steeper slopes of cortical thinning with
advancing age whereas no cortical thinning was
apparent in the healthy control group for left postcentral, right temporal pole, and right fusiform regions.
Interestingly, although during young adulthood cortical thickness was similar between migraine and
healthy control groups, with migraineurs showing
Figure 1. Age-related thinning and thickening patterns shown for healthy controls and migraine subjects for the right (a) and left (b)
hemisphere separately. The color blue indicates age-related thinning and red indicates age-related thickening along the cortical mantle.
In order to visually demonstrate the widespread age-related changes, significance thresholds were set at p < 0.01, uncorrected.
Chong et al.
perhaps slightly increased cortical thickness, this pattern reverses past the third decade of life, when
migraineurs demonstrate accelerated cortical thinning
relative to healthy controls.
Discussion
The main finding of this study is that migraine interacts
with the aging process, resulting in localized regions of
cortical thinning in temporal (right fusiform and right
temporal pole) and parietal (bilateral postcentral)
regions.
Continuous decline of cortical thickness with advancing age is a well-established process (18,19,46).
Although our results of predominant frontal thinning
in healthy controls and migraineurs are in agreement
with several normal aging studies (17–19,47), those studies have also found widespread areas of cortical thinning throughout the cortex. Unlike our study, which
included age ranges 20 through 60, those studies had
larger sample sizes including age ranges of subjects
from late childhood through the eighth decade of life,
and were henceforth able to show more pronounced
5
cortical thinning patterns than demonstrated by our
study.
Prior studies have investigated age-related cortical
thinning and volume loss in a variety of disease states
and hypothesized that disease-specific factors may
modulate healthy aging patterns. For example, compared to healthy controls, steeper slopes of cortical
thinning have been found in healthy carriers of the
apolipoprotein E gene (ApoE) (20), in granulin
(GRN) gene carriers (48), and in patients with temporomandibular joint disorder (22). Similarly, accelerated volume decline with advancing age was noted in
patients with schizophrenia (49).
Since migraine typically begins in late childhood to
early adulthood and lasts for at least many decades, it is
hypothesized that migraine-specific clinical aspects such
as disease duration and migraine frequency may have a
cumulative or progressive effect on cortical decline (9).
Several studies to date, using voxel-based morphometry, investigated the effect of disease duration on cortical integrity in migraine subjects. Schmitz et al.
showed a positive correlation between disease duration
and WM density in the cerebellum as well as decreased
Figure 2. Age-by-group interaction in migraine subjects and healthy controls for the right and left hemisphere separately. For
illustration, significance thresholds were set at p < 0.01, uncorrected. Light blue areas encircled in red survive cluster correction
(p < 0.025). The color blue indicates regions where migraine subjects show more age-related cortical decline compared to healthy
controls. The color red indicates the opposite. One (a) and (b) show the inflated maps of the lateral brain surface, 2) dorsal view, and
3) ventral view of the left and right hemisphere. For healthy controls and migraine patients, mean cortical thickness was calculated
for regions surviving cluster correction (p < 0.025).
rPC: right postcentral; lPC: left postcentral; rFS: right fusiform; rTP: right temporal pole.
6
Cephalalgia 0(0)
Table 2. Right and left hemisphere clusters are reported separately.
Clusters p < 0.001
Right hemisphere
1
2
3
4
5
6
7
8
Left hemisphere
1
2
3
4
5
6
Cortical region
Region size
(mm2)
Postcentrala (p < 0.0001)
Supramarginal
Temporal polea(p < 0.0001)
Middle temporal
Paracentral
Superior temporal
Inferior temporal
Fusiforma (p < 0.0004)
191.7
93.2
72.4
72.9
65.9
22.5
18.47
18.5
Postcentrala (p < 0.0001)
Superior parietal
Fusiform
Postcentral
Superior temporal
Fusiform
118.2
79.0
28.2
18.4
10.7
3.7
x
y
z
29.5
45.8
38.1
49.7
6.2
45.7
50.5
32.8
34.1
38.7
6.2
6.2
26.8
32.7
9.8
53.9
65.4
39.8
29.2
35
51.9
10.1
38.0
15.8
9.8
34.7
36.5
27.3
50.2
30.1
37.9
47.0
60.9
34.2
2.1
42.8
74.9
57.4
15.9
62.6
12.4
20.8
Cortical regions, size of the surface area (mm2), and Tailarach (x, y, z) coordinates that show significantly more agerelated thinning in migraine subjects compared to healthy controls.
a
Regions that survived Monte Carlo cluster correction for both hemispheres (p < 0.025). Cluster corrected p values are
indicated in parenthesis ().
GM density in the brain stem and lentiform nucleus
while controlling for age (10). Similarly, results by
Rocca et al. demonstrated a negative correlation
between putamen volume and disease duration in pediatric migraine, but an absence of a relationship between
fusiform atrophy and disease duration (15). The study
reported herein is novel since it examined the effects of
age on cortical thickness in episodic migraine subjects.
Although cortical thickness and volume both provide
measurements of changes in GM integrity, these measures do not necessarily track each other (50). It is suggested that cortical thickness might provide a more
reliable measure of cortical integrity (51) and better
detection of slight GM reductions.
Our results indicate that migraine patients, but not
healthy control subjects, show age-associated thinning
of temporal (right fusiform and right temporal pole)
and parietal (bilateral postcentral) cortex. We conclude
that these regions may undergo age-related modification in response to the migraine disease process. Our
findings are in agreement with several other studies that
found abnormal GM aging in patients with chronic
pain due to temporomandibular disorder (TMD) (22)
and fibromyalgia (23,25), albeit age-related atrophy
patterns were demonstrated for different regions (thalamus, cingulate, insular cortex).
Our study identified advanced age-related cortical
thinning of the postcentral region in migraineurs,
which is part of the primary somatosensory cortex,
predominantly participating in sensory-discriminative
pain processing. Functional magnetic resonance imaging (fMRI) studies demonstrate activation of postcentral regions in response to painful stimulation (52,53).
Our finding of postcentral thinning with advancing age
in migraineurs is in accord with other studies showing
cortical volume loss and cortical thinning in patients
with other types of chronic pain (54,55). However, it is
noteworthy that the location of bilateral postcentral
cortex thinning demonstrated in this study does not
seem to be confined to areas subserving processing of
head pain.
Our results also demonstrate age-related cortical
thinning in temporal regions in episodic migraineurs.
The fusiform is a region important for assigning emotional valance (56) and imagery (57). Increased activation in the fusiform gyrus has been reported in chronic
pain subjects during mental visualization of pain (57),
and fusiform gyrus GM volume reductions have been
identified in cluster headache patients who were between
bouts (58). Fusiform atrophy was also identified in children with migraine, the extent of atrophy not correlating
with migraine duration (15). The presence of fusiform
atrophy in very young migraineurs combined with the
absence of a correlation to migraine duration lead the
authors to hypothesize that the fusiform abnormalities
might serve as a migraine disease biomarker.
Structural and functional temporal pole abnormalities are shown in several migraine studies to date.
7
Chong et al.
3.6
3.1
3.4
rh temporal pole (mm)
Ih postcentral (mm)
2.9
2.7
2.5
2.3
*
2.1
1.9
1.7
3.2
3
*
2.8
2.6
2.4
2.2
1.5
2
15
25
35
45
55
65
15
25
35
Age
Controls
2.7
*
2.5
2.3
2.1
1.9
1.7
1.5
45
35
55
65
Age
Mean cortical thickness (mm)
rh fusiform (mm)
2.9
25
55
65
Migraineurs
3.1
15
45
Age
2.75
2.7
2.65
2.6
2.55
2.5
2.45
2.4
2.35
2.3
2.25
2.2
*
15
25
45
35
55
65
Age
Figure 3. Scatterplots indicating differences in age-related slopes for migraine subjects (red) and healthy controls (blue). *Indicates
significant differences in the correlation slopes between healthy controls and migraine patients.
Using positron emission tomography (PET) imaging,
increased activation in the temporal pole was first
demonstrated for migraineurs during the acute migraine
phase relative to migraineurs in the interictal state (59).
These data were corroborated in a more recent fMRI
study that demonstrated temporal pole hyperexcitability
as well as increased functional connectivity to other
brain regions for migraineurs in response to painful
heat stimulation (60). Temporal pole activation
increased in migraineurs during the ictal state, leading
the authors to suggest that the temporal pole might
become sensitized by repeated migraine attacks.
Similarly, increased regional homogeneity in the temporal pole was found using resting-state fMRI for
migraineurs with longer disease durations relative to
migraineurs with shorter disease durations (61). It is of
note that migraine subjects with longer disease durations
were significantly older than those with shorter disease
durations, which complicates the interpretation of the
findings. Lastly, in a study investigating the interregional
correlations of cortical networks, results indicated altered
temporal pole structural connectivity in female migraine
patients relative to healthy control subjects (62).
In summary, this study demonstrates that migraine
patients show structural alterations of temporal and
parietal regions and that cortical decline of these
regions becomes more pronounced with advancing
age. Although the physiological mechanisms underlying cortical thinning are not completely understood,
thinning may reflect cytoarchitectural changes of neuronal density or synaptic pruning, possibilities that need
to be addressed in future studies. It is noteworthy that
migraine prevalence starts declining in the fifth decade
(5) and that migraine symptoms often become ‘‘milder’’
with decreasing frequency during this stage of life (63).
Future longitudinal studies are needed to evaluate
whether age-related cortical thinning contributes to
the decline and/or remission of migraine symptoms in
older patients.
There are several limitations in this study. First,
although the results of this study are intriguing, they
should be interpreted with caution because of the relatively small sample size and cross-sectional design.
Inferences toward a larger population base can only
be warranted using a longitudinal design with a larger
study population.
8
Cephalalgia 0(0)
Migraine is a complex disorder that usually has an
onset in early adulthood and continues for many years
thereafter. Many studies have attempted to correctly
assess the ‘‘migraine burden,’’ which we calculated as
years with migraine headache frequency. It should be
noted that this is a rough estimate and potentially
flawed measure of lifetime headache burden. This measure assumes that the headache frequency reported by
the patient remains the same (or is unaltered) over
many years. In order to have a more precise understanding of the headache pattern over a patient’s lifetime, a longitudinal assessment would facilitate the
interpretation of how declining, progressing, or stable
patterns of attack frequency over someone’s lifetime are
related to structural decline.
Possibly confounding variables that we could not
assess in our analysis are the influence of medication
on cortical thickness. Although we enrolled only
subjects who did not use preventive migraine medication, and were headache free at the time of the scan
(and thus not in need of taking pain-related medication), we cannot account for the effect of medication
that subjects used on occasion to treat migraine attacks.
Since opioid dependency is indicated to decrease regional brain volume (64), we included only subjects who
were not taking scheduled opioid medication. Only one
migraine participant included in this study treated
migraines with opiates on occasion.
Conclusion
Our results suggest that migraine interacts with age to
induce microstructural brain changes. Migraine
patients show age-related cortical thinning patterns
regardless of disease duration and frequency.
Clinical implications
. Migraine interacts with age to induce microstructural brain changes.
. Migraineurs show age-related thinning in regions that do not thin in healthy controls.
. Localized regions of cortical thinning in migraineurs include temporal (right fusiform and right temporal
pole) and parietal (bilateral postcentral) regions.
Funding
This work was supported by a grant from the National
Institutes of Health (K23NS070891) to TJS.
Conflict of interest
None declared.
Authorship statement
We declare that all authors made significant contributions
toward the study design, acquisition, and interpretation of
the data. All authors were involved in drafting and revising
the manuscript, and approved the final version to be published.
Acknowledgments
We would like to thank the study participants and imaging
technicians for their time commitment and dedication to this
study.
References
1. Bigal ME, Rapoport AM, Lipton RB, et al. Assessment of
migraine disability using the migraine disability assessment
(MIDAS) questionnaire: A comparison of chronic
migraine with episodic migraine. Headache 2003; 43:
336–342.
2. Lipton RB, Bigal ME, Diamond M, et al. Migraine prevalence, disease burden, and the need for preventive therapy.
Neurology 2007; 68: 343–349.
3. Lipton RB, Hamelsky SW, Kolodner KB, et al.
Migraine, quality of life, and depression: A populationbased case-control study. Neurology 2000; 55: 629–635.
4. Stewart WF, Linet MS, Celentano DD, et al. Age- and
sex-specific incidence rates of migraine with and without
visual aura. Am J Epidemiol 1991; 134: 1111–1120.
5. Stewart WF, Lipton RB, Celentano DD, et al. Prevalence
of migraine headache in the United States. Relation to
age, income, race, and other sociodemographic factors.
JAMA 1992; 267: 64–69.
6. Maleki N, Becerra L, Brawn J, et al. Concurrent functional and structural cortical alterations in migraine.
Cephalalgia 2012; 32: 607–620.
7. Maleki N, Linnman C, Brawn J, et al. Her versus his
migraine: Multiple sex differences in brain function and
structure. Brain 2012; 135: 2546–2559.
8. Messina R, Rocca MA, Colombo B, et al. Cortical
abnormalities in patients with migraine: A surface-based
analysis. Radiology 2013; 268: 170–180.
9. Rocca MA, Ceccarelli A, Falini A, et al. Brain gray
matter changes in migraine patients with T2visible lesions: A 3-T MRI study. Stroke 2006; 37:
1765–1770.
10. Schmitz N, Admiraal-Behloul F, Arkink EB, et al. Attack
frequency and disease duration as indicators for brain
damage in migraine. Headache 2008; 48: 1044–1055.
11. Schmidt-Wilcke T, Ganssbauer S, Neuner T, et al. Subtle
grey matter changes between migraine patients and
healthy controls. Cephalalgia 2008; 28: 1–4.
9
Chong et al.
12. Schmitz N, Arkink EB, Mulder M, et al. Frontal lobe
structure and executive function in migraine patients.
Neurosci Lett 2008; 440: 92–96.
13. Kim JH, Suh SI, Seol HY, et al. Regional grey matter
changes in patients with migraine: A voxel-based morphometry study. Cephalalgia 2008; 28: 598–604.
14. Maleki N, Becerra L, Nutile L, et al. Migraine attacks the
basal ganglia. Mol Pain 2011; 7: 71.
15. Rocca MA, Messina R, Colombo B, et al. Structural
brain MRI abnormalities in pediatric patients with
migraine. J Neurol 2014; 261: 350–357.
16. DaSilva AF, Granziera C, Snyder J, et al. Thickening in
the somatosensory cortex of patients with migraine.
Neurology 2007; 69: 1990–1995.
17. Hogstrom LJ, Westlye LT, Walhovd KB, et al. The structure of the cerebral cortex across adult life: Age-related
patterns of surface area, thickness, and gyrification.
Cereb Cortex 2013; 23: 2521–2530.
18. Salat DH, Buckner RL, Snyder AZ, et al. Thinning of
the cerebral cortex in aging. Cereb Cortex 2004; 14:
721–730.
19. Sowell ER, Peterson BS, Thompson PM, et al. Mapping
cortical change across the human life span. Nat Neurosci
2003; 6: 309–315.
20. Espeseth T, Westlye LT, Fjell AM, et al. Accelerated
age-related cortical thinning in healthy carriers of
apolipoprotein E epsilon 4. Neurobiol Aging 2008; 29:
329–340.
21. Kubota M, Miyata J, Yoshida H, et al. Age-related cortical thinning in schizophrenia. Schizophr Res 2011; 125:
21–29.
22. Moayedi M, Weissman-Fogel I, Salomons TV, et al.
Abnormal gray matter aging in chronic pain patients.
Brain Res 2012; 1456: 82–93.
23. Kuchinad A, Schweinhardt P, Seminowicz DA, et al.
Accelerated brain gray matter loss in fibromyalgia
patients: Premature aging of the brain? J Neurosci 2007;
27: 4004–4007.
24. Wallace GL, Dankner N, Kenworthy L, et al. Agerelated temporal and parietal cortical thinning in autism
spectrum disorders. Brain 2010; 133: 3745–3754.
25. Ceko M, Bushnell MC, Fitzcharles MA, et al. Fibromyalgia interacts with age to change the brain. Neuroimage
Clin 2013; 3: 249–260.
26. Lado FA, Sankar R, Lowenstein D, et al. Age-dependent
consequences of seizures: Relationship to seizure
frequency, brain damage, and circuitry reorganization.
Ment Retard Dev Disabil Res Rev 2000; 6: 242–252.
27. Headache
Classification
Subcommittee
of
the
International Headache Society. The International
Classification of Headache Disorders, 2nd edition.
Cephalalgia 2004; 24(Suppl 1): 9–160.
28. Beck AT, Steer RA and Brown G. Manual for the Beck
Depression Inventory-II. San Antonio, TX: Psychological
Corporation, 1996.
29. Spielberger C and Gorsuch R. State trait anxiety inventory for adults: Sampler set, manual, test, scoring key.
Redwood City, CA: Mind Garden, 1983.
30. Dale AM and Sereno MI. Improved localization of cortical activity by combining EEG and MEG with MRI
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
cortical surface reconstruction: A linear approach.
J Cogn Neurosci 1993; 5: 162–176.
Dale AM, Fischl B and Sereno MI. Cortical surfacebased analysis. I. Segmentation and surface reconstruction. Neuroimage 1999; 9: 179–194.
Fischl B and Dale AM. Measuring the thickness of the
human cerebral cortex from magnetic resonance images.
Proc Natl Acad Sci U S A 2000; 97: 11050–11055.
Fischl B, Liu A and Dale AM. Automated manifold surgery: Constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE
Trans Med Imaging 2001; 20: 70–80.
Fischl B, Salat DH, Busa E, et al. Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron 2002; 33: 341–355.
Fischl B, Salat DH, van der Kouwe AJ, et al. Sequenceindependent segmentation of magnetic resonance images.
Neuroimage 2004; 23 (Suppl 1): S69–S84.
Fischl B, Sereno MI and Dale AM. Cortical surfacebased analysis. II: Inflation, flattening, and a surfacebased coordinate system. Neuroimage 1999; 9: 195–207.
Fischl B, Sereno MI, Tootell RB, et al. High-resolution
intersubject averaging and a coordinate system for the
cortical surface. Hum Brain Mapp 1999; 8: 272–284.
Fischl B, van der Kouwe A, Destrieux C, et al. Automatically parcellating the human cerebral cortex. Cereb
Cortex 2004; 14: 11–22.
Han X, Jovicich J, Salat D, et al. Reliability of MRIderived measurements of human cerebral cortical thickness: The effects of field strength, scanner upgrade and
manufacturer. Neuroimage 2006; 32: 180–194.
Jovicich J, Czanner S, Greve D, et al. Reliability in multisite structural MRI studies: Effects of gradient nonlinearity correction on phantom and human data.
Neuroimage 2006; 30: 436–443.
Segonne F, Dale AM, Busa E, et al. A hybrid approach
to the skull stripping problem in MRI. Neuroimage 2004;
22: 1060–1075.
Sled JG, Zijdenbos AP and Evans AC. A nonparametric
method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging 1998; 17:
87–97.
Hagler DJ Jr, Saygin AP and Sereno MI. Smoothing and
cluster thresholding for cortical surface-based group analysis of fMRI data. Neuroimage 2006; 33: 1093–1103.
Fisher RA. On the ‘probable error’ of a coefficient of
correlation deduced from a small sample. Metron 1921;
1: 3–32.
Cohen JC and Cohen P. Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ:
Erlbaum, 1983.
Westlye LT, Walhovd KB, Dale AM, et al. Differentiating maturational and aging-related changes of the cerebral cortex by use of thickness and signal intensity.
Neuroimage 2010; 52: 172–185.
Fjell AM, Westlye LT, Grydeland H, et al. Accelerating
cortical thinning: Unique to dementia or universal in
aging? Cereb Cortex 2014; 24: 919–934.
Moreno F, Sala-Llonch R, Barandiaran M, et al.
Distinctive age-related temporal cortical thinning in
10
49.
50.
51.
52.
53.
54.
55.
56.
Cephalalgia 0(0)
asymptomatic granulin gene mutation carriers. Neurobiol
Aging 2013; 34: 1462–1468.
Whitford TJ, Grieve SM, Farrow TF, et al. Progressive
grey matter atrophy over the first 2–3 years of illness in
first-episode schizophrenia: A tensor-based morphometry
study. Neuroimage 2006; 32: 511–519.
Hutton C, Draganski B, Ashburner J, et al. A comparison between voxel-based cortical thickness and voxelbased morphometry in normal aging. Neuroimage 2009;
48: 371–380.
Winkler AM, Kochunov P, Blangero J, et al. Cortical
thickness or grey matter volume? The importance of
selecting the phenotype for imaging genetics studies.
Neuroimage 2010; 53: 1135–1146.
Carlsson K, Andersson J, Petrovic P, et al.
Predictability modulates the affective and sensorydiscriminative neural processing of pain. Neuroimage
2006; 32: 1804–1814.
Wager TD, Atlas LY, Lindquist MA, et al. An fMRIbased neurologic signature of physical pain. N Engl J
Med 2013; 368: 1388–1397.
Baliki MN, Schnitzer TJ, Bauer WR, et al. Brain morphological signatures for chronic pain. PLoS One 2011; 6:
e26010.
Frøkjær JB, Bouwense SA, Olesen SS, et al. Reduced
cortical thickness of brain areas involved in pain
processing in patients with chronic pancreatitis. Clin
Gastroenterol Hepatol 2012; 10: 434–438 e1.
Godinho F, Magnin M, Frot M, et al. Emotional modulation of pain: Is it the sensation or what we recall?
J Neurosci 2006; 26: 11454–11461.
57. Shimo K, Ueno T, Younger J, et al. Visualization of
painful experiences believed to trigger the activation of
affective and emotional brain regions in subjects with low
back pain. PLoS One 2011; 6: e26681.
58. Yang FC, Chou KH, Fuh JL, et al. Altered gray matter volume in the frontal pain modulation network
in patients with cluster headache. Pain 2013; 154:
801–807.
59. Afridi SK, Giffin NJ, Kaube H, et al. A positron emission tomographic study in spontaneous migraine. Arch
Neurol 2005; 62: 1270–1275.
60. Moulton EA, Becerra L, Maleki N, et al. Painful heat
reveals hyperexcitability of the temporal pole in
interictal and ictal migraine states. Cereb Cortex 2011;
21: 435–448.
61. Zhao L, Liu J, Dong X, et al. Alterations in regional
homogeneity assessed by fMRI in patients with migraine
without aura stratified by disease duration. J Headache
Pain 2013; 14: 85.
62. Liu J, Zhao L, Li G, et al. Hierarchical alteration of brain
structural and functional networks in female migraine
sufferers. PLoS One 2012; 7: e51250.
63. Kelman L. Migraine changes with age: IMPACT on
migraine classification. Headache 2006; 46: 1161–1171.
64. Upadhyay J, Maleki N, Potter J, et al. Alterations in
brain structure and functional connectivity in prescription opioid-dependent patients. Brain 2010; 133:
2098–2114.