Neurological Sciences

https://doi.org/10.1007/s10072-019-03851-1

REVIEW ARTICLE

Imaging the migrainous brain: the present and the future

Bruno Colombo 1 & Roberta Messina 1 & Maria Assunta Rocca 1 & Massimo Filippi 1

# Fondazione Società Italiana di Neurologia 2019

Abstract
In the last 20 years, we observed significant improvements in the use of magnetic resonance imaging (MRI) for the evaluation of
patients affected by migraine. Before these technological advances, knowledge of the pathogenesis of migraine was particularly
based on clinical assessment. Complementary to clinical evaluation, conventional MRI provides both specific information for
differential diagnosis (particularly if cortical or subcortical lesions are detected in the migrainous brain) and unsurpassable
opportunities in migraine research. However, the correlations between brain structural and functional alterations and both the
clinical manifestation of the disease and the individual history of the patient remains uncertain. Both quantitative and functional
MR-based techniques have a great potential to better provide insights into human brain structures and possible links between
brain areas and complex brain networks that could be involved in the pathophysiology of migraine. Morphometric and functional
MRI approaches are contributing to better elucidate the mechanisms that underlie the pain mechanisms and functional adaptation
in migraine patients. All these information support the view of migraine as a complex brain disorder involving different cortical
and subcortical areas.

Keywords Migraine . Diagnosis . Structural MRI . Functional MRI

Introduction and historical background
Forty-five years ago, Sir Peter Mansfield (along with postdoc
fellow P.K. Grannell in Nottingham) and Paul C. Lauterbur
(Professor of Chemistry in New York) independently studied
the technique that later became known as magnetic resonance
imaging (MRI) [1, 2]. They devised a method to encode the
nuclear magnetic resonance relaxation information in an ob-
ject spatially and to reproduce it as an image, defining a third
dimension directly. The immediate practical application in-
volves imaging the distribution of hydrogen nuclei (protons).
The image brightness (or intensity) in a given region is usually
dependent (weighted) jointly on the spin density and the re-
laxation times, with their relative importance determined by
the particular technique employed. Neither of these papers
received a great deal of attention at that time. Subsequent
works focused on technological improvements, such as the
experiment to enable imaging of live animals. The first image
* Bruno Colombo
colombo.bruno@hsr.it
1 in patients with atypical headache patterns, a history of sei-
Department of Neurology, IRCCS Ospedale San Raffaele,
Vita-salute University, Via Olgettina 48, 20100 Milan, Italy
of human anatomy was a human finger published in 1975 [3].
MRI gave, in fact, the opportunity to create images by use of
the nuclear magnetic resonance phenomenon.
The 2003 Nobel Prize for Physiology or Medicine was
awarded to Lauterbur and Mansfield for their studies which
led to the development of MRI, one of the greatest innovations
in diagnostic medicine. This technology has become an in-
valuable research and clinical tool, particularly because of its
non-invasive capability. MRI plays a pivotal role in the diag-
nosis and prognosis of many neurological diseases. The intro-
duction of this technique has completely revolutionized the
diagnostic process of inflammatory, degenerative, and cere-
brovascular diseases. However, the proliferation of MRI scan-
ners in the more wealthy countries has sometimes led to their
overuse and to misinterpretation of the images.
The use of MRI to study patients with migraine varies
widely in the clinical settings. MRI examinations are often
obtained in migraine patients because of fear of missing seri-
ous underlying diseases (e.g., stroke, tumors, infections), es-
pecially in patients with migraine with aura. Its use is manda-
tory in the differential diagnosis of acute headache versus
symptomatic headache. However, in patients suffering of
non-acute headache, the use of MRI should be indicated only
zures, and/or focal neurological symptoms or signs [4].

MRI techniques can provide insights into human brain
structures and possible links between brain anatomy and brain
network activities that could be involved in pathophysiology
of migraine. The extensive application of conventional and
advanced MRI techniques to the study of patients affected
by migraine, both in the course of an acute attack and during
the interictal phase, has contributed to improve the under-
standing of the pathophysiology of this condition and to elu-
cidate novel mechanisms which might become a target of
future therapeutic interventions. Neuroimaging techniques fall
broadly into two categories examining either brain anatomy
(structure) or function. In particular, modern morphometric
techniques such as voxel-based, surface-based morphometry,
and diffusion tensor imaging revealed widespread gray matter
(GM) and white matter (WM) structural abnormalities in cor-
tical and subcortical areas and WM tracts involved in multi-
sensory processing, including pain [5, 6].
Functional MRI (fMRI) studies have largely provided new
insights into functional organization of different pain process-
ing networks in migraine patients. Blood oxygenation level–
dependent (BOLD) fMRI is entirely non-invasive, no radio-
active tracers are needed and a good spatial and temporal
resolution is obtained. Functional connectivity (FC) fMRI da-
ta provide information about the interplay between different
brain areas. Their application in studying migraine patients
has shed light on the mechanisms responsible for initiation
and propagation of migraine attacks and has disclosed the
activity of cortical and subcortical regions during the different
phases of headache [7]. In recent years, various alterations
regarding resting-state (RS) FC were discovered, especially
in pain-related regions and networks showing altered func-
tional connectivity in migraine [8]. Whether such alterations
represent a predisposing trait or the consequence of the recur-
rence of headache attacks is still a matter of debate.
Admittedly, these findings might represent a balance between
an intrinsic predisposition and disease-related processes. This
review will discuss the potential of the use of MRI techniques
in the context of the current knowledge of migraine.
Differential diagnosis based on MRI
Seminal MRI studies on migraine patients reported an increased
risk to present diffuse signal abnormalities in brain WM, espe-
cially in subjects affected by migraine with aura [9]. In partic-
ular, different MRI structural alterations are described such as
infra-tentorial T2-hyperintense lesions, silent posterior circula-
tion territory infarcts, and supra-tentorial deep WM lesions
[10]. A common source of difficulty in differential diagnosis
is the presence of small areas of high signal in the cerebral WM
of subjects affected by many other neurological disturbances.
Admittedly, many diseases of the CNS that result in WM le-
sions seen by MRI are often erroneously diagnosed:
Neurol Sci
inflammatory diseases (LES, Sjogren’s disease, Behcet’s dis-
ease, polyarthritis nodosa), granulomatous diseases
(neurosarcoidosis, Wegener granulomatosis), vascular diseases
(prothrombotic states, CADASIL, mitochondrial encephalopa-
thies), infectious diseases (neuroborreliosis, Whipple’s disease,
PML), deficiency diseases (B12 deficiency), hereditary dis-
eases (leukodystrophies). The differential diagnosis can be a
real challenge, considering that in some patients affected by
migraine, the distribution of lesions may even mimic some or
all neuroradiological features of multiple sclerosis (MS) [11]. In
fact, MRI features suggestive of MS are described as scattered
WM lesions, hyperintense on T2, and proton-density scans. In
this specific differential diagnosis, some important features are
to be considered [12]. In MS, lesions are large (more than 3 mm
in size), multiple, and ovoid shaped in specific locations such as
periventricular (98% of patients), infratentorial, brainstem,
juxtacortical, and corpus callosum (best seen on sagittal scan,
lesions pointing away, and moth-eaten appearance). In MS,
lesions tend to involve the deep rather than the more peripheral
WM, although focal lesions are common enough in the periph-
ery. Moreover, it is unusual to see basal ganglia (25%) or inter-
nal capsule lesions (10%). Spinal cord is frequently involved,
particularly in cervical and thoracic regions. Spinal cord lesions
usually involve less than 50% of cross-sectional diameter, le-
sions are less than two segments in length and are located in
lateral, posterior, and anterior columns. Unlike the brain,
asymptomatic spinal cord areas of high signal are very uncom-
mon as an incidental finding with increasing age. A previous
MRI study of tissue damage in the cervical cord of patients with
migraine showed that migraineurs do not have macro and mi-
croscopic tissue abnormalities in the cervical cord if compared
with MS patients [13]. MRI scanning of the cervical cord in
patients with migraine and hyperintense WM lesions of un-
known etiology may be a useful investigation to facilitate the
diagnostic workout. Another aspect investigated in migraine
patients is whether the presence of focal lesions in the cerebral
cortex might have a role in the diagnostic process, particularly
in the differential diagnosis with MS. A pivotal study showed
that, if compared with MS patients, no cortical lesions were
identified in migraine patients using double inversion recovery
(DIR) imaging [14]. By suppressing the signal from both WM
and cerebrospinal fluid, DIR sequences allow a better in vivo
detection of cortical lesions. The results of this study support
the notion that DIR sequences should be used in the diagnostic
workup of patients presenting with brain WM lesions and the
presence of cortical lesions should alert physician to a possible
diagnosis of MS. New imaging features, such as the recently
described Bcentral venous sign,^ may help the clinician to dif-
ferentiate migraine from MS [15]. In particular, the presence of
the Bcentral venous sign^ on combined T2*-weighted and
FLAIR images acquired with gadolinium contrast on a 3
Tesla has demonstrated specificity for MS in a number of pop-
ulation, differentiating MS people from migraine patients. A
Neurol Sci

central vein exhibits the following properties on T2*-weighted

images: appears as a thin hypointense line or small hypointense
dot, is positioned centrally in the lesion, has a small apparent
diameter (< 2 mm), and runs entirely or partially through the
lesion. In particular, the venocentric distribution of lesions ex-
ists in all MS clinical phenotypes, with prevalence in
periventricular and deep WM lesions. On the other hand, a
central vein is absent from most of the lesions in patients affect-
ed by migraine, as confirmed in a recent study [16].

Structural imaging

Quantitative non-invasive MRI techniques provided new in-

sights in WM and GM structural alterations. Whether such
abnormalities represent a predisposing migrainous trait or
are the direct consequence of the recurrence of migraine at-
tacks is still an intriguing and debatable argument. The appli-
cation of Voxel-based morphometry (VBM) demonstrated
that patients affected by migraine might have structural brain
abnormalities that extend beyond lesions usually detectable
with conventional MRI sequences. Specific findings were de-
tected in patients affected by migraine both with and without
WM hyperintensities: in particular, decreased GM volume
density has been detected in several temporal, frontal, and
parietal areas, whereas an increased volume was detected in
periaqueductal gray (PAG) and dorsolateral pons (Fig. 1) [17].
Interestingly, the increased GM volume in dorsal pons paral-
lels to the location activated in seminal positron emission to-
mography studies [18, 19]. The strict anatomical
colocalization of functional and structural abnormalities has
supported a potential role of the dorsal pons as migraine
Bgenerator.^ In a seminal study, brain volumetric changes
were investigated in a pediatric migraine cohort. Results
showed reduced GM volume in frontotemporal areas but
greater putamen volume when compared with age-matched
non-migraine controls. Volume in the putamen correlated with
the duration of the disease. The presence of such abnormalities
early in the disease course suggests that brain alterations may
represent a phenotypic biomarker of migraine [20, 21].
Surface-based morphometric found increased cortical thick-
ness of motion-processing visual areas both in patients affect-
ed by migraine with and without aura. These structural chang-
es were particularly evident in V3A area, most likely involved
as a source of spreading changes involved in visual aura. It is
speculated that these structural changes may be due to brain
plasticity or account for the brain hyperexcitability in migraine
patients [22–24]. In order to better understand the dynamics of
these abnormalities, a recent longitudinal study was planned
to map modifications of GM volume in a cohort of patients
affected by migraine in over 4 years of observation. The re-
sults of this study showed that, if compared to controls, pa-
tients with migraine showed higher GM volume of regions of
Fig. 1 Brain regions showing decreased gray matter volume in several
frontal and temporal areas in migraine patients compared to controls.
(Reproduced from Rocca et al., Stroke 2006, with permission)
the frontotemporal lobes and lower volume in cerebellum
(quadrangular lobule) at baseline. At follow-up, patients de-
veloped an increased volume of frontotemporoparietal re-
gions, which was more evident in patients with higher attack
frequency and longer disease duration at baseline (Fig. 2).
Patients with an increased attack frequency at follow-up ex-
perienced both increased and decreased volume of nociceptive
regions. Another interesting finding was that the level of pain
intensity, baseline disease duration, and number of migraine
attacks could influence GM volume changes of cortical visual
areas. These data support the notion that the brain of patients
affected by migraine can be remodeled over time, suggesting
that cellular and molecular mechanisms can occur in response
to patient’s disease severity [25]. Diffusion weight (DW) MRI
is a structural MRI technique that allows to study specific
parameters reflecting the microscopic organization of brain
WM tracts. This technique allows to analyze the so-called
normal-appearing WM in order to detect occult brain damages
and to evaluate if this damage extends beyond visible WM
hyperintensities. In a pivotal study, a selective change of dif-
fusivity measures of the visual pathways was detected in pa-
tients affected by migraine using tractography based on diffu-
sion tensor MRI [26]. In another study, brain microstructural
alterations were detected in right frontal WM although not
correlated with disease duration and attack frequencies [27].

Fig. 2 Brain areas showing gray matter volume modifications over
4 years in patients with migraine compared to controls. a Regions of
increased gray matter volume and b and regions of decreased gray
Functional imaging
Functional imaging techniques can be divided into two dis-
tinct categories: task-based or task-free analysis. The first is
the measure of brain responses to a specific action or stimulus,
the second is the analysis of the FC of the brain at rest (RS
fMRI). Over the past 15 years, the number of RS studies in
migraine patients has increased. This technique allows identi-
fying intrinsic brain activity and connectivity by scanning a
migrainous brain Btask-free^ with three major approaches.
The first is based on independent component analysis, where
group differences can be determined without a previous ana-
tomical hypothesis to identify specific network of interest (da-
ta-driven approach). The second approach is a region-of-
interest analysis, whereas connectivity of particular
predetermined areas is examined based on specific anatomical
knowledge in a sort of hypothesis-driven approach. Finally,
another data-driven method (regional homogeneity) gives the
possibility to evaluate low-frequency BOLD fluctuations to
characterize regional homogeneity and relative connectivity
in neighboring voxels. These specific techniques can focus
on changes in neuronal activity, giving insights into functional
organization of pain-processing networks [28]. A large num-
ber of studies showed different alterations regarding RS FC in
patients affected by migraine, particularly in pain-related re-
gions such as the PAG area and insular cortex. Altered con-
nectivity was detected in various networks, depending on sex,
disease duration, or chronification: in particular, the default
mode network, executive control salience (a core network that
includes limbic and paralimbic regions) and visual networks
are involved. All these data confirm that although single brain
Neurol Sci
matter volume. Abbreviation: GM, gray matter. (Reproduced from
Messina et al., Neurology 2018, with permission)
regions may have a specific role in migraine course, a dys-
function of brain networks seems more likely [29]. Among
different brain networks, a sort of putative driver of migraine
attacks lies in the hypothalamus-brainstem connectivity par-
ticularly in the link to the spinal trigeminal nuclei and the
trigeminovascular system, as well as the Bmigraine generator^
in the dorsal rostral pons. There is also evidence of an abnor-
mal functional organization in networks relevant for emotion-
al, cognitive, and attentional aspect of pain during a migraine
attack. In particular, functional reorganization of brain areas
involved in pain and multisensory processing such as the hip-
pocampus, cingulate cortex, PAG, and cerebellum was de-
scribed, as well as in cognitive and limbic networks that can
have some influence on pain experience and integration in
migraine patients. Moreover, abnormal connectivity between
the posterior thalamus (where the ascending pain pathway
converges) and pain encoding and modulating cortical areas
was reported [30–34]. In recent works, fMRI techniques were
utilized to evaluate the neural basis of pain perception in mi-
graine patients by using different specific nociceptive stimuli.
These techniques gave the possibility to explore the functional
organization of the pain matrix, comprising the cingulate cor-
tex (representing the emotional response to pain), thalamus,
insula, primary and secondary somatosensory cortices
(representing the discriminative aspects of pain perception),
and prefrontal cortex [24]. Admittedly, the perception of pain
emerges from the dynamic flow and integration of informa-
tion, and fMRI may give us reliable information about brain
mechanisms that are related to anticipation, expectation, and
expression of migrainous pain. It is reasonable that, in the next
future, fMRI will provide information about the flow of earlier
Neurol Sci
events activating the pain system in migraine attacks. A
pioneering study demonstrated that during induced and spon-
taneous visual aura, an increased BOLD signal developed
within the V3A area and progressed contiguously and slowly
over the cortex, congruent with the retinotopy of the visual
percept. Following the same retinotopic progression it dimin-
ished [35]. This spreading signal disturbance had striking sim-
ilarity with cortical spreading depression phenomenon, thus
further supporting the theory that cortical spreading depres-
sion was the electrophysiological correlate of visual aura.
Numerous fMRI studies showed also hyper-responsiveness
of the primary and extrastriate visual cortex and a lack of
habituation to visual stimuli in migraine patients with and
without aura. This cortical hyperexcitability may represent
the biological basis for the visual symptoms commonly expe-
rienced by migraine patients during and outside the attacks.
Conclusions
Both structural and functional brain imaging gave us a solid
contribution to a greater understanding of migraine patho-
physiology. Brain imaging is an ideal tool for a noninvasive
evaluation of brain morphology and functionality, in order to
ameliorate both the diagnostic setting and the research field.
MRI techniques demonstrated brain structural and functional
alterations in migraine patients, particularly in areas associated
with pain and sensory elaboration. All these data emphasize
the view that migraine is a severe episodic disturbance of CNS
sensory processing. Even if fMRI has no value in clinical
settings, in the next future, it may be very important to provide
new imaging markers both to better discriminate patients with
complex headache diagnosis and to identify group of patients
most likely to achieve good results to a given treatment.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
References
1. Mansfield P, Grannell PK (1973) MMR diffraction in solids? J Phys
C Solid State Phys 6:L422–L427
2. Lauterbur PC (1973) Image formation by induced local interac-
tions: examples employing nuclear magnetic resonance. Nature
242:190–191
3. Hinshaw WS, Bottomley PA, Holland GN (1977) Radiographic
thin-section imaging of the human wrist by nuclear magnetic reso-
nance imaging. Nature 270:722–723
4. Sandrini G, Friberg L, Coppola G, Jänig W, Jensen R, Kruit M,
Rossi P, Russell D, Sanchez del Rìo M, Sand T, Schoenen J (2011)
Neurophysiological tests and neuroimaging procedures in non-
acute headache (2nd edition). Eur J Neurol 18(3):373–381
5. Colombo B, Dalla Costa G, Dalla Libera D, Comi G (2012) From
neuroimaging to clinical setting: what we have learned from mi-
graine pain? Neurol Sci 33:S95–S97
6. Tedeschi G, Russo A, Tessitore A (2012) Functional neuroimaging
in migraine: usefulness for the clinical neurologist. Neurol Sci 33:
S91–S94
7. Messina R, Filippi M, Goadsby PJ (2018) Recent advances in head-
ache neuroimaging. Curr Opin Neurol 31:379–385
8. Colombo B, Rocca MA, Messina R, Guerrieri S, Filippi M (2015)
Resting-state fMRI functional connectivity: a new perspective to
evaluate pain modulation in migraine? Neurol Sci 36:S41–S45
9. Cooney BS, Grossman MD, Farber RE et al (1996) Frequency of
magnetic resonance imaging abnormalities in patients with mi-
graine. Headache 36:616–621
10. Colombo B, Dalla Libera D, Comi G (2011) Brain white matter
lesions in migraine: what’s the meaning? Neurol Sci 32:S37–S40
11. Tortorella P, Rocca MA, Colombo B, Annovazzi P, Comi G, Filippi
M (2006) Assessment of MRI abnormalities of the brainstem from
patients with migraine and multiple sclerosis. J Neurol Sci 244:
137–141
12. Filippi M, Rocca MA, Ciccarelli O, de Stefano N, Evangelou N,
Kappos L, Rovira A, Sastre-Garriga J, Tintorè M, Frederiksen JL,
Gasperini C, Palace J, Reich DS, Banwell B, Montalban X, Barkhof
F (2016) MRI criteria for the diagnosis of multiple sclerosis:
MAGNIMS consensus guidelines. Lancet Neurol 15:292–303
13. Rovaris M, Bozzali M, Rocca MA, Colombo B, Filippi M (2001)
An MR study of tissue damage in the cervical cord of patients with
migraine. J Neurol Sci 183:43–46
14. Absinta M, Rocca MA, Colombo B, Copetti M, Feo D, Falini A,
Comi G, Filippi M (2012) Patients with migraine do not have MRI-
visible cortical lesions. J Neurol 259:2695–2698
15. Sati P, Oh J, Todd Constable R et al (2016) The central vein sign
and its clinical evaluation for the diagnosis of multiple sclerosis: a
consensus statement from the North American Imaging in Multiple
Sclerosis Cooperative. Nat Neurol 12:714–722. https://doi.org/10.
1038/nmeurol2016.166
16. Solomon AJ, Watts R, Ontaneda D, Absinta M, Sati P, Reich DS
(2018) Diagnostic performance of central vein sign for multiple
sclerosis with a simplified three-lesion algorithm. Mult Scler J
24(6):750–757
17. Rocca MA, Ceccarelli A, Falini A, Colombo B, Tortorella P,
Bernasconi L, Comi G, Scotti G, Filippi M (2006) Brain gray mat-
ter changes in migraine patients with T2-visible lesions: a 3-T MRI
study. Stroke 37(7):1765–1770
18. Zhang J, Wu Y-L, Su J, Yao Q, Wang M, Li GF, Zhao R, Shi YH,
Zhao Y, Zhang Q, Lu H, Xu S, Qin Z, Cui GH, Li J, Liu JR, du X
(2017) Assessment of gray and white matter structural alterations in
migraineurs without aura. J Headache Pain 18. https://doi.org/10.
1186/s10194-017-0783-5
19. Tso AR, Goadsby PJ (2015) Recent neuroimaging advances in the
study of primary headaches. Curr Pain Headache Rep 19:15. https://
doi.org/10.1007/s11916-015-0487-y
20. Rocca MA, Messina R, Colombo B, Falini A, Comi G, Filippi M
(2014) Structural brain MRI abnormalities in pediatric patients with
migraine. J Neurol 261:350–357
21. Messina R, Rocca MA, Colombo B, Pagani E, Falini A, Comi G,
Filippi M (2015) White matter microstructure abnormalities in pe-
diatric migraine patients. Cephalalgia 35(4):1278–1286
22. Da Silva AF, Granziera C, Snjder J, Hadjikhani N (2007)
Thickening in the somatosensory cortex of patients. Neurology
69(21):1990–1995
23. Granziera C, Da Silva AF, Snyder J et al (2006) Anatomical alter-
ations of the visual motor processing network in migraine with and
without aura. PLoS Med 3(10):e402

24. Maleki N, Becerra L, Brawn J, Bigal M, Burstein R, Borsook D
(2012) Concurrent functional and structural cortical alterations in
migraine. Cephalalgia 32(8):607–620
25. Messina R, Rocca MA, Colombo B, Pagani E, Falini A, Goadsby
PJ, Filippi M (2018) Gray matter modifications in migraine: a cross-
sectional and longitudinal study. Neurology 91(3):e280–e292.
https://doi.org/10.1212/WNL.000000000005819
26. Rocca MA, Pagani E, Colombo B, Tortorella P, Falini A, Comi G,
Filippi M (2008) Selective diffusion changes of the visual pathway
in patients with migraine: a 3-T tractography study. Cephalalgia 28:
1061–1068
27. Rocca MA, Colombo B, Inglese M et al (2003) A diffusion tensor
magnetic resonance imaging study of brain tissue from patients
with migraine. J Neurol Neurosurg Psychiatry 74(4):501–503
28. Russo A, Silvestro M, Tessitore A, Tedeschi G (2018) Advances in
migraine neuroimaging and clinical utility: from the MRI to the
bedside. Expert Rev Neurother 18(7):533–544
29. Russo A, Silvestro M, Tedeschi G, Tessitore A (2017) Physiopathology
of migraine: what have we learned from functional imaging? Curr
Neurol Neurosci Rep 17:95. 10.1007s11910-017-0803-5
30. Chen Z, Chen X, Liu M, Liu S, Ma L, Yu S (2017) Disrupted
functional connectivity of periaqueductal gray subregions in epi-
sodic migraine. J Headache Pain 18:36
Neurol Sci
31. Chong CD, Dumkrieger GM, Schwedt TJ (2017) Structural co-
variance patterns in migraine: a cross-sectional study exploring
the role of hippocampus. Headache 57:1522–1531
32. Yu ZB, Lu YB, Song LH et al (2017) Functional connectivity dif-
ferences in the insular sub-regions in migraine without aura: a
resting-state functional magnetic resonance imaging study. Front
Behav Neurosci 11:124
33. Zhang J, Su J, Wang M, Zhao Y, Zhang QT, Yao Q, Lu H, Zhang H,
Li GF, Wu YL, Liu YS, Liu FD, Zhuang MT, Shi YH, Hou TY,
Zhao R, Qiao Y, Li J, Liu JR, du X (2017) The sensorimotor net-
work dysfunction in migraineurs without aura: a resting state fMRI
study. J Neurol 264:654–663
34. Androulakis XM, Krebs K, Peterlin BL, Zhang T, Maleki N, Sen S,
Rorden C, Herath P (2017) Modulation of intrinsic resting-state
fMRI networks in women with chronic migraine. Neurology 89:
163–169
35. Hadjikhani N, Sanchez Del Rio M, Wu O et al (2001) Mechanisms
of migraine aura revealed by functional MRI in human visual cor-
tex. Proc Natl Acad Sci USA 98(8):4687–4692
Publisher’s note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.