Functional MRI of migraine

Review

Functional MRI of migraine Todd J Schwedt, Chia-Chun Chiang, Catherine D Chong, David W Dodick

Migraine is a disabling neurological condition manifesting with attacks of headache, hypersensitivities to visual, auditory, olfactory and somatosensory stimuli, nausea, and vomiting. Exposure to sensory stimuli, such as odours, visual stimuli, and sounds, commonly triggers migraine attacks, and hypersensitivities to sensory stimuli are prominent during migraine attacks, but can persist with less magnitude between attacks. Functional MRI (fMRI) has been used to investigate the mechanisms that lead to migraine sensory hypersensitivities by measuring brain responses to visual, olfactory, and painful cutaneous stimulation, and functional connectivity analyses have investigated the functional organisation of specific brain regions and networks responsible for sensory processing. These studies have consistently shown atypical brain responses to sensory stimuli, absence of the normal habituating response between attacks, and atypical functional connectivity of sensory processing regions. Identification of the mechanisms that lead to migraine sensory hypersensitivities and that trigger migraine attacks in response to sensory stimuli might help to better understand neural dysfunction in migraine and provide new targets for migraine prevention, and could provide fMRI biomarkers that indicate early responses to preventive therapy.

Introduction Migraine is a disabling and common neurological disorder with a 1-year prevalence of 12% in the general population.1 A migraine attack comprises moderate-tosevere intensity headache, with a combination of nausea, vomiting, and hypersensitivities to visual, auditory, olfactory, and somatosensory stimuli.2,3 Visual, olfactory, and auditory stimuli are also common triggers of migraine attacks.4–6 About a third of patients with migraine have aura associated with at least some of their migraine attacks.7 Although many different neurological symptoms can occur during a migraine aura, visual symptoms are the most common.8 Between migraine attacks, migraineurs often have persistent but less prominent migraine symptoms, including hypersensitivity to visual, auditory, olfactory, and somatosensory stimuli.3 Because migraine is mainly a disorder of brain function, brain functional MRI (fMRI) studies are useful to study the underlying mechanisms of migraine. Although the aura and headache associated with migraine have been attributed to abnormal vasoconstriction and vasodilation of intracranial arteries, the symptoms are now known to be mostly due to brain dysfunction.9 Minor and transient changes in the calibre of extracranial and intracranial arteries might occur during a migraine attack, but such changes are not always a component of migraine.10 fMRI is also useful to study sensory hypersensitivities in migraine. The sensory hypersensitivities that trigger and are present both during and between migraine attacks are specific to migraine and are not noted to the same extent in other headache or pain disorders. A description of the mechanisms underlying these features of migraine should lead to an improved understanding of pathophysiology and possibly to mechanistic distinction of migraine from other headache and pain disorders. Furthermore, fMRI localisation of atypical stimulusinduced activations and atypical functional connectivity in migraineurs might identify targets for preventive www.thelancet.com/neurology Vol 14 January 2015

Lancet Neurol 2015; 14: 81–91 Mayo Clinic, 5777 East Mayo Boulevard, Phoenix, USA (T J Schwedt MD, C-C Chiang MD, C D Chong PhD, Prof D W Dodick MD) Correspondence to: Dr Todd J Schwedt, Mayo Clinic, 5777 East Mayo Boulevard, Phoenix, AZ 85054, USA [email protected]

therapies in migraine. Normalisation of these atypical imaging findings might serve as biomarkers of an early response to preventive therapies in migraine. In the past few years there has been a substantial increase in the number of published migraine fMRI studies, which have helped to identify the location and clinical significance of migraine-associated brain dysfunction. Many studies examined stimulus-induced brain activation; most have used noxious thermal stimulation of the skin, trigemino-nociceptive activation by intranasal ammonia, olfactory stimuli, or visual stimuli. Other studies have used analyses of functional connectivity to investigate the organisation of specific brain regions and functional networks implicated in migraine pathophysiology. Most fMRI studies of migraine have focused on the migraineur between migraine attacks, in the so-called interictal phase, whereas only a few have been done during the migraine attack (ie, in the so-called ictal phase). fMRI studies of migraine have enhanced our understanding of hypersensitivities in migraine, including identification of brain regions and networks that contribute to atypical processing of sensory stimuli. This kind of processing is a key feature of migraine that leads to increased sensitivity to painful and non-painful touch, and visual and olfactory stimuli, and enables typically non-noxious environmental stimuli, such as flashing lights and odours, to trigger migraine attacks. In this Review we summarise the findings of fMRI studies that investigated migraine hypersensitivities, describe how these have helped to clarify understanding of the anatomy and biology of migraine, discuss the limitations of these studies, and propose avenues for future research using fMRI to study migraine.

Painful stimuli Atypical processing of somatosensory stimuli is suggested by physiological studies showing that, during a migraine attack, most migraineurs are hypersensitive to stimulation of the skin that would not normally be 81

Review

considered noxious, and that a proportion of these migraineurs maintain a state of hypersensitivity to somatosensory stimuli between migraine attacks.11,12 Hypersensitivity occurs in body regions innervated by the trigeminal nerve and in extracranial regions, implicating central sensitisation.13,14 Somatosensory hypersensitivity results in the development of cutaneous allodynia in around two-thirds of migraineurs during a migraine attack.15–17 A migraineur with cutaneous allodynia finds normally non-noxious stimuli of the skin to be painful, and thus experiences pain or discomfort in

response to stimuli such as a light touch of the face or scalp, wearing hair in a tight ponytail, wearing heavy earrings or eyeglasses, or a shirt collar buttoned tightly. Brain activation patterns in response to painful stimuli have been investigated in several migraine fMRI studies (table 1). Pain-inducing heat applied to the skin of the head, face, or upper extremity has been used in fMRI studies of migraine. Typically, heat is applied with an MRI-compatible contact thermode with the temperature individualised to each patient to elicit pain of moderate or severe intensity.

Cohort (n)

Region studied

Timing

Stimulus

Main findings

Episodic migraine (24) Control (27)

Whole brain

Interictal

Heat

Stronger activation in lentiform nuclei, fusiform gyrus, subthalamic nucleus, hippocampus, middle cingulate cortex, somatosensory cortex, dorsolateral prefrontal cortex; and weaker activation in precentral gyrus and superior temporal gyrus in episodic migraineurs than in healthy controls

Stankewitz et al (2013)19 Episodic migraine interictal (20) Episodic migraine ictal (10) Control (20)

Whole brain

Interictal and ictal

Ammonia

Activation patterns in anterior insular cortex, middle cingulate cortex, and thalamus suggest sensitisation to pain stimuli in interictal episodic migraineurs and habituation to pain stimuli in controls

Maleki et al (2012)20

High-frequency episodic migraine* (10) Low-frequency episodic migraine† (10)

Hippocampus

Interictal

Heat

Stronger hippocampal deactivation in low-frequency episodic migraineurs than in high-frequency episodic migraineurs

Maleki et al (2012)21

Episodic migraine: Men (11) Women (11)

Whole brain

Interictal

Heat

Stronger activation in insular cortex, primary somatosensory cortex, and putamen in men with episodic migraine than in women with episodic migraine Stronger activation in caudate, superior temporal, superior frontal, precuneus, posterior cingulate, sensory nucleus, and spinal trigeminal nucleus of brainstem in women with episodic migraine than in men with episodic migraine

Maleki et al (2012)22

High-frequency episodic migraine*(10) Low-frequency episodic migraine† (10)

Results reported as ROI: somatosensory cortex, cingulate cortex, anterior insula, and temporal pole

Interictal

Heat

Stronger activation of primary somatosensory cortex and temporal pole; and weaker activation of anterior insular cortex and cingulate gyrus in high-frequency episodic migraine than in low-frequency episodic migraine

Russo et al (2012)23

Episodic migraine (16) Control (16)

Whole brain

Interictal

Heat

Stronger activation in anterior cingulate cortex; and weaker activation in secondary somatosensory cortex and pons in episodic migraineurs than in healthy controls

Maleki et al (2011)24

High-frequency episodic migraine*(10) Low-frequency episodic migraine† (10)

Whole brain

Interictal

Heat

Weaker activation in caudate, putamen, and pallidum in highfrequency episodic migraineurs than in low-frequency episodic migraineurs

Moulton et al (2011)25

Episodic migraine ictal (8) Episodic migraine interictal (8) Control (8)

Whole brain (interictal vs controls) ROI (ictal vs interictal): temporal pole and parahippocampal gyrus

Interictal and ictal

Heat

Weaker activation in dorsolateral pons (probably the nucleus cuneiformis) in episodic migraineurs with ictal allodynia than in controls Stronger activation in temporal pole and parahippocampal gyrus in ictal episodic migraineurs than in interictal episodic migraineurs

Stankewitz et al (2011)26 Episodic migraine interictal (20) Episodic migraine pre-ictal (10) Episodic migraine ictal (13) Control (20)

Whole brain (ictal vs interictal) Spinal trigeminal nuclei (pre-ictal vs ictal vs interictal vs control)

Interictal and ictal

Ammonia

Weaker activation of spinal trigeminal nuclei in interictal or ictal migraineurs than in controls Stronger activation in spinal trigeminal nuclei in pre-ictal than in interictal migraineurs The stronger the activity in the trigeminal nuclei, the closer the migraineur was in days to their next migraine attack

Aderjan et al (2010)27

Episodic migraine (15) Control (15)

Whole brain

Interictal

Ammonia

Decreased activation in prefrontal cortex, anterior cingulate cortex, red nucleus, and ventral medulla from day 1 to day 8 in episodic migraineurs Increased activation in these regions from day 1 to day 8 in controls

Burstein et al (2010)28

Allodynic episodic migraine ictal (8) Allodynic episodic migraine interictal (8)

Thalamus

Interictal and ictal

Heat and brush

Stronger activation of several thalamic regions in ictal phase than in interictal phase

Moulton et al (2008)29

Episodic migraine ictal allodynia (12) Controls (12)

Brainstem

Interictal

Heat

Weaker activation in dorsolateral pons (probably the nucleus cuneiformis) in episodic migraineurs with ictal allodynia than in controls

Schwedt et al (2014)18

ROI=region of interest. *High-frequency episodic migraine is defined as 8–14 headache days per month. †Low-frequency episodic migraine is defined as 1–2 headache days per month.

Table 1: fMRI studies of migraine using pain stimuli

82

www.thelancet.com/neurology Vol 14 January 2015

Review

In 2010, Stankewitz and colleagues30 published a new method of eliciting trigeminal nerve pain in fMRI studies by using intranasal ammonia. According to the investigators, the ammonia, “stimulates the nasal mucosa, leading to irritation of the first and second branches of the trigeminal nerve, resulting in short-lasting, stinging or stabbing pain sensations”.30 Similar to the brain activation in response to noxious heat, pain elicited by intranasal ammonia results in activation in several pain processing regions, including the insular cortex, thalamus, middle cingulate cortex, amygdala, precentral gyrus, calcarine sulcus, cerebellum, middle temporal gyrus, rostral medulla, lower pons, caudate nucleus, supramarginal gyrus, anterior cingulate cortex, postcentral gyrus, and pallidum.30 Studies of thermal pain-induced activation and intranasal ammonia-induced activation in specific brain regions in the interictal period showed that migraineurs have regional brain activation that differs in intensity from controls.18,19,23,25–27,29 Increased thermal pain-induced activation was localised to regions within the temporal pole, parahippocampal gyrus, anterior cingulate cortex, lentiform nuclei, fusiform gyrus, subthalamic nucleus, hippocampus, middle cingulate cortex, somatosensory cortex, and dorsolateral prefrontal cortex (figure 1).18,23,25 Reduced thermal paininduced activation in migraineurs was localised in migraineurs to areas in the secondary somatosensory cortex, precentral gyrus, superior temporal gyrus, and brainstem.18,23,29 Although few patients with migraine have been studied during a migraine attack, some studies have shown increased thermal pain-induced activation of the temporal pole, parahippocampal gyrus and many thalamic areas compared with interictal activation patterns in the same patients.25,28 Thus,

migraineurs have atypical pain-induced activation of brain regions that participate in various aspects of pain processing, including sensory-discriminative, affective, cognitive, and modulatory processing of pain. fMRI data suggest that an imbalance in the facilitation and the inhibition of pain signalling might contribute to hypersensitivity in migraine. Migraineurs in the interictal period who reported symptoms of allodynia during migraine attacks, but no allodynia in the interictal period, had less thermal pain-induced activation in the dorsolateral pons, a region containing the nucleus cuneiformis, than did controls.29 Because the nucleus cuneiformis is predominantly an inhibitory region in the descending pathways, hypoactivation of this region suggests decreased inhibition of the pain response in migraineurs, which could lead to allodynia during a migraine attack. Results from a study27 in which trigemino-nociception in migraineurs was stimulated with ammonia delivered daily for 8 consecutive days support the notion that migraineurs have inadequate inhibition of pain signalling pathways. Recurrent stimulation decreased activation of the prefrontal cortex, rostral anterior cingulate cortex, red nucleus, and ventral medulla in patients with migraine, whereas controls had increased activation in these regions with stimulation.27 The decreased activation in these brain regions, which are involved in endogenous pain control, might represent inadequate pain inhibition in migraineurs and suggests that the recurrent pain of migraine leads to progressively less pain inhibition over time. Furthermore, physiological studies have shown an absence of normal habituation (decrement in response to repetitive stimuli) in the brains of migraineurs in the interictal period in response to various stimuli. The Middle cingulate cortex

Postcentral gyrus

Anterior cingulate cortex

Precentral gyrus

Dorsolateral prefrontal cortex

Secondary somatosensory cortex

Lentiform nuclei Hippocampus Fusiform gyrus Temporal pole Superior temporal gyrus

Parahippocampal gyrus Subthalamic nucleus

Pons Ventral medulla Greater activation in migraine Less activation in migraine

Figure 1: Pain-induced brain activations that differ in migraineurs versus controls Brain regions with interictal pain-induced activity in migraineurs differing from that in controls are depicted on the surface of the brain and midline. Red areas show more activation in migraineurs than in controls. Blue areas show less activation in migraineurs than in controls. Shaded areas do not represent the exact location, size, or extent of differential activity in brain regions.

www.thelancet.com/neurology Vol 14 January 2015

83

Review

decrease in habituation probably relates to dysfunctional inhibition or increased facilitation of sensory information and perhaps further contributes to atypical pain processing in migraine.31 A lack of habituation in interictal migraineurs was shown in an fMRI study19 that exposed migraineurs and healthy controls to recurrent painful stimulation with intranasal ammonia. Controls habituated to the ammonia during the course of the study, whereas the interictal migraineurs showed increased activation in the anterior insula, middle cingulate, and thalamus. These fMRI results lend support to the notion that recurrent attacks of pain or prolonged pain associated with migraine could increase sensitivity to pain interictally because of decreased habituation and the development of sensitisation.

investigated in an fMRI study (table 2).38 No differences in brain activation were noted when interictal migraineurs were compared with healthy controls. However, during spontaneous and untreated migraine attacks, migraineurs had greater activation in the amygdala, insular cortex, temporal pole, superior temporal gyrus, rostral pons, and cerebellum than they did in the interictal state (figure 2). Consistent with the frequent reports of osmophobia during an attack, the study showed hyper-responsiveness to olfactory stimuli in specific brain regions.38 Furthermore, because a region in the rostral pons might be one of the earliest brain regions activated in a migraine attack (ie, a so-called migraine generator), odour-induced activation of the rostral pons in migraineurs might be a mechanism by which odours trigger migraine attacks.

Olfactory stimuli

Visual stimuli

Migraineurs are hypersensitive to odours during and between migraine attacks. 25–43% of migraineurs report olfactory hypersensitivity during an attack, whereas about a third of migraineurs report olfactory hypersensitivity between migraine attacks (table 2).41–43 Furthermore, half of migraineurs report that odours, such as cigarette smoke, perfumes, and certain food smells, can trigger migraines.41,44 The processing of olfactory stimuli (rose odour) by migraineurs during and between migraine attacks was

Migraineurs are sensitive to visual stimuli, such as lights and patterns, during and between migraine attacks. Light of less intensity is needed to cause visual discomfort interictally in migraineurs compared with controls, and light of even less intensity causes visual discomfort during an attack compared with the interictal state.45,46 Around 45% of migraineurs report symptoms of light hypersensitivity in the interictal state, and about 90% report that they have these symptoms during a migraine attack.6,45,47,48 Additionally, 40% of migraineurs report that

Cohort (n)

Region studied

Timing

Stimulus

Main findings

Griebe et al (2014)

Episodic migraine with aura (18) Control (18)

Whole brain

Interictal

Visual (pattern)

Stronger activation in V5, V3, precuneus, and middle frontal, superior occipital, and intraparietal sulcus in migraineurs than in controls

Datta et al (2013)33

Episodic migraine with aura (25) Episodic migraine without aura (25) Control (25)

Whole brain Primary visual cortex, lateral geniculate

Interictal

Visual (pattern)

Stronger activation of primary visual cortex and lateral geniculate in migraineurs with aura than in both other groups

Hougaard et al (2013)34

Episodic migraine with aura (20) Control (20)

Whole brain (symptomatic vs asymptomatic hemisphere)

Interictal

Visual (pattern)

Stronger activation in inferior parietal lobe, superior parietal lobe or intraparietal sulcus, and inferior frontal lobe in symptomatic than in asymptomatic hemisphere; similar results in symptomatic hemisphere compared with controls No difference in activation of asymptomatic hemisphere versus equivalent hemisphere in controls

Antal et al (2011)35

Episodic migraine (24 [12 with aura]) Control (12)

Motionresponsive middle temporal area

Interictal

Visual (pattern and Stronger activation in left superior-anterior portion of middle temporal motion) complex in migraineurs with aura and migraineurs without aura than in controls Weaker activation in inferior-posterior portion of middle temporal complex in migraineurs with aura and migraineurs without aura than in controls

Huang et al (2011)36

Episodic migraine (11 [7 with aura]) Control (11)

Visual cortex

Interictal

Visual (pattern)

Stronger activation in visual cortex in migraineurs than in controls; combined analysis of individuals with and without aura

Martin et al (2011)37

Episodic migraine (19 [7 with aura]) Control (19)

Occipital cortex

Interictal

Visual (light)

Higher number of voxels corresponding to activation (but no increase in signal intensity in migraineurs); combined analysis of individuals with and without aura

Stankewitz et al (2011)38

Episodic migraine interictal (20) Episodic migraine ictal (13) Control (20)

Whole brain

Interictal and ictal

Odour

Brain activation in interictal migraineurs and controls did not differ Stronger activation of amygdala, insular cortex, temporal pole, superior temporal gyrus, rostral pons, and cerebellum in ictal than interictal migraineurs

Vincent et al (2003)39 Episodic migraine with aura (5) Control (5)

ROI not reported

Interictal

Visual (pattern)

Activation of contralateral extrastriate visual cortex shown in more migraineurs than healthy controls

Cao et al (2002)40

Brainstem, occipital cortex

Interictal and ictal

Visual (pattern)

Activation of red nucleus and substantia nigra precedes occipital cortex activation and migraine symptoms in ictal migraineurs

32

Migraineurs with visually triggered attacks (12)

ROI=region of interest. V3=visual area 3. V5=visual area 5.

Table 2: fMRI studies of migraine using visual or olfactory stimuli

84

www.thelancet.com/neurology Vol 14 January 2015

Review

Parahippocampal gyrus (painful heat)

Superior temporal gyrus (odour)

Insular cortex (floor of the Sylvian fissure) (odour)

Posterior thalamus (painful heat)

Amygdala (odour) Occipital cortex (visual)

Temporal pole (painful heat, odour)

Cerebellum (odour)

Pons (odour)

Midbrain: red nucleus and substantial nigra (visual) Greater activation during migraine attacks

Figure 2: Stimulus-induced brain activations that differ in ictal versus interictal migraineurs Brain regions with pain, visual, and odour-induced activation that differ in migraineurs during a migraine versus between migraines are depicted on the surface of the brain and midline. Red areas show more activation during a migraine attack than in the interictal period. No areas have less activity during a migraine attack than in the interictal period. Shaded areas do not represent the exact location, size, or extent of differential activity in brain regions.

visual stimuli can trigger their migraines.7 In view of these associations, several studies have used fMRI to investigate migraineurs’ responses to visual stimuli. Some of these studies specifically investigated patients with migraine who had aura, because visual cortex hyperexcitability might predispose the brain to visual hypersensitivity and visual aura (table 2, figure 3).49 fMRI studies of visual stimuli-induced brain activation noted that migraineurs were hyper-responsive to such stimuli. Results from several studies32,36,37,39 show that migraineurs viewing visually stimulating patterns (eg, black and white stripes) have high activation in the primary and extrastriate visual cortices. Results from one study37 that used light as the stimulus showed migraineurs had a larger photoresponsive area (ie, a greater number of voxels corresponding to areas of brain activation) than did controls, but there was no difference with respect to intensity of the blood oxygenlevel dependent (BOLD) signal. A study33 comparing visual pattern-induced brain activity in migraineurs with aura, migraineurs without aura, and controls, showed that migraineurs with aura had greater activation in the primary visual cortex and lateral geniculate than did the migraineurs without aura or controls. There were no significant differences in brain activation between the migraineurs without aura, and controls. These results lend support to the notion that visual cortex hyperexcitability is associated with migraine with aura. Finally, researchers in one study35 investigated activation of the motion-responsive middle temporal complex in response to a moving visual stimulus and noted that migraineurs had greater activation of the left superior-anterior portion of this region than did non-migraineurs. These data show that hyper-responsiveness to visual stimuli in migraine extends beyond the visual cortex. www.thelancet.com/neurology Vol 14 January 2015

To add to the understanding of how visual stimuli might trigger a migraine attack, brainstem activation in association with visually triggered migraine attacks was studied in 12 migraineurs. Four patients developed migraine with visual aura and eight developed headaches without aura in response to viewing a flickering checkerboard.40 Activation in the red nucleus and substantia nigra was identified in 75% of these migraineurs. Red nucleus and substantia nigra activation preceded the onset of the symptoms of visually triggered migraine and increased signal intensity in the occipital cortex. Similar to the studies of the neural processing of odour, these results indicate that brainstem structures have a role early in or during initiation of a migraine attack and suggest a potential pathway in which exposure to external stimuli could trigger a migraine attack.38 A potential clinical use of fMRI is in the measurement of the early effects of treatment. An fMRI study36 of visual stimulation showed that migraineurs who were shown a striped visual pattern had higher ratings of discomfort and activation in the visual cortex than did nonmigraineurs.36 However, when these migraineurs wore precision ophthalmic tints (ie, eyeglasses with tinted lenses that best reduce visual discomfort for each patient) their visual discomfort decreased and activation in the visual cortex normalised. This study shows that visual hypersensitivity in migraineurs can be reduced, and effectiveness can be measured both with subjective patient reports of reduced visual discomfort and objectively by normalisation of measures of visual cortex hyper-reactivity. Lending support to the notion that normal habituation to recurrent stimuli is absent in migraineurs in the interictal period, haemodynamic refractory effects (ie, less BOLD activity in response to a second stimulus that 85

Review

is delivered in close temporal proximity to the first stimulus) on fMRI were absent in migraineurs in a study in which they were exposed to paired visual stimuli.50 Whereas controls had reduced activation in response to the second stimulation within each pair, haemodynamic refractory effects were not seen in migraineurs. Thus, the absence of interictal habituation in people with migraine goes beyond the processing of pain to include other senses, such as vision.

Functional connectivity in migraine Analysis of functional connectivity with fMRI investigates the functional organisation of the brain based on temporal correlations in BOLD signal fluctuations in different brain regions.51 Most functional connectivity analyses are done when the brain is at rest, when the person being studied is not performing a task and is not being stimulated. In the resting state there is continuous low-frequency fluctuation in the BOLD signal throughout the brain. Brain regions with temporal correlations in BOLD signal are deemed to be functionally connected or functionally communicating.52 The presence of functional connections and the strength of such functional connections can be atypical in the presence of neurological diseases including migraine. Studies of functional connectivity in migraine have consistently shown migraineurs to have aberrant functional organisation in the brain, mostly in regions that participate in the processing of pain.20,22,24,53–67 Many studies have shown positive correlations between the frequency of migraine attacks or number of years with migraine and the extent of atypical functional connectivity.54,56,57,58,60,62–66 A direct relation between migraine and atypical functional connectivity is suggested by these correlations; however, longitudinal studies are needed to establish whether atypical

functional connectivity might predispose individuals to migraine or is the result of recurrent migraines. Functional connectivity studies comparing migraineurs with healthy controls show atypical connectivity between many regions involved in the processing of pain.21–22,53–66 Brain regions shown to have atypical functional connectivity in migraineurs include those involved in sensory-discriminative processing of pain (eg, somatosensory cortex, posterior insula), affectiveemotional processing (eg, anterior insula, anterior cingulate cortex, and amygdala), cognitive processing (eg, hippocampus, parahippocampal gyrus, and orbitofrontal cortex), and pain modulation (eg, periaqueductal grey, nucleus cuneiformis).53,54,56,57,60,61,65–67 Migraineurs have atypical functional connectivity of several resting-state networks including the salience network, default mode network, central-executive network, somatomotor network, and frontoparietal attention network (table 3).53,54,56,57,58,60–67 Two small studies57,59 investigated functional connectivity of regions in the pain-modulating descending pathways of the brainstem in patients with migraine who report symptoms of cutaneous allodynia during their attacks. These studies showed atypical functional connectivity of the periaqueductal grey and nucleus cuneiformis to other pain processing regions in migraineurs with allodynia. Atypical functional connectivity in these brainstem regions suggests that altered functional organisation of areas involved in pain inhibition is associated with the development of central sensitisation and cutaneous allodynia in migraine, but these results will need to be replicated in larger studies.

Limitations of fMRI studies of migraine Several limitations of the published fMRI studies make drawing of conclusions about stimulus-induced brain activity and functional connectivity in migraine difficult.

Superior parietal lobule

Middle frontal gyrus

Intraparietal sulcus Inferior parietal lobule

Precuneus

Lateral geniculate nucleus Inferior frontal gyrus Visual cortex

Greater activation in migraine

Figure 3: Visual stimuli-induced brain activations that differ in migraineurs versus controls Brain regions with interictal visual stimuli-induced activation in migraineurs that differ from those in healthy controls are depicted on the surface of the brain and midline. Red areas show more activation in migraineurs than in controls. No areas have less activity in migraineurs than in controls. Shaded areas do not represent the exact location, size, or extent of differential activity in brain regions.

86

www.thelancet.com/neurology Vol 14 January 2015

Review

First, fMRI studies of migraine generally include small numbers of patients, which limits the statistical power of the study, often resulting in less than optimum methods to determine significance and limiting the generalisability of

the results. Furthermore, there are few replication studies confirming the fMRI results. Additional studies are needed to increase confidence in the results of fMRI studies and to establish the generalisability of these findings.

Cohorts (n)

Analysis

Main findings

Schwedt (2014)59

Migraine with severe allodynia (8) Migraine with no allodynia (8)

ROI: periaqueductal grey and nucleus cuneiformis

Stronger functional connectivity of periaqueductal grey and nucleus cuneiformis to other sensory discriminative regions in brainstem, thalamus, insular cortex, and cerebellum and with high-order pain processing regions in frontal and temporal lobes in migraineurs with severe allodynia than in migraineurs with no allodynia

Hadjikhani (2013)53

Migraine (22) Control (20) Carpal tunnel syndrome (11) Trigeminal neuralgia (9)

ROI: amygdala

Stronger functional connectivity of amygdala to visceroceptive insular cortex in migraineurs than in all other cohorts; stronger functional connectivity of amygdala with insular cortex, secondary somatosensory cortex, thalamus, Heschl’s gyrus, and temporal pole in migraineurs than in controls

Jin (2013)54

Episodic migraine (21) Control (21)

ROI: anterior cingulate cortex

Stronger functional connectivity of anterior cingulate cortex to middle temporal, orbitofrontal cortex, and dorsolateral prefrontal cortex in migraineurs than in contols

Maleki (2013)20

High-frequency episodic migraine* (10) ROI: hippocampus Low-frequency episodic migraine† (10)

Weaker functional connectivity of hippocampus to supramarginal gyrus, temporal pole, fronto-orbital, nucleus accumbens, anterior insular cortex, middle frontal, and paracingulate gyrus in high-frequency episodic migraineurs than in low-frequency epidsodic migraine

Schwedt (2013)60

Chronic migraine (20) Control (20)

ROI: anterior cingulate cortex, anterior insula, and amygdala

Atypical functional connectivity of ROI to pain-facilitating and pain-inhibiting regions in sensory-discriminative, cognitive, and integrative domains (anterior insular cortex, amygdala, pulvinar, mediodorsal thalamus, middle temporal, and periaqueductal grey) in migraineurs

Tessitore (2013)61

Episodic migraine (20) Control (20)

Intrinsic connectivity networks in default mode network

Weaker functional connectivity of superior frontal gyrus and temporal pole to other regions of the default mode network in migraineurs than in controls

Xue (2013)62

Episodic migraine (18) Control (18)

Amplitude of low-frequency fluctuation and functional connectivity in ROI (anterior cingulate cortex, thalamus, prefrontal cortex, and insular cortex)

Decreased amplitude of low-frequency fluctuation in prefrontal cortex, and rostral anterior cingulate cortex and increased amplitude of low-frequency fluctuation in thalamus in migraineurs Stronger functional connectivity of rostral anterior cingulate cortex to frontal lobe and parietal lobe; thalamus with caudate, temporal lobe, and putamen; prefrontal cortex with precuneus, parietal lobe, and temporal lobe; and insular cortex with temporal pole, frontal lobe, and parietal lobe in migraineurs than in controls

Yuan (2013)66

Episodic migraine (40) Control (40)

ROI: basal ganglia

Stronger functional connectivity of caudate nucleus to parahippocampal gyrus, amygdala, insular cortex, and putamen; and nucleus accumbens with parahippocampal, anterior cingulate cortex, orbitofrontal cortex, and posterior cingulate cortex in migraineurs than in controls

Zhao (2013)67

Episodic migraine (40) Control (20)

Regional homogeneity (whole brain)

Abnormal regional homogeneity in thalamus, inferior frontal, middle occipital, insular cortex, caudate, middle frontal gyrus, middle temporal gyrus, inferior occipital gyrus, anterior cingulate cortex, medial frontal lobe, superior temporal lobe, amygdala, lentiform nucleus, uncus, superior frontal lobe, temporal pole, cerebellum, pons, medulla, midbrain, hippocampus, lingual gyrus, cuneus, inferior parietal gyrus, postcentral gyrus, precuneus, fusiform gyrus, and posterior cingulate cortex in migraineurs

Liu (2012)56

Episodic migraine (43) Control (43)

90 ROI

Brain hubs related to pain processing have abnormal nodal centrality (precentral gyrus, inferior frontal gyrus, parahippocampal gyrus, anterior cingulate cortex, thalamus, temporal pole, and inferior parietal gyrus) in migraineurs

Maleki (2012)21

Episodic migraine: Men (11) Women (11)

ROI: insular cortex and precuneus

Stronger negative functional connectivity of insular cortex to primary somatosensory cortex, posterior cingulate cortex, precuneus, temporal pole; and stronger functional connectivity of precuneus with amygdala and primary somatosensory cortex in women with episodic migraine than in men with episodic migraine

Maleki (2012)22

High-frequency episodic migraine* (10) ROI: postcentral gyrus, anterior insula, Stronger functional connectivity of postcentral gyrus to anterior cingulate cortex, posterior Low-frequency episodic migraine† (10) temporal pole, and anterior cingulate insular cortex, pulvinar, parahippocampus, hypothalamus, putamen, frontal pole, and weaker functional connectivity to substantia nigra; stronger anterior cingulate cortex cortex functional connectivity with frontal pole, temporal pole, inferior temporal gyrus, pulvinar, and parahippocampal gyrus; stronger anterior insula functional connectivity with anterior cingulate cortex, putamen, parahippocampal gyrus, hippocampus; stronger temporal pole functional connectivity with postcentral gyrus, middle and superior temporal gyrus, and frontal pole in high-frequency episodic migraine than in low-frequency episodic migraine

Russo (2012)58

Episodic migraine (14) Control (14)

Frontoparietal network

Weaker functional connectivity in right frontoparietal network, specifically in middle frontal gyrus and dorsal anterior cingulate cortex, in migraineurs than in controls

Xue (2012)63

Episodic migraine (23) Control (23)

Independent components analysis of default mode network, central executive network, and salience network

Intrinsic connectivity differed in the three intrinsic connectivity networks in migraineurs compared with healthy controls: greater intranetwork functional connectivity of middle frontal gyrus in the right lateralised central executive network and greater intranetwork functional connectivity of inferior frontal gyrus in the left lateralised central executive network; and decreased intranetwork functional connectivity of supplementary motor area in the salience network; and greater intrinsic default mode network and right central executive network connectivity to right anterior insula (Table 3 continues on next page)

www.thelancet.com/neurology Vol 14 January 2015

87

Review

Cohorts (n)

Analysis

Main findings

(Continued from previous page) Yu (2012)64

Episodic migraine (26) Control (26)

Regional homogeneity (whole brain)

Decreased regional homogeneity in anterior cingulate cortex, prefrontal cortex, orbitofrontal cortex, and supplementary motor area in episodic migraineurs

Yuan (2012)66

Episodic migraine (21) Control (21)

Voxel-mirrored homotopic connectivity and ROI (anterior cingulate cortex)

Weaker interhemispheric functional connectivity in the anterior cingulate cortex in episodic migraineurs than in healthy controls; with bilateral anterior cingulate cortex as ROI, migraineurs had stronger functional connectivity between left anterior cingulate cortex and bilateral orbitofrontal cortex and right dorsolateral prefrontal cortex, and stronger functional connectivity between right anterior cingulate cortex and bilateral orbitofrontal cortex than did controls

Liu (2011)55

Episodic migraine: Men (18) Women (20) Controls (38)

Graph theory analysis (whole brain; 90 ROI)

Abnormal topological organisation, including small-world properties, resilience, nodal centrality, and inter-regional connections in migraineurs compared with controls

ROI: periaqueductal grey

Stronger functional connectivity of periaqueductal grey to ventrolateral prefrontal cortex, supramarginal gyrus, anterior insula, precentral gyrys, postcentral gyrus, and thalamus in episodic migraineurs than in healthy controls Weaker functional connectivity of periaqueductal grey to prefrontal, anterior cingulate cortex, and anterior insula in migraineurs with allodynia than in migraineurs without allodynia

Mainero (2011)57 Episodic migraine (17) Controls (17) Episodic migraine with allodynia (5) Episodic migraine with no allodynia (5) Maleki (2011)24

High-frequency episodic migraine* (10) ROI: basal ganglia, periaqueductal Low-frequency episodic migraine† (10) grey, pulvinar nuclei, and hypothalamus

Weaker functional connectivity of caudate nucleus to middle frontal, insular cortex, temporal pole, and parahippocampus; weaker nucleus accumbens functional connectivity to posterior cingulate cortex, superior parietal, and hippocampus; stronger putamen functional connectivity with hippocampus, caudate, middle frontal gyrus, anterior insula; and stronger globus pallidus functional connectivity to middle temporal lobe, supramarginal gyrus, thalamus, hippocampus, insular cortex, and temporal pole in high-frequency episodic migraineurs than in low-frequency episodic migraineurs

All studies were done when patients with migraine were interictal (ie, between migraine attacks). *High-frequency episodic migraine is defined as 8–14 headache days per month. †Low-frequency episodic migraine is defined as 1–2 headache days per month. ROI=region of interest.

Table 3: Functional connectivity MRI studies of migraine

Substantial variation in methods of data collection and analysis in fMRI studies makes it difficult to develop a cohesive model of the specific brain regions and networks that are atypical in migraine. For example, variation in the methods used does not enable a proper meta-analysis of studies showing brain regions that have atypical activation in response to painful stimuli in people with migraine. The use of analyses based on the region of interest, rather than whole-brain analyses, is a substantial limitation of these studies. Additional fMRI studies using whole-brain analyses are needed to understand atypical stimulus-induced activity and atypical functional connectivity in migraine. Furthermore, the laterality of migraine symptoms should be taken into account in fMRI studies. A study34 comparing the symptomatic brain hemisphere (ie, brain hemisphere contralateral to the hemifield in which visual aura symptoms occur) and the asymptomatic brain hemisphere in migraineurs, and the same brain areas in controls, showed that the symptomatic hemisphere was hyper-reactive to visual stimuli compared with the asymptomatic hemisphere in migraineurs and compared with controls. However, brain activation in response to stimuli did not differ between the asymptomatic hemisphere in migraineurs and the same brain hemisphere in controls. Thus, fMRI studies of migraine activation might need to focus on the usual side of migraine symptoms. Additionally, fMRI studies of migraine vary in the methods used to create a stimulus, the resolution used in MRI, and the thresholds set for statistical significance. 88

Further heterogeneity might also be introduced by the different analytical techniques used in functional connectivity analyses of migraine: 1) region-of-interest (functional connectivity among chosen regions or between chosen regions and the rest of the brain); 2) independent components analysis (masked computational separation of the whole brain into functional networks or into its functional subcomponents); 3) voxel-mirrored homotopic connectivity (measurement of functional connectivity between voxels in one hemisphere of the brain and symmetric counterparts on the other hemisphere as a measure of interhemispheric functional connectivity); and 4) regional homogeneity (measurement of synchronization of fluctuations in BOLD in local voxels). Similarly, the use of medications and the potential effects of comorbidities in migraine (eg, anxiety, depression, and myofascial pain) have not been adequately taken into account in many studies. Differences in the timing of data collection are another limitation in migraine fMRI studies. Migraineurs can be studied during a migraine attack or between migraine attacks. However, the exact intervals between fMRI, the preceding migraine, and the next migraine probably affect the results. Investigators used trigeminonociceptive stimulation by intranasal ammonia to study patients with migraine at several timepoints during and between migraine attacks.26 The recorded strength of spinal trigeminal nuclei activation in response to this stimulus was dependent on the timing of fMRI: interictal migraineurs had weaker activation in the spinal www.thelancet.com/neurology Vol 14 January 2015

Review

trigeminal nuclei than did healthy controls, but activation in the spinal trigeminal nuclei strengthened as the migraineurs neared their next migraine attack. Similarly, if one presumes that brain activation changes as a patient progresses through a migraine attack, the exact timing of fMRI in relation to migraine attacks probably has a substantial effect on the imaging results. Most studies do not explicitly account for sex-specific differences in migraineurs. However, such differences have been shown in migraine fMRI studies.21,55 One study21 showed that women have greater pain-induced activation compared with men in the caudate nucleus, superior temporal lobe, superior frontal lobe, precuneus, posterior cingulate, sensory nucleus, and spinal trigeminal nuclei of the brainstem, whereas men had greater activation than women in the insular cortex, primary somatosensory cortex, and putamen. Studies investigating the differences in functional connectivity between male and female migraineurs showed that functional connectivity differed according to sex.21,55 Sexrelated differences in the topological organisation of resting-state networks and in the functional connectivity of the insular cortex and of the precuneus have been identified. fMRI studies should, therefore, carefully match participants according to sex or investigate male and female migraineurs independently. Longitudinal studies are needed to establish whether the differences in brain activation or functional connectivity detected by fMRI in patients with migraine predispose a person to migraine or result from recurrent migraine attacks. Prospective longitudinal studies to assess associations in migraine patterns (eg, increasing or decreasing headache frequency) with changes in brain activation and changes in functional connectivity, shown by fMRI, might help to elucidate the direction of the relation and help to identify early biomarkers to predict improvements in, or worsening, migraine patterns. Such studies should take into account the effects on brain activation of drugs to treat migraine, ageing, and the development of comorbidities. The large sample sizes needed for this type of analysis almost certainly necessitate multicentre collaborative efforts. The inability to establish whether the results of fMRI studies are specific to migraine or represent other types of pain is a shortcoming in the medical literature on migraine. One functional connectivity study53 compared migraineurs not only with healthy controls, but also with patients with carpal tunnel syndrome and patients with trigeminal neuralgia. The migraineurs had stronger functional connectivity between the amygdala and visceroceptive insula compared with people in the other groups. Future studies comparing migraineurs with patients with other headache types and other pain types are needed to establish the specificity of fMRI results in migraine studies. These studies should focus on brain regions and networks that might contribute to the unique features of migraine, such as the co-occurrence www.thelancet.com/neurology Vol 14 January 2015

and interaction of headache with visual, olfactory, and auditory symptoms. Studies show the presence and intensity of stimuli in one sensory modality (eg, visual) mediate the presence and intensity of symptoms in other sensory domains (eg, intensity of headache pain in migraine), suggesting that multisensory integration might play a part in migraine pathophysiology.3 Thus, fMRI studies of brain regions with roles in multisensory integration might be useful to differentiate migraine from other headache types and other pain disorders. Notably, the temporal pole is involved in multisensory integration and has atypical activation in response to stimuli and atypical functional connectivity in several migraine fMRI studies.20,22,24,25,38,53,56,61,63,67

Conclusions and future research fMRI studies consistently show that migraine is associated with atypical brain activation, in response to painful, olfactory, and visual stimuli, and atypical functional connectivity. Atypical brain activity and functional connectivity involve several areas of the brain, a finding consistent with migraine being a complex neurological disorder with atypical processing of several types of sensory stimuli (somatosensory, visual, and olfactory). fMRI studies of migraine show a combination of enhanced sensory facilitation and reduced inhibition in response to sensory stimuli, and reduced or absent habituation to stimuli interictally. Correlations between the extent of brain abnormalities on fMRI and headache frequency or number of years with migraine suggest that migraine has a cumulative effect on brain function or that the extent of underlying abnormalities on fMRI positively correlates with the risk of more severe migraine. Several avenues exist for future fMRI research on migraine to improve the quality and clinical significance of these data. fMRI studies with large numbers, more stringent statistical analyses, and replicate studies would increase confidence in the validity of study results. Large, multicentre, longitudinal studies are needed to investigate associations between changes in the frequency and number of years with migraine and corresponding changes on fMRI. Such studies would advance understanding of the mechanisms of migraine transformation (moving from less frequent headaches to more frequent headaches) and migraine reversion (moving from more frequent to less frequent headaches) and might help to identify baseline biomarkers predictive of transformation and reversion. Similarly, studies of this kind should investigate the effects of migraine therapies on fMRI measures and could help to identify baseline biomarkers that predict treatment response. Neuroimaging studies that differentiate migraine from other headache types and other types of pain are also needed. Whether the findings from fMRI studies of migraine are specific to migraine or are shared with other headache types and other types of chronic pain is unknown. Investigation of the 89

Review

10

Search strategy and selection criteria We searched PubMed for English language articles on patients with migraine published between 1966 and March 25, 2014. The following search terms were used: “migraine and MRI”, “migraine and fMRI”, “migraine and blood oxygen-level dependent”, “migraine and functional connectivity”. The reference lists of included articles and the authors’ own files were searched for additional articles. Articles that used fMRI to investigate migraine hypersensitivities were considered for inclusion in this Review. Publications were selected for inclusion on the basis of their relevance to the topic, originality, and the extent to which the study results were deemed to contribute to the migraine neuroimaging specialty.

11

12

13

14 15

16

specificity of fMRI findings in migraine is a crucial step before fMRI can be used in the diagnosis and management of patients with migraine. Although migraine is a symptom-based diagnosis, and neuroimaging would typically not be needed to assign a diagnosis, a diagnostic test would be very useful in some situations for which the diagnosis is otherwise uncertain. Analyses of functional connectivity in the resting state might be particularly useful with respect to diagnosis because collection of these imaging data does not require patients to participate in tasks or to be stimulated and the imaging does not require significant acquisition time. Contributors TJS and C-CC searched the medical literature, wrote, edited, and approved the Review. CDC and DWD wrote, edited, and approved the Review.

17 18

19

20

21

22

23

Declaration of interests We declare no competing interests. Acknowledgments Our work is partly funded by a grant from US National Institutes of Health (K23NS070891) to TJS. References 1 Lipton RB, Bigal ME, Diamond M, Freitag F, Reed ML, Stewart WF. Migraine prevalence, disease burden, and the need for preventive therapy. Neurology 2007; 68: 343–49. 2 The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808. 3 Schwedt TJ. Multisensory integration in migraine. Curr Opin Neurol 2013; 26: 248–53. 4 Friedman DI, De ver Dye T. Migraine and the environment. Headache 2009; 49: 941–52. 5 Kelman L. The triggers or precipitants of the acute migraine attack. Cephalalgia 2007; 27: 394–402. 6 Martin PR, Reece J, Forsyth M. Noise as a trigger for headaches: relationship between exposure and sensitivity. Headache 2006; 46: 962–72. 7 Launer LJ, Terwindt GM, Ferrari MD. The prevalence and characteristics of migraine in a population-based cohort: the GEM study. Neurology 1999; 53: 537–42. 8 Russell MB, Olesen J. A nosographic analysis of the migraine aura in a general population. Brain 1996; 119 (pt 2): 355–61. 9 Noseda R, Burstein R. Migraine pathophysiology: anatomy of the trigeminovascular pathway and associated neurological symptoms, cortical spreading depression, sensitization, and modulation of pain. Pain 2013; 154 (suppl 1): 44–53.

90

24 25

26

27

28

29

30

31 32

33

Schoonman GG, van der Grond J, Kortmann C, van der Geest RJ, Terwindt GM, Ferrari MD. Migraine headache is not associated with cerebral or meningeal vasodilatation—a 3T magnetic resonance angiography study. Brain 2008; 131 (pt 8): 2192–200. Schwedt TJ, Krauss MJ, Frey K, Gereau RWt. Episodic and chronic migraineurs are hypersensitive to thermal stimuli between migraine attacks. Cephalalgia 2011; 31: 6–12. Weissman-Fogel I, Sprecher E, Granovsky Y, Yarnitsky D. Repeated noxious stimulation of the skin enhances cutaneous pain perception of migraine patients in-between attacks: clinical evidence for continuous sub-threshold increase in membrane excitability of central trigeminovascular neurons. Pain 2003; 104: 693–700. Burstein R, Yarnitsky D, Goor-Aryeh I, Ransil BJ, Bajwa ZH. An association between migraine and cutaneous allodynia. Ann Neurol 2000; 47: 614–24. Mathew NT, Kailasam J, Seifert T. Clinical recognition of allodynia in migraine. Neurology 2004; 63: 848–52. Ashkenazi A, Silberstein S, Jakubowski M, Burstein R. Improved identification of allodynic migraine patients using a questionnaire. Cephalalgia 2007; 27: 325–29. Bigal ME, Ashina S, Burstein R, et al. Prevalence and characteristics of allodynia in headache sufferers: a population study. Neurology 2008; 70: 1525–33. Lipton RB, Bigal ME, Ashina S, et al. Cutaneous allodynia in the migraine population. Ann Neurol 2008; 63: 148–58. Schwedt TJ, Chong CD, Chiang CC, Baxter L, Schlaggar BL, Dodick DW. Enhanced pain-induced activity of pain-processing regions in a case-control study of episodic migraine. Cephalalgia 2014; 34: 947–58. Stankewitz A, Schulz E, May A. Neuronal correlates of impaired habituation in response to repeated trigemino-nociceptive but not to olfactory input in migraineurs: an fMRI study. Cephalalgia 2013; 33: 256–65. Maleki N, Becerra L, Brawn J, McEwen B, Burstein R, Borsook D. Common hippocampal structural and functional changes in migraine. Brain Struct Funct 2013; 218: 903–12. Maleki N, Linnman C, Brawn J, Burstein R, Becerra L, Borsook D. Her versus his migraine: multiple sex differences in brain function and structure. Brain 2012; 135 (pt 8): 2546–59. Maleki N, Becerra L, Brawn J, Bigal M, Burstein R, Borsook D. Concurrent functional and structural cortical alterations in migraine. Cephalalgia 2012; 32: 607–20. Russo A, Tessitore A, Esposito F, et al. Pain processing in patients with migraine: an event-related fMRI study during trigeminal nociceptive stimulation. J Neurol 2012; 259: 1903–12. Maleki N, Becerra L, Nutile L, et al. Migraine attacks the Basal Ganglia. Mol Pain 2011; 7: 71. 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: 435–48. Stankewitz A, Aderjan D, Eippert F, May A. Trigeminal nociceptive transmission in migraineurs predicts migraine attacks. J Neurosci 2011; 31: 1937–43. Aderjan D, Stankewitz A, May A. Neuronal mechanisms during repetitive trigemino-nociceptive stimulation in migraine patients. Pain 2010; 151: 97–103. Burstein R, Jakubowski M, Garcia-Nicas E, et al. Thalamic sensitization transforms localized pain into widespread allodynia. Ann Neurol 2010; 68: 81–91. Moulton EA, Burstein R, Tully S, Hargreaves R, Becerra L, Borsook D. Interictal dysfunction of a brainstem descending modulatory center in migraine patients. PLoS One 2008; 3: e3799. Stankewitz A, Voit HL, Bingel U, Peschke C, May A. A new trigemino-nociceptive stimulation model for event-related fMRI. Cephalalgia 2010; 30: 475–85. Coppola G, Di Lorenzo C, Schoenen J, Pierelli F. Habituation and sensitization in primary headaches. J Headache Pain 2013; 14: 65. Griebe M, Flux F, Wolf ME, Hennerici MG, Szabo K. Multimodal assessment of optokinetic visual stimulation response in migraine with aura. Headache 2014; 54: 131–41. Datta R, Aguirre GK, Hu S, Detre JA, Cucchiara B. Interictal cortical hyperresponsiveness in migraine is directly related to the presence of aura. Cephalalgia 2013; 33: 365–74.

www.thelancet.com/neurology Vol 14 January 2015

Review

34

35

36

37

38

39

40

41

42

43 44

45 46

47

48

49

50

Hougaard A, Amin FM, Hoffmann MB, et al. Interhemispheric differences of fMRI responses to visual stimuli in patients with side-fixed migraine aura. Hum Brain Mapp 2013; 35: 2714–23. Antal A, Polania R, Saller K, et al. Differential activation of the middle-temporal complex to visual stimulation in migraineurs. Cephalalgia 2011; 31: 338–45. Huang J, Zong X, Wilkins A, Jenkins B, Bozoki A, Cao Y. fMRI evidence that precision ophthalmic tints reduce cortical hyperactivation in migraine. Cephalalgia 2011; 31: 925–36. Martin H, Sanchez del Rio M, de Silanes CL, Alvarez-Linera J, Hernandez JA, Pareja JA. Photoreactivity of the occipital cortex measured by functional magnetic resonance imaging-blood oxygenation level dependent in migraine patients and healthy volunteers: pathophysiological implications. Headache 2011; 51: 1520–28. Stankewitz A, May A. Increased limbic and brainstem activity during migraine attacks following olfactory stimulation. Neurology 2011; 77: 476–82. Vincent M, Pedra E, Mourao-Miranda J, Bramati IE, Henrique AR, Moll J. Enhanced interictal responsiveness of the migraineous visual cortex to incongruent bar stimulation: a functional MRI visual activation study. Cephalalgia 2003; 23: 860–68. Cao Y, Aurora SK, Nagesh V, Patel SC, Welch KM. Functional MRI-BOLD of brainstem structures during visually triggered migraine. Neurology 2002; 59: 72–78. De Carlo D, Dal Zotto L, Perissinotto E, et al. Osmophobia in migraine classification: a multicentre study in juvenile patients. Cephalalgia. 2010; 30: 1486–94. Demarquay G, Royet JP, Giraud P, Chazot G, Valade D, Ryvlin P. Rating of olfactory judgements in migraine patients. Cephalalgia 2006; 26: 1123–30. Kelman L. Osmophobia and taste abnormality in migraineurs: a tertiary care study. Headache 2004; 44: 1019–23. Zanchin G, Dainese F, Trucco M, Mainardi F, Mampreso E, Maggioni F. Osmophobia in migraine and tension-type headache and its clinical features in patients with migraine. Cephalalgia 2007; 27: 1061–68. Main A, Dowson A, Gross M. Photophobia and phonophobia in migraineurs between attacks. Headache 1997; 37: 492–95. Vanagaite J, Pareja JA, Storen O, White LR, Sand T, Stovner LJ. Light-induced discomfort and pain in migraine. Cephalalgia 1997; 17: 733–41. Russell MB, Rasmussen BK, Fenger K, Olesen J. Migraine without aura and migraine with aura are distinct clinical entities: a study of four hundred and eighty-four male and female migraineurs from the general population. Cephalalgia 1996; 16: 239–45. Wober-Bingol C, Wober C, Karwautz A, et al. Clinical features of migraine: a cross-sectional study in patients aged three to sixty-nine. Cephalalgia 2004; 24: 12–17. Aurora SK, Ahmad BK, Welch KM, Bhardhwaj P, Ramadan NM. Transcranial magnetic stimulation confirms hyperexcitability of occipital cortex in migraine. Neurology 1998; 50: 1111–14. Descamps B, Vandemaele P, Reyngoudt H, et al. Absence of haemodynamic refractory effects in patients with migraine without aura: an interictal fMRI study. Cephalalgia 2011; 31: 1220–31.

www.thelancet.com/neurology Vol 14 January 2015

51

52

53

54 55

56

57

58

59

60

61

62

63

64

65

66

67

Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci USA 2005; 102: 9673–78. Cohen AL, Fair DA, Dosenbach NU, Miezin FM, et al. Defining functional areas in individual human brains using resting functional connectivity MRI. Neuroimage 2008; 41: 45–57. Hadjikhani N, Ward N, Boshyan J, et al. The missing link: Enhanced functional connectivity between amygdala and visceroceptive cortex in migraine. Cephalalgia 2013; 33: 1264–68. Jin C, Yuan K, Zhao L, et al. Structural and functional abnormalities in migraine patients without aura. NMR Biomed 2013; 26: 58–64. Liu J, Qin W, Nan J, Li J, Yuan K, Zhao L, et al. Gender-related differences in the dysfunctional resting networks of migraine suffers. PLoS One 2011; 6: e27049. 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. Mainero C, Boshyan J, Hadjikhani N. Altered functional magnetic resonance imaging resting-state connectivity in periaqueductal gray networks in migraine. Ann Neurol 2011; 70: 838–45. Russo A, Tessitore A, Giordano A, et al. Executive resting-state network connectivity in migraine without aura. Cephalalgia 2012; 32: 1041–48. Schwedt TJ, Larson-Prior L, Coalson RS, et al. Allodynia and descending pain modulation in migraine: a resting state functional connectivity analysis. Pain Med 2014; 15: 154–65. Schwedt TJ, Schlaggar BL, Mar S, et al. Atypical resting-state functional connectivity of affective pain regions in chronic migraine. Headache 2013; 53: 737–51. Tessitore A, Russo A, Giordano A, et al. Disrupted default mode network connectivity in migraine without aura. J Headache Pain 2013; 14: 89. Xue T, Yuan K, Cheng P, et al. Alterations of regional spontaneous neuronal activity and corresponding brain circuit changes during resting state in migraine without aura. NMR Biomed 2013; 26: 1051–58. Xue T, Yuan K, Zhao L, et al. Intrinsic brain network abnormalities in migraines without aura revealed in resting-state fMRI. PLoS One 2012; 7: e52927. Yu D, Yuan K, Zhao L, et al. Regional homogeneity abnormalities in patients with interictal migraine without aura: a resting-state study. NMR Biomed 2012; 25: 806–12. Yuan K, Qin W, Liu P, et al. Reduced fractional anisotropy of corpus callosum modulates inter-hemispheric resting state functional connectivity in migraine patients without aura. PLoS One 2012; 7: e45476. Yuan K, Zhao L, Cheng P, et al. Altered structure and resting-state functional connectivity of the basal ganglia in migraine patients without aura. J Pain 2013; 14: 836–44. 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.

91