Functional neuroimaging of headaches

Functional neuroimaging of headaches

Reviews Functional neuroimaging of headaches Margarita Sánchez del Rio and Juan Alvarez Linera Functional neuroimaging, mainly PET and functional MRI...

828KB Sizes 0 Downloads 13 Views

Reviews Functional neuroimaging of headaches

Margarita Sánchez del Rio and Juan Alvarez Linera Functional neuroimaging, mainly PET and functional MRI, is the main tool that allows the capturing of neurovascular events during a headache attack. In migraine, functional imaging has clarified the underlying pathophysiology of the visual aura, whereas in migraine without aura, brainstem findings suggest a dysfunctional pain system. In cluster headache, the activation and morphological changes seen in a region posterior and inferior to the hypothalamus has provided a useful therapeutic target using deep-brain stimulation. We will discuss the main neuroimaging findings pertaining to the pathophysiology of these two common headache disorders, migraine and cluster headache.

Meninges-dural vessels 2 NO Ach VIP



Lancet Neurol 2004; 3: 645–51


The diagnosis of the various headache disorders is based on clinical features as described by the International Headache Society criteria.1 Functional neuroimaging of patients with headache is done to study the underlying pathophysiology rather than for diagnostic purposes. The task of imaging a paroxysmal disorder of relatively short duration is challenging, and more so if spontaneous attacks are to be captured. These factors therefore determine the selection of the imaging technique and study design. Despite the inherent difficulties, a few studies have provided important information about the pathophysiology of migraine and cluster headache. Functional activation of brain regions are thought to be indicated by increases in the regional cerebral blood flow (CBF) in PET and perfusion weighted imaging (PWI) studies, and in the blood-oxygen-level dependent (BOLD) signal in functional MRI. Therefore, blood flow is a useful surrogate to detect synaptic activity, without determining whether the underlying physiological event is excitation or inhibition, or any other energy-consuming process. Before reviewing the recent neuroimaging studies, we will briefly describe the structures involved in the transmission and modulation of the cephalic pain signal. The trigeminovascular system is key since headaches depend on the activation of this pathway. The trigeminovascular system is comprised of sensory fibres of the ophthalmic division of the trigeminal nerve that densely innervate the blood vessels of the dura mater (figure 1). Paradoxically, it is the only system that has not been shown to be activated during a headache attack. A possible explanation is that brainstem motion due to cardiac pulsation and gross head motion may produce a significant artefact that impairs analysis of the signal from structures in this location. Trigeminal fibres investing the dural vessels are not within

Neurology Vol 3 November 2004






Figure 1. Pathogenetic theory of migraine with and without aura. Migraine with aura: (1) the primary event occurs in the cortex with cortical spreading depression (CSD) and release of NO, K+, and adenosine, which are released into the extracellular and perivascular space and (2) reach the trigeminal fibres in the dura mater. (3) The trigeminal fibres are sensitised, which leads both to a sterile inflammation in the meninges and to the transmission of the signal antidromically to the trigenimal nucleus caudalis (TNC). (4) From the TNC, the signal is transmitted to the thalamus and from here to the cortex for the subjective perception of pain. Connections from the TNC to the superior salivary nucleus (SSN) would control dural vasodilatation and autonomic symptoms observed in migraine. Other nuclei, such as the hypothalamus (Hyp), periaqueductal grey matter (PAG), and locus coeruleus modulate the pain signal. Migraine without aura: (4) the primary event is thought to be in the brainstem in dorsal midbrain regions where structures, such as the dorsal raphe nucleus, locus coeruleus, and PAG modulate pain. (3) A dysfunction of these structures would lead to an anomalous activation of the trigeminal system. Ach=acetylcholine; VIP=vasoactive intestinal protein; NK=neurokinine; CGRP=calcitonin gene-related peptide; SP=substance P; TG=trigeminal ganglion; SPG=sphenopalatine ganglion.

the resolution of the currently available imaging techniques. As a consequence, other nuclei distributed within the brainstem need to be imaged. To interpret the headache data it is necessary to take into account pain structures known to participate in painful conditions other than headache. MSdR is Director of the Headache Programme, at the Department of Neurology, and JAL is at the Department of Neuroradiology, Hospital Ruber Internacional, Madrid, Spain. Correspondence: Dr Margarita Sánchez del Rio, Neurology Department, Hospital Ruber Internacional, C/ La Masó 38, 28034 Mirasierra, Madrid, Spain. Tel +34 91 387 52 50; fax +34 91 387 53 33; email [email protected]

For personal use. Only reproduce with permission from Elsevier Ltd



Functional neuroimaging of headaches

Migraine Migraine with aura

The visual aura is one of the most spectacular and feared symptoms reported during a migraine attack. This has prompted researchers to study the underlying pathogenetic mechanisms. Previous limitations to studying acute migraine attacks have included the difficulty in the triggering of typical attacks under controlled conditions, and studying the actual event with appropriate techniques. Since Milner associated the migraine visual aura with cortical spreading depression (CSD) in 1958,2 no study had clearly corroborated this hypothesis. Until the mid-1990s, the possibility of a primary vascular event attaining ischaemic levels still prevailed. CBF studies with single photon emission CT (SPECT), subject to radiation artefacts, have supported this hypothesis.3 Figure 2. Spreading suppression of cortical activation during migraine aura. (A) Shows the progression over 20 min of the scintillations and the visual field defect affecting the left hemifield, as described by the patient. (B) A reconstruction of the same patient’s brain, based on anatomical magnetic resonance (MR) data. The posterior medial aspect of occipital lobe is shown in an inflated cortex format. In this format, the cortical sulci and gyri appear in darker and lighter grey, respectively, on a computationally inflated surface. MR signal changes over time are shown to the right. Each time course was recorded from one in a sequence of voxels that were sampled along the calcarine sulcus, in the primary visual cortex (V1), from the posterior pole to more anterior location, as indicated by the arrows. A similar BOLD response was found within all of the extrastriate areas, differing only in the time of onset of the MR perturbation. The MR perturbations developed earlier in the foveal representation, compared with more eccentric representations of retinotopic visual cortex. This finding was consistent with the progression of the aura from central to peripheral eccentricities in the corresponding visual field (A and C). (C) The MR maps of retinotopic eccentricity from this same patient, acquired during interictal scans. As shown in the upper left corner, voxels that show retinotopically specific activation in the fovea are coded in red (centred at 1·5° eccentricity). Parafoveal eccentricities are shown in blue, and more peripheral eccentricities are shown in green (centred at 3·8° and 10·3°, respectively). Reproduced with permission from the National Academy of Sciences.5

Relatively well established functional networks control different aspects of pain: (1) the lateral pain pathway that includes the trigeminal system, ventral posterior nucleus of the thalamus, and somatosensory cortex, which is involved in localisation of pain and for motor defence system; (2) the hypothalamus, periaqueductal grey matter, and ventral tegmental area, which control the autonomic and antinociceptive component of pain; (3) emotional and affective responses, which are modulated mostly by the amygdala and hippocampus; (4) arousal and attention, which are encoded in the anterior nuclei of the thalamus; and (5) motor preparation to escape pain, which is controlled by the basal ganglia and cerebellum. There is substantial variability in the pattern of activation of pain nuclei among studies. Only the anterior cingulated cortex and insula are consistently activated. The functions attributed are the processing of pain unpleasantness, attention and planning, and evaluation of emotional conflict. In summary, pain triggers complex responses that can be functionally categorised into sensory, adaptive, and affective components.


BOLD functional MRI

The introduction of BOLD functional MRI allowed almost continuous study of brain activity, which provided information not only on vascular but also on neuronal metabolic responses. Furthermore, BOLD functional MRI provides the best ratio between temporal and spatial resolution among available imaging techniques. The possibility of imaging the typical visual aura in migraine with BOLD functional MRI has revealed multiple neurovascular events in the occipital cortex within a single attack that closely resemble CSD: (1) an initial hyperaemia lasting 3·0–4·5 min, spreading at a rate of 3·5 mm per min, (2) followed by mild hypoperfusion lasting 1–2 h, (3) an attenuated response to visual activation, and (4) like CSD, in migraine aura, the first affected area is the first to recover (figure 2). Although this technique does not directly record electrical activity in the cortex to measure the spreading depolarisation that defines CSD,4,5 all the characteristics listed above are nevertheless seen only in CSD and not in any other neurovascular phenomena. To further corroborate these findings, studies with magnetoencephalography, which is able to measure changes in cortical magnetic fields, have also revealed that migraine visual aura is associated with shifts in direct current neuromagnetic field potentials similar to those seen during CSD.6,7 PWI

PWI estimates haemodynamic changes based on signal loss caused by the first pass of a bolus injection of a paramagnetic contrast agent through the brain parenchyma. With this technique, multiple measurements are made possible during a migraine attack without the need of a radioactive tracer. Seven spontaneous visual auras were imaged as early as 20 min into the attack, as well as seven headache phases after the aura.8,9 There was a 27% (range 14–35%) and 15% (range 2–33%) decrement in regional CBF and regional cerebral blood volume, respectively, and an increase in the mean transit time of 32% (±0·3) in occipital cortex contralateral to the affected hemifield.8,9 Changes in blood flow persisted up to 2·5 h into the headache phase (figure 3). Other cortical and brainstem regions did not show significant perfusion changes.

Neurology Vol 3 November 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Functional neuroimaging of headaches


PWI and diffusion weighted imaging (DWI) have been used extensively to study rCBF rCBV MTT cerebral-tissue ischaemia and its viability. Many factors are known to play an important part in the degree of tissue lesion, including the duration of the haemodynamic changes and the degree of the blood flow decrease, measured with PWI, and the local metabolic Visual and haemostatic changes, measured by the aura apparent diffusion coefficient (ADC). It is generally accepted that DWI measures the irreversible damaged infarct core as areas with decreased ADC, whereas PWI measures the 37% 20% 31% complete area of hypoperfusion. The mismatch between the PWI and DWI maps is a frequently used technique to identify the ischaemic penumbra. The PWI studies in migraine have confirmed that CBF changes remained well above the threshold associated with ischaemic injury (>50% decrease in CBF) Interictal and DWI showed no changes in ADC.8 Until recently, a source of controversy has been the link of migraine aura with the subsequent headache. Recent research has shown how a cortical event such as CSD, which Figure 3. Perfusion-weighted imaging of migraine aura. Regional cerebral blood flow is implicated in migraine aura, is able to (rCBF), regional cerebral blood volume (rCBV), and mean transit time (MTT) maps obtained 35 min after onset of visual aura affecting the contralateral visual hemifield to the perfusion activate trigeminal meningeal afferents and defect. White arrows point to the perfusion defect. The changes in rCBF (–37%), rCBV trigger brainstem events consistent with the (–20%), and MTT (+31%) were estimated with respect to the same side interictally. The development of headache.10 CSD induces an upper panel shows the study during the visual aura symptoms. The lower panel represents increase in cortical metabolic activity, which the interictal study of the same patient. releases neurotransmitters and metabolites that are able to reach the perivascular trigeminal fibres. The out ischaemia and to better characterise the underlying activation of the trigeminal fibres leads to an inflammatory pathophysiology. PWI studies have shown either normal process in the dura mater known as neurogenic CBF or unilateral hyperperfusion.17–20 DWI, which is inflammation (reviewed by Pietrobon and Striessnig11). particularly sensitive to water movement, has been either Unfortunately until now, attempts to image neurogenic normal18,21 or altered, showing a decreased water diffusion inflammation in humans have proved fruitless. However, in contralateral to the hemiparesis, in turn indicating that one case of typical migraine with sensorimotor aura12 and at cerebral oedema is most likely.22,23 To date, the only reported least three cases of prolonged migraine aura, cortical focal PET study that used radiolabelled glucose in a patient with hyperaemia and augmented vasogenic leakage from the FHM after 6 days of hemiparesis showed reversible leptomeningeal vessels has been shown with multiple hypometabolism on the contralateral frontobasal cortex, imaging techniques (MRI, SPECT, and angiogram).13,14 caudate nucleus, and thalamus.18 In this study, there was Imaging in all cases was done during the symptoms. slowing of direct current neuromagnetic fields over the Intraparenchymal extravasation of the contrast agent contralateral hemisphere during active symptoms, which showed a mild breakdown of the blood–brain barrier seen returned to normal with the clinical improvement. In only 2–4 h after the administration of gadolinium. The timing for two cases, angiography has shown vasospasm without other the detection of the extravasation was a key factor since the data suggestive of ischaemia, one case in a pregnant woman21 impairment of the blood–brain barrier in migraine is mild, and the second in an 11-year-old girl with coma and left hemiparesis.24 In another study, the angiogram showed as expected. Familiar hemiplegic migraine (FHM) is a dominantly vasodilation of branches of the middle and posterior cerebral inherited subtype of migraine with aura characterised by arteries over the hyperperfused cerebral hemisphere.25 Taken motor weakness. FHM is caused either by missense mutation together, these data provide evidence for a primary neuronal in CACNA1A (50% of families), the gene that encodes the reversible dysfunction in FHM. pore-forming α1-subunit of a P/Q-type calcium channel,15 or by mutations in the gene ATP1A2 that encodes the ␣2 Migraine without aura subunit of the Na+/K+ pump (15% of families).16 Increased A hitherto unsolved problem is how to explain the duration of hemiparesis has raised the possibility of an pathogenesis of headache in patients with migraine without underlying ischaemic insult. This has motivated the use of aura. One hypothesis defends the presence of a clinically multimodal functional imaging during the attacks to rule silent CSD that is able to activate the trigeminal system as in

Neurology Vol 3 November 2004

For personal use. Only reproduce with permission from Elsevier Ltd


Review migraine with aura. Evidence in favour of this hypothesis is based on the results of three studies showing cortical perfusion changes similar to those seen in migraine aura. The first study is a single case of migraine without aura imaged with PET using 15O-labelled water.26 A bilateral decrease in regional CBF was observed to spread forward across the cortical surface from visual associative cortex to parietal and occipitotemporal areas at a relatively constant rate. This result supports the conjecture that subclinical spreading hypoperfusion can occur in migraine without aura, but does not reach a threshold for perceptual deficits. A PET study27 in nine patients, who were studied within 13·3 h of the onset of a spontaneous migraine attack without aura, showed global hypoperfusion. However, the study was not designed to detect temporal changes in CBF. Finally, a study with BOLD functional MRI during five attacks of migraine without aura triggered with visual stimulus (alternating red and green checkerboard pattern) revealed a spreading hyperaemia.4,28 Both the increase and decrease in blood-flow changes slowly proceeding along the cortex are characteristic of CSD.29 Only one study with PWI in 14 attacks of spontaneous migraine without aura studied 1–11 h after onset (mean 4·5 h) did not show significant changes in CBF.9 The timing of the acquisition of the images and the technique used may explain the apparently different results. A second theory proposes a primary brainstem dysfunction as the origin of migraine without aura. Since the initial PET study performed by Weiller and colleagues30 in nine patients with spontaneous migraine without aura, further studies have shown activation in brainstem structures, although not at the same location.31,32 In all cases two main conditions are compared: pain (either spontaneous or triggered with nitroglycerine), and absence of pain. Weiller and colleagues30 showed increased regional CBF in regions that could well correspond to the dorsal raphe nucleus, periaqueductal grey, and locus coeruleus. A second study, in a single patient during a migraine attack triggered with nitroglycerine, showed activation in the dorsal rostral pons.31 Furthermore, BOLD functional MRI studies in migraine without aura have shown activation in the red nucleus and substantia nigra.28 More recently, eight patients with chronic migraine were studied with PET while switching on (no pain) and off (pain) a bilateral suboccipital stimulator implanted for pain control. A significant increase in regional CBF that correlated with pain scores was observed in the dorsal rostral pons, anterior cingulated cortex, and cuneus.32 Interestingly, no increase in regional CBF was reported in periaqueductal grey or other midbrain regions. The finding of pontine activation in two studies31,32 has suggested its relevance in migraine pathogenesis. It is important to note that other facial pain conditions also showed activation of brainstem structures as part of the autonomic and antinociceptive response to pain.33 Two main arguments have been used in favour of the brainstem as migraine generator. First, brainstem activation is specific to migraine without aura because it is not observed in other headache disorders. In both cluster headache34 and short-lasting neuralgiform headache with


Functional neuroimaging of headaches

conjunctival injection and tearing (SUNCT),35 specific activation lies posterior and inferior to the hypothalamus. In an acute experimental model of head pain in healthy volunteers, injection of capsaicin in the forehead, which activated first division trigeminal afferents, did not show brainstem activation.36 Other structures also related with pain processing, such as insula, thalamus, and cingulated cortex, were activated in these studies. Evidence against this specificity is the substantial variability in the site of activation within the brainstem. Activation has been described in nuclei located in the midbrain, such as the red nucleus, substantia nigra28 or locus coeruleus, dorsal raphe nucleus, and periaqueductal grey,30 and most recently in the dorsolateral pons.31 The absence of an apparent structural abnormality with automated segmentation techniques most likely confirms this variability.37 The second argument in favour of a brainstem generator is the persistence of activation in the midbrain after successful treatment with sumatriptan.30 Activation in cingulate cortex and in visual and auditory association cortices seen during the attack were not detectable after treatment and were thought to be secondary to pain activation. Nevertheless, the possibility remains that the persistent midbrain activation may be necessary for a sustained pain-free response after treatment, and its absence may result in recurrence of the attack. There is evidence in favour of this last argument, since there was an almost immediate re-emergence of pain when the stimulator was switched off in patients with chronic migraine and bilateral suboccipital stimulators.32 To date, these data indicate that the most rational conclusion is that different brainstem structures participate in headache pathogenesis, probably in a dysfunctional mode, either by lowering the threshold, which renders the system hyperexcitable, or by decreasing the inhibitory nociceptive pathways. Increased iron deposition in the periaqueductal grey matter in patients with chronic daily headache supports the repeated activation of the antinociceptive network.38

Cluster headache Cluster headache is a unique primary pain syndrome characterised by strictly unilateral severe periorbital pain with ipsilateral autonomic phenomena. Some of the most striking features of this syndrome are the circadian and circannual rhythmicity of attacks, the relapsing-remitting course, and abnormal rhythms of hormones such as cortisol, testosterone, growth hormone, or endorphins.39 All these alterations have pointed to the involvement of the hypothalamus as the primum movens structure in cluster headache.40,41 The first branch of the trigeminal system may be responsible for the periorbital distribution of the pain, whereas the parasympathetic system would be responsible for the autonomic accompaniments observed (figure 4). The possibility of imaging an attack has proved of great value in confirming the a priori hypothesis of the pathogenetic involvement of the hypothalamic area. Only recently has cluster headache been imaged with PET. The triggering of cluster headache with nitroglycerine has allowed comparison of a baseline condition (before pain) with pain condition, and after treatment of the headache.

Neurology Vol 3 November 2004

For personal use. Only reproduce with permission from Elsevier Ltd


Functional neuroimaging of headaches

Significant increase in regional CBF was seen during pain in the ipsilateral Meninges-dural vessels inferior hypothalamic grey matter, the contralateral ventral posterior thalamus, the anterior cingulate cortex, NK and bilaterally in the insula.34,42 This CGRP region located posterior and inferior to SP the hypothalamus (figure 5) is argued NO to be specific to cluster headache Ach VPM because it is imaged during active VIP cluster pain but not in pain-free conditions (in healthy volunteers and RN PAG patients with cluster headache studied Hyp outside an active period, when SN SSN VTA nitroglycerine was not able to trigger pain). Curiously, a previous PET study in patients with cluster headache SPG failed to detect activity in this Autonomic symptoms TNC 2 TG region.43 Certainly, activation of the hypothalamus per se has been reported Pain in many pain conditions other than cluster headache, including after injecting ethanol in the forearm of Figure 4. Proposed pathogenetic theory of cluster headache. The hypothalamus (Hyp) is a key healthy volunteers,44 or in the structure in cluster headache. It can explain the circadian rhythmicity of attacks and the neuroendocrine dysfunction observed in cluster headache patients. The hypothalamus has important mandibular region of trigeminal connections with the pain modulation system of the brainstem: trigeminal nucleus caudalis (TNC), supply.45 periaqueductal grey matter (PAG), ventral tegmental area (VTA). The pain is likely to be mediated by With the knowledge that the activation of the ophthalmic division of the trigeminal nerve, whereas the autonomic symptoms are hypothalamus is involved in the caused by activation of the cranial parasympathetic nerve. NK=neurokinine; CGRP=calcitonin genepeptide; SP=substance P; Ach=acetylecholine; VIP=vasoactive intestinal peptide; RN=raphe autonomic response to pain, it is related nucleus; TG=trigeminal ganglion; SPG=sphenopalatine ganglion; SSN=superior salivary nucleus. unsurprising to find increased regional CBF in this region. Present controversy lies in conjugated ocular deviation and thoughts of impending discriminating the precise location of this activation, which doom, elicited during macrostimulation of the region lies in midbrain tegmentum, in close proximity to the posterior and inferior to the hypothalamus, resemble those hypothalamus but is not hypothalamus per se (figure 5). observed during midbrain stimulation.48 Therefore, on the The anatomical boundaries of the hypothalamus extend from lamina terminalis in the region of the optic chiasm to the mamillary bodies, and include the structures forming A B the lateral walls and the floor of the third ventricle. The functional boundaries of the hypothalamus are less well defined. The hypothalamus has extensive fibre connections Y=⫺18 X=⫺2 with other brain regions involved in visceral activities. Among these connections are fibres originating and C D departing to the midbrain and tegmentum that run through Hyp Hyp the periaqueductal grey matter. These brain regions are well RN established to participate in pain signal processing. Electrical stimulation of the periventricular grey–periaqueductal grey, a region with high-level opiate MB (⫺2,⫺18,⫺8) Z=⫺8 binding, exerts analgesia ascribed to the release of endogenous opioids and the activation of descending inhibitory systems. In animals, lesion of the ventral tegmental area enhances self-injury behaviour, whereas Figure 5. Anatomical location of cluster headache activation. (A–C) The site electrical stimulation of the same nucleus facilitates of activation in cluster headache according to the reported stereotactic coordinates of Talairach and Tournoux (–2, –18, –8). The red lines point to analgesic processes.46 Therefore, the ventral tegmental area, the site of activation in the three planes of a normalised brain: (A) the functionally linked to the hypothalamus among other location in the coronal brain slice 18 mm posterior to the anterior nuclei, is a midbrain structure involved in the processing commissure (AC; y = –18); (B) the location in the sagittal brain slice 2 mm and modulation of persistent pain information.47 Analgesia to the left of the midline (x = –2); (C) the location in the axial brain slice mm inferior to the AC–posterior commissure (PC) plane (z = –8). (D) An for chronic pain in individuals can be attained by 8axial brain slice corresponding to z = –8. Note that the cluster headache stereotactic electrical stimulation of the periventricular grey activation lies in the midbrain posterior to the hypothalamus (Hyp) and region of the diencephalons.47 Side-effects, such as medial to the red nucleus (RN). MB=mamillary bodies. 34

Neurology Vol 3 November 2004

For personal use. Only reproduce with permission from Elsevier Ltd



Functional neuroimaging of headaches

Conclusion Search strategy and selection criteria References for this review were identified by searches of MEDLINE from Jan 1997 to May 2004, references from relevant articles, and the authors’ own files. The search terms “migraine with aura”, “migraine without aura”, “cluster headache”, “functional imaging”, “PET”, and “fMRI” were used. Abstracts and reports from meetings were included only if they related directly to the previously published work. Only original papers published in English, as well as data from three presentations (one oral presentation and two posters), have been included.

basis of previous evidence, we should expect to find activation in midbrain regions functionally related to the hypothalamus. A second finding that strongly implicates this location as relevant to cluster headache is the result of the voxel-based morphometric analysis. This technique is based on an automated method for the segmentation of grey and white matter. 25 patients with cluster headache compared with 29 control individuals showed a structural difference in approximately the same area observed with PET.49 The importance of this locus in the pathogenesis of cluster headache, and more recently also of SUNCT,35,50 is emphasised by the satisfactory long-term response to deepbrain stimulation.48,51,52 References 1










11 12

13 14

Headache Classification Subcommittee of the International Headache Society. Τhe international classification of headache disorders. Cephalalgia 2004; 24 (suppl 1): 1–152. Milner P. Note on a possible correspondence between the scotomas of migraine and spreading depression of Leao. EEG Clin Neurophysiol 1958; 10: 705. Olesen J, Friberg L, Olsen TS, et al. Timing and topography of cerebral blood flow, aura, and headache during migraine attacks. Ann Neurol 1990; 28: 791–98. Cao Y, Welch KM, Aurora S, Vikingstad EM. Functional MRI-BOLD of visually triggered headache in patients with migraine. Arch Neurol 1999; 56: 548–54. Hadjikhani N, Sanchez del Rio M, Wu O, et al. Mechanisms of migraine aura revealed by functional MRI in human visual cortex. Proc Natl Acad Sci USA 2001; 98: 4687–92. Bowyer SM, Tepley N, Papuashvili N, et al. Analysis of MEG signals of spreading cortical depression with propagation constrained to a rectangular cortical strip. II. Gyrencephalic swine model. Brain Res 1999; 843: 79–86. Bowyer SM, Aurora KS, Moran JE, Tepley N, Welch KM. Magnetoencephalographic fields from patients with spontaneous and induced migraine aura. Ann Neurol 2001; 50: 582–87. Cutrer FM, Sorensen AG, Weisskoff RM, et al. Perfusion-weighted imaging defects during spontaneous migrainous aura. Ann Neurol 1998; 43: 25–31. Sanchez del Rio M, Bakker D, Wu O, et al. Perfusion weighted imaging during migraine: spontaneous visual aura and headache. Cephalalgia 1999; 19: 701–07. Bolay H, Reuter U, Dunn AK, Huang Z, Boas DA, Moskowitz MA. Intrinsic brain activity triggers trigeminal meningeal afferents in a migraine model. Nat Med 2002; 8: 136–42. Pietrobon D, Striessnig J: Neurobiology of migraine. Nat Rev Neurosci 2003; 4: 386–98. Arnold G, Reuter U, Kinze S, Wolf T, Einhaupl KM. Migraine with aura shows gadolinium enhancement which is reversed following prophylactic treatment. Cephalalgia 1998; 18: 644–46. Smith M, Cros D, Sheen V. Hyperperfusion with vasogenic leakage by fMRI in migraine with prolonged aura. Neurology 2002; 58: 1308–10. Iizuka T, Sakai F, Yamakawa K, Suzuki K, Igarashi H,









22 23



Neuroimaging of different headache disorders is beginning to provide important information about the pathogenesis of cephalic pain. We are still far away from discovering the primum movens structure, most probably because headache is a complex disorder in which no single factor or structure plays a key role. The variability in the activation patterns may be the result of this situation. Further studies with larger sample sizes may reduce the inherent variability of functional studies, and thereby decrease confounding factors in the interpretation of the results. In future, the possibility of using high temporal and spatial techniques to image patients minutes before the onset of the headache may provide valuable information of the earlier events that activate the pain system. These findings should reveal a new set of therapeutic targets. Authors’ contributions

MSdR drafted the paper and did the reference search. MSdR and JAL contributed to the final draft. Conflict of interest

We have no conflict of interest. Role of funding source

No funding body was involved in the preparation of this review or the decision to submit it for publication.

Suzuki N. New insight in the mechanism of prolonged migraine aura: augmented vasogenic leakage associated with trigemionovascular activation. Neurology 2004; 62: 152 (abstr P01). Ophoff RA, Terwindt GM, Vergouwe MN, et al. Familial hemiplegic migraine and episodic ataxia type-2 are caused by mutations in the Ca2+ channel gene CACNL1A4. Cell 1996; 87: 543–52. De Fusco M, Marconi R, Silvestri L, et al. Haploinsufficiency of ATP1A2 encoding the Na+/K+ pump alpha2 subunit associated with familial hemiplegic migraine type 2. Nat Genet 2003; 33: 192–96. Masuzaki M, Utsunomiya H, Yasumoto S, Mitsudome A. A case of hemiplegic migraine in childhood: transient unilateral hyperperfusion revealed by perfusion MR imaging and MR angiography. AJNR Am J Neuroradiol 2001; 22: 1795–97. Gutschalk A, Kollmar R, Mohr A, et al. Multimodal functional imaging of prolonged neurological deficits in a patient suffering from familial hemiplegic migraine. Neurosci Lett 2002; 332: 115–18. Lindahl AJ, Allder S, Jefferson D, Allder S, Moody A, Martel A. Prolonged hemiplegic migraine associated with unilateral hyperperfusion on perfusion weighted magnetic resonance imaging. J Neurol Neurosurg Psychiatry 2002; 73: 202–03. Hayashi R, Tachikawa H, Watanabe R, Honda M, Katsumata Y. Familial hemiplegic migraine with irreversible brain damage. Intern Med 1998; 37: 166–68. Gonzalez-Alegre P, Tippin J. Prolonged cortical electrical depression and diffuse vasospasm without ischemia in a case of severe hemiplegic migraine during pregnancy. Headache 2003; 43: 72–75. Butteriss DJ, Ramesh V, Birchall D. Serial MRI in a case of familial hemiplegic migraine. Neuroradiology 2003; 45: 300–03. Chabriat H, Vahedi K, Clark CA, et al. Decreased hemispheric water mobility in hemiplegic migraine related to mutation of CACNA1A gene. Neurology 2000; 54: 510–12. Prodan CI, Holland NR, Lenaerts ME, Parke JT. Magnetic resonance angiogram evidence of vasospasm in familial hemiplegic migraine. J Child Neurol 2002; 17: 470–72. Masuzaki M, Utsunomiya H, Yasumoto S, Mitsudome A. A case of hemiplegic migraine in childhood: transient unilateral hyperperfusion revealed by perfusion MR imaging and MR angiography. AJNR Am J Neuroradiol 2001; 22: 1795–97.

26 Woods RP, Iacoboni M, Mazziotta JC. Brief report: bilateral spreading cerebral hypoperfusion during spontaneous migraine headache. N Engl J Med 1994; 331: 1689–92. 27 Bednarczyk EM, Remler B, Weikart C, Nelson AD, Reed RC. Global cerebral blood flow, blood volume, and oxygen metabolism in patients with migraine headache. Neurology 1998; 50: 1736–40. 28 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. 29 Lauritzen M. Pathophysiology of the migraine aura. The spreading depression theory. Brain 1994; 117: 199–210. 30 Weiller C, May A, Limmroth V, et al. Brain stem activation in spontaneous human migraine attacks. Nat Med 1995; 1: 658–60. 31 Bahra A, Matharu MS, Buchel C, Frackowiak RS, Goadsby PJ. Brainstem activation specific to migraine headache. Lancet 2001; 357: 1016–17. 32 Matharu MS, Bartsch T, Ward N, Frackowiak RS, Weiner R, Goadsby PJ. Central neuromodulation in chronic migraine patients with suboccipital stimulators: a PET study. Brain 2004; 127: 220–30. 33 Kupers R. Facial pain: from animal models to functional neuroimaging studies. Acta Neurol Belg 2001; 101: 32–38. 34 May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. Hypothalamic activation in cluster headache attacks. Lancet 1998; 352: 275–78. 35 May A, Bahra A, Buchel C, Turner R, Goadsby PJ. Functional magnetic resonance imaging in spontaneous attacks of SUNCT: short-lasting neuralgiform headache with conjunctival injection and tearing. Ann Neurol 1999; 46: 791–94. 36 May A, Kaube H, Buchel C, et al. Experimental cranial pain elicited by capsaicin: a PET study. Pain 1998; 74: 61–66. 37 Matharu MS, Good CD, May A, Bahra A, Goadsby PJ. No change in the structure of the brain in migraine: a voxel-based morphometric study. Eur J Neurol 2003; 10: 53–57. 38 Welch KM, Nagesh V, Aurora SK, Gelman N. Periaqueductal gray matter dysfunction in migraine: cause or the burden of illness? Headache 2001; 41: 629–37. 39 Leone M, Bussone G. A review of hormonal findings in cluster headache. Evidence for hypothalamic involvement. Cephalalgia 1993; 13: 309–17. 40 Leone M, Zappacosta BM, Valentini S, Colangelo AM, Bussone G. The insulin tolerance

Neurology Vol 3 November 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Functional neuroimaging of headaches



43 44

test and the ovine corticotrophin-releasing hormone test in episodic cluster headache. Cephalalgia 1991; 11: 269–74. Leone M, Frediani F, D’Amico D, et al. Dexamethasone suppression test, melatonin and TRH-test in cluster headache. Ital J Neurol Sci 1992; 13: 227–32. Sprenger T, Boecker H, Tolle TR, Bussone G, May A, Leone M. Specific hypothalamic activation during a spontaneous cluster headache attack. Neurology 2004; 62: 516–17. Hsieh JC, Hannerz J, Ingvar M. Right-lateralized central processing for pain of nitroglycerin induced cluster headache. Pain 1996; 67: 59–68. Hsieh JC, Stahle-Backdahl M, Hagermark O, Stone-Elander S, Rosenquist G, Ingvar M. Traumatic nociceptive pain activates the hypothalamus and the periaqueductal gray: a positron emission tomography study. Pain 1996;

Neurology Vol 3 November 2004

64: 303–14. 45 Zubieta JK, Smith YR, Bueller JA, et al. Regional mu opioid receptor regulation of sensory and affective dimensions of pain. Science 2001; 293: 311–15. 46 Sotres-Bayon F, Torres-Lopez E, Lopez-Avila A, del Angel R, Pellicer F. Lesion and electrical stimulation of the ventral tegmental area modify persistent nociceptive behavior in the rat. Brain Res 2001; 898: 342–49. 47 Baskin DS, Mehelr WR, Hosobuchi Y, Richardson DDE, Adams JE, Flitter MA. Autopsy analysis of the safety, efficacy and cartography of electrical stimulation of the central gray in humans. Brain Res 1986; 371: 231–36. 48 Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003; 52: 1095–101.

Review 49 May A, Ashburner J, Buchel C, et al. Correlation between structural and functional changes in brain in an idiopathic headache syndrome. Nat Med 1999; 5: 836–38. 50 Leone M, Franzini A, D’Amico D, et al. Hypothalamic deep brain stimulation to relieve intractable chronic SUNCT: the first case. Neurology 2004; 62: S43 (abstr 003). 51 Leone M, Franzini A, Bussone G. Stereotactic stimulation of posterior hypothalamic gray matter in a patient with intractable cluster headache. N Engl J Med 2001; 345: 1428–29. 52 Leone M, Franzini A, Broggi G, Bussone G. Hypothalamic deep brain stimulation for intractable chronic cluster headache: a 3-year follow-up. Neurol Sci 2003; 24 (suppl 2): 143–45.

For personal use. Only reproduce with permission from Elsevier Ltd