Abstract
Local morphological alterations of the brain have recently been detected in cluster headache and chronic tension-type headache, but not in migraine. We investigated 35 patients suffering from migraine and compared them with 31 healthy controls with no headache history. Using magnetic resonance imaging and voxel based morphometry, we found a significant decrease of grey matter in areas ascribable to the transmission of pain (cingulate cortex), but not in areas specific for migraine, such as the brainstem. Our data are in line with recent findings in chronic pain states, such as chronic phantom pain and chronic back pain. We suggest that the grey matter change in migraine patients is the consequence of frequent nociceptive input and should thus be reversible when migraine attacks cease.
Introduction
In the past decade of pain research a network of pain-transmitting areas within the central nervous system has been established, based on both animal studies and findings from functional imaging studies in humans. Consequently, the neurobiology of pain is increasingly understood as an integration of activity in distinct neuronal structures. Evidence of altered local brain chemistry and functional and structural reorganization in chronic back pain patients supports the idea that chronic pain could be understood not only as a state of (dys)functionality, but also as a consequence of central plasticity (1).
Functional imaging of headache has demonstrated that neuronal activation patterns in primary headache syndromes fall into two groups: areas known to be generally involved in pain processing, such as the cingulate, insula cortex and thalamus, and areas activated specifically in primary headache such as the hypothalamus in cluster headache attacks (2) and the brainstem in acute migraine (3, 4). These regions are not thought to be involved in the response to pain impulses, but seem to play a permissive, possibly even causative role, which supports the idea of an underlying neurovascular rather than purely vascular mechanism. Voxel-based morphometric (VBM) analysis of structural T1-weighted magnetic resonance imaging (MRI) scans has been shown in a colocalization of morphometric alterations and functional activation in cluster headache patients (2). A pioneering study by Mathatru et al. found no significant morphometric changes in grey or white matter in patients suffering from episodic migraine (5), whereas patients with chronic tension-type headache showed a decrease of brain grey matter in several pain-transmitting structures (6).
In this study, we investigated patients suffering from migraine without aura and subdivided them into patients with menstrual migraine (mM) and patients whose migraine attacks were not connected to the menstrual cycle. Using VBM, we compared them with healthy controls, including only female volunteers entirely without headache or any other perimenstrual symptoms during menstruation.
Methods
Study population and clinical assessment
Thirty-five patients (32 female, aged 18–49 years, mean 32.4 ± 9.2) and 31 healthy volunteers (31 female, aged 21–56 years, mean 32.3 ± 12.6) participated. The group of migraine patients consisted of 19 patients with mM and 16 with no correlation between headache attacks and menstrual cycle. The diagnosis followed the International Headache Society (IHS) criteria (7). Only three of our patients suffered from chronic migraine following the IHS criteria, and none overused medication. Seven migraine patients were on preventative medication. None of the participants had a history of head injury or any neurological disorders other than migraine.
Healthy controls were chosen carefully. They had either no or extremly few headache days per year (not resembling a migraine). Controls with relatives who suffered regularly from a migraine or other headaches were excluded. All patients and healthy volunteers received a T1-weighted structural image of the brain.
Data acquisition, preprocessing of structural data and statistical analysis
In all patients and controls, a 3-D structural MRI (1.5 T, Siemens Symphony scanner) was acquired on each subject using a T1-weighted rapid gradient echo (MP-RAGE) sequence. Data preprocessing and analysis were performed with SPM2 (Wellcome Department of Cognitive Neurology, London, UK) running under Matlab (Mathworks, Sherborn, MA, USA), described in detail elsewhere (8). Preprocessing involved spatial normalization, grey matter segmentation and smoothing with a Gaussian kernel of 12 mm full width at half maximum. For the preprocessing steps, a previously described optimized protocol (8) was used and a scanner- and study-specific grey matter template. A voxel-by-voxel
Results
Grey matter analysis (P < 0.05, corrected for multiple comparisons)
Patients with migraine showed no significant increase or decrease in grey matter throughout the whole cortex compared with healthy controls. No significant changes were found comparing patients suffering from mM with patients suffering from migraine attacks without obvious hormonal influence, or between the migraine attack compared with the migraine-free interval.
Grey matter analysis (P < 0.05, small volume correction)
Patients with migraine showed a significant decrease in grey matter in the anterior [x = 9, y = 43, z = −2; t = 4.2 (d.f. 64) and x = −5, y = 42, z = 18; t = 4.1 (d.f. 64); P < 0.05 small volume correction (SVC)] and posterior [x = −1, y = 0, z = 47; t = 3.88 (d.f. 64); P < 0.05 SVC] part of the cingulate cortex and the right insular cortex [x = 43, y = 15, z = 4; t = 3.3 (d.f. 64); P < 0.001 SVC] compared with healthy controls.
White matter analysis
No significant changes were found between either migraine without aura or mM patients and controls.
Discussion
Using VBM in migraine patients, we found significant changes in grey matter in areas ascribable to the transmission of pain (cingulate cortex), but not in areas specific for migraine, such as the brainstem.
Our data are in line with a recent study by Rocca et al., who reported reduced grey matter density in the anterior cingulate cortex, anterior insulae and temporal lobes (11). They also described increased density of the periaqueductal grey and of the dorsolateral pons in migraine patients (11), a finding we could not reproduce. One possibility for this discrepancy is the fact that we (the Matharu study as well as this study) used a 1.5-T scanner, whereas the Italian group used a magnet with a higher field strength (3 T), which may be an advantage in detecting exceptionally subtle differences in cohort studies. However, the migraine cohort of the Italian group was small (n = 16) and comprised only patients with T2-visible brain lesions, a finding which is not part of the IHS criteria is rather unusual in migraine patients. As migraine has a strong genetic component, the ideal inclusion criteria for future studies to render groups as homogeneous as possible could be based on genotype (cohort study) or response to treatment (longitudinal study including controls).
Although our main finding (decrease of grey matter in pain-transmitting structures) is very similar to that of Rocca et al., our interpretation of the results is different. The authors of the former study have suggested that grey matter changes may be the consequence of repeated brain insults or damage during migraine attacks (11). According to their interpretation, the topographical distributions of grey matter changes would be the consequence of cortical regions having varying susceptibilities. This mechanism was described as facilitated by a ‘retrograde degeneration of axons passing through T2-visible lesions of the white matter’, which was an inclusion criterion in the above-mentioned study. However, a striking feature of both studies is the fact that the grey matter changes were not randomly distributed, but concerned defined and functionally highly specified brain areas, namely involvement in supraspinal nociceptive processing.
It is remarkable that the alterations (i.e. decrease in grey matter) seen in the anterior cingulate cortex in migraine patients (Fig. 1) are similar to a decrease in this region in chronic back pain (1), chronic phantom pain (12) and chronic tension-type headache (6). From these data, it is possible to argue that the alteration of this region is a consequence, rather than a cause, of frequent nociceptive input. As all of our patients had suffered from migraine attacks since youth or childhood, it was not possible to search for correlations between changes in grey matter volume and pain duration. In this regard, the migraineurs that we studied were ‘chronic pain patients’, although the IHS defines chronic migraine as migraine headache occurring on ≥15 days per month, irrespective of the overall duration of years that the attacks occur.
Statistical parametric maps demonstrating the structural difference in grey matter between migraine patients and unaffected control subjects. Significant grey matter decrease (P < 0.05, small volume correction) is shown superimposed in white on a normalized image of a healthy control subject. The left side of the picture is the left side of the brain.
It has been suggested that the central reorganization processes in chronic pain syndromes could involve a ‘degeneration’ of antinociceptive brain areas (i.e. the anterior cingulate cortex), which might, among other mechanisms, contribute to a chronification process (6, 12). It is not known why migraine usually remits with age. It is a very interesting question for future studies whether the morphological changes reverse when migraine, and hence the disproportionate amount of nociceptive signals, stops. In this context, it is important to mention that the degree of functional reorganization in chronic pain syndromes could be reduced by a sensory discrimination training programme (13).
Acknowledgements
The authors thank all patients and volunteers for their participation in this study, Gerhard Schuierer and Bogdan Draganski for technical support and E. Schoell for reading the manuscript. A.M. is supported by a grant of the Deutsche Forschungsgemeinschaft (MA 1862/2). This study issupported by a grant of the Federal Ministry of Education and Research (BMBF project no. 371 57 01).