Sage Journals: Discover world-class research

Abstract

Background

Specific white-matter tract alterations in migraine remain to be elucidated. Using diffusion tensor imaging (DTI), this study investigated whether the integrity of white-matter tracts that underlie regions of the “pain matrix” is altered in migraine and interrogated whether the number of years lived with migraine modifies fibertract structure.

Methods

Global probabilistic tractography was used to assess the anterior thalamic radiations, the corticospinal tracts and the inferior longitudinal fasciculi in 23 adults with migraine and 18 healthy controls.

Results

Migraine patients show greater mean diffusivity (MD) in the left and right anterior thalamic radiations, the left corticospinal tract, and the right inferior longitudinal fasciculus tract. Migraine patients also show greater radial diffusivity (RD) in the left anterior thalamic radiations, the left corticospinal tract as well as the left and right inferior longitudinal fasciculus tracts. No group fractional anisotropy (FA) differences were identified for any tracts. Migraineurs showed a positive correlation between years lived with migraine and MD in the right anterior thalamic radiations (r = 0.517; p = 0.012) and the left corticospinal tract (r = 0.468; p = 0.024).

Conclusion

Results indicate that white-matter integrity is altered in migraine and that longer migraine history is positively correlated with greater alterations in tract integrity.

Keywords

Migraine pain neuroimaging white matter fibertracts diffusion-tensor imaging

Introduction

Recent advances in neuroimaging techniques have allowed for a better understanding of brain-related changes in migraine. As such, several structural and functional imaging studies have indicated loss of brain integrity in a variety of multifocal cortical (1 –3) and subcortical pain-processing regions (3 –6) that are often referred to as being part of the “pain matrix.” It has been hypothesized that recurrent headache pain alters the brain “pain matrix,” and several studies indeed suggest that there is a correlation between migraine-specific disease markers (years lived with migraine and headache frequency) and abnormal brain structure and function (7,8). As the “pain matrix” consists of an interspersed network of cortical and subcortical brain regions, it is likely that the white-matter tracts that underlie and inter-connect regions of the “pain matrix” are altered in patients with migraine. Diffusion-tensor imaging (DTI) provides an excellent technique for investigating the microscopic characteristics of white-matter pathways by assessing the diffusion of water along and across white-matter tracts. Several migraine studies to date have investigated diffusion-related brain alterations using DTI (5,7,9 –17) and have found regional differences in diffusivity metrics of migraine patients relative to controls. Yet, tract-specific differences are still poorly understood and insufficiently characterized. To date, DTI studies have shown people with migraine to have altered white-matter integrity in midline white-matter tracts including the corpus callosum and the corona radiata, as well as in subportions of white-matter tracts such as the internal capsule. However, studies that have investigated structural integrity over the entire path length are sparse. (For a summary of migraine DTI findings, see Table 1.) In addition, most published migraine DTI studies have either used a region-of-interest approach to look at specific subregions of white-matter tracts or have used tract-based-spatial-statistics (TBSS), which uses a skeletonized fibertract template for estimating and modeling fibertract location and tract-specific diffusion characteristics. In this study, we use a novel fibertract tool that enables modeling of fibertract anatomy within each person’s own diffusion space by using each person’s own structural anatomy as estimation “priors.” The advantage of this methodology is that it does not require a template approach for image reconstruction, thereby increasing the accuracy of precisely locating and estimating tract diffusion characteristics.

Table 1.

Migraine studies using DTI.

Author	Study cohort	Data analysis	Results
Coppola et al. 2014 (9)	24 MWoA 10 ictal 14 interictal 15 HCs	ROI	Interictal MWoA vs. HCs: ↗FA and ↘MD in bilateral thalami. Ictal MWoA vs. HCs: no difference in FA and MD All MWoA: positive correlation between FA in right thalamus and number of days since last migraine attack
Kara et al. (10) 2013	14 MWoA 15 HCs	ROI	MWoA during ictal phase vs. HCs: ↗MD in red nuclei. No correlation between diffusion metrics and migraine disease parameters
Liu et al. (13) 2013	36 MWoA (newly diagnosed migraineurs)	VBM/TBSS longitudinal study time-point I: 8–14 weeks post-diagnosis time-point II: One-year follow-up	Episodic MWoA: No changes in white matter FA, MD, AD, and RD over one year.
Yu et al. (12) 2013	40 MWoA 40 HCs	TBSS	MWoA vs. HCs: ↘ AD in several tracts. MWoA with depression vs. MWoA without depression: ↘ FA, ↗MD and ↗RD in corpus callosum, superior longitudinal fasciculus and anterior corona radiata. MWoA: Regional tract integrity was correlated with symptoms of depression.
Yu et al. (35) 2013	20 MWoA 20 HCs	TBSS	MWoA vs. HCs: ↘ FA, MD, AD in corpus callosum and internal capsule. Negative correlation between regional diffusivity (MD, AD) and headache frequency and disease duration
Yuan et al. 2012 (16)	21 MWoA 21 HCs	TBSS	MWoA vs. HC: ↘ FA in corpus callosum
Rocca et al. 2008 (17)	7 MWA 8 MWoA 11 HCs	ROI	MWA vs. HCs: ↘ FA and ↗MD in optic radiations No correlation between diffusion metrics and clinical parameters
Schmitz et al. 2008 (7)	8 MWA 20 MWoA 28 HCs	VBM	All migraineurs vs. HCs: ↘ FA in superior frontal lobes and medial frontal lobe, brainstem and cerebellum. No differences in MD
Granziera et al. 2006 (11)	12 MWA 12 MWoA 15 HCs	ROI	All migraineurs vs. HCs: ↘FA in superior colliculus and lateral geniculate nucleus
Rocca et al. 2003 (15)	6 MWA 28 MWoA 17 HCs	Whole-brain MD and FA histogram-derived metrics	All migraineurs vs. HC: ↘MD peak height of normal-appearing brain tissue. No differences in FA.

DTI: diffusion tensor imaging; AD: axial diffusivity; FA: fractional anisotropy; HCs: healthy controls; MD: mean diffusivity; MWA: migraine with aura; MWoA: migraine without aura; RD: radial diffusivity; ROI: region of interest; TBSS: tract-based spatial statistics; VBM: voxel-based-morphometry; ↘: lower; ↗: higher.

The aim of this study was to investigate several major fibertracts with high potential to be involved in the migraine disease process. Three major fibertracts were interrogated, each known to connect regions of the “pain matrix.” Using DTI, this study interrogated the integrity of bilateral fibertracts including the anterior thalamic radiations, the corticospinal tracts, and the inferior longitudinal fasciculi. DTI-derived metrics using a novel tractography reconstruction tool, TRActs Constrained by Underlying Anatomy (TRACULA), was used to assess fibertract integrity by measuring fractional anisotropy (FA), radial diffusivity (RD) and mean diffusivity (MD) in migraine patients and healthy controls and to investigate whether fibertract-specific diffusion metrics are associated with number of years with migraine.

Methods

Eighteen healthy controls (12 female, six male) and 23 migraineurs (15 female, eight male) were included in this study after providing informed consent in accordance with the Mayo Clinic Institutional Review Board guidelines. Migraine diagnoses were made using the International Classification of Headache Disorders II (ICHD-II) diagnostic criteria (18). Nineteen episodic and four chronic migraine patients were included. None of the migraineurs were taking migraine prophylactic medications and none were overusing migraine-abortive medications in accordance with ICHD-II specifications. Imaging was conducted when migraine patients and healthy controls were pain free. All migraine patients were free of migraines during the time of imaging and had been so for a minimum of two days prior to their imaging appointment. All patients had migraines for a minimum of three years.

Prior to undergoing consent, all participants were screened for history of brain trauma, psychiatric disorders and chronic or acute pain disorders (other than migraine). If screening identified any of these conditions, individuals were excluded from study participation. Initially, this study included 19 healthy controls and 24 migraine patients but two individuals were excluded from this study because of excessive movement in the scanner (one migraineur) and abnormal imaging findings (one control individual) thus leaving 18 healthy controls and 23 migraine patients in the final analysis. Feelings of depression, general anxiety and situational anxiety were recorded for all participants using the Beck Depression Inventory (BDI-II) (19) and the State-Trait Anxiety Inventory (STAI), Form Y1 and Form Y-2 (20).

Group differences for demographic data were estimated using independent Student’s t-tests or Fisher’s exact tests, wherever appropriate. Group differences between fibertract metrics were estimated using independent Student’s t-tests. Because of the sensitivity of DTI to head motion, between-group differences in translational (mm) and rotational (degrees) movement were investigated using independent t-tests. Pearson correlations were used to assess correlations between age and tract metrics in healthy controls in order to assess whether there is a relationship between age and fibertract integrity. Pearson correlations were used to assess correlations between tract metrics and number of years with migraine. Statistical significance levels were set at p < 0.05 for estimating between-group differences of demographic data and fibertract data.

Image acquisition

Imaging was conducted on a single Skyra Siemens (Erlangen, Germany) 3-Tesla whole-body magnetic resonance imaging (MRI) scanner using a 20-channel head/neck coil. All imaging was conducted during a time period of 14 months, and all participants were scanned in one single session. No scanner or imaging sequence updates were performed during this time period. Imaging parameters included the following series: 1) three-dimensional (3D) T1-weighted sagittal magnetization-prepared-rapid acquisition with gradient echo (MP-RAGE) (echo time (TE) = 3.03 ms, repetition time (TR) = 2400 ms, flip angle = 8 degrees, 128 slices, slice thickness = 1.25 mm, 1 × 1 × 1.3 mm³ voxels, 256 mm² field of view (FOV), matrix size = 256 × 256). 2) Axial T2-weighted imaging (TE = 84 ms, TR = 6800 ms, flip angle = 150 degrees, 38 slices with 1 × 1 × 4 mm³ voxels, slice thickness = 4 mm, 256 mm² FOV, matrix size = 256 × 256). 3) 30-non-linear direction (b = 1000 s/mm²) axial DTI (TE = 71 ms, TR = 5100 ms, 38 slices, slice thickness = 4 mm, 1.7 × 1.7 × 4 mm³ voxels, 220 mm² FOV, matrix size = 128 × 128). One image was acquired without diffusion weighting. T2-weighted imaging and DTI images were acquired with two averages. All T1- and T2-imaging scans were reviewed by a board-certified neuroradiologist and data were excluded from analyses if there were abnormalities (including T2 hyperintensities). Individuals were excluded for analyses if motion exceeded more than two degrees rotational and/or two mm translational movement at the time of scanning.

Data post-processing methods

DTI data were post-processed using the recently developed automated global tractography toolbox TRACULA which is freely downloadable online (http://surfer.nmr.mgh.harvard.edu/) (21).. In order to prevent post-processing irregularities stemming from using multiple workstations (22), all imaging post-processing was conducted on a single Mac workstation (2 × 2.4 GHz 6-core Intel Xeon processor) running OS X Lion 10.75. A priori selected white-matter tracts were automatically reconstructed using global probabilistic tractography using the T1 anatomical priors of each individual. First, using each participant’s own high-resolution T1-weighted images, the structural brain anatomy was reconstructed using the automated brain segmentation and parcellation package in FreeSurfer, which has been previously described in detail (23). Briefly, data processing steps of this technique include: removal of non-brain tissue (skull stripping) (24), Talairach transformation (25), segmentation and parcellation of gray and white matter (26), intensity normalization and brain boundary tesselation (26) and surface deformation (23). Using the structural brain anatomy of each person in combination with a pre-labeled training set, the distribution of fibertracts was estimated using a Markov Chain Monte Carlo algorithm (27). The accuracy of the brain segmentation output was manually inspected by a trained technician before including patient data for group analyses. This ensures the accuracy of the automated brain reconstruction process and prevents the inclusion of erroneous datasets. The robustness of the TRACULA package was validated in prior papers (21,28). TRACULA is able to provide results comparable to voxel-based analyses techniques using tract-based spatial statistics (28) and has been shown to accurately and reliably identify white-matter tract damage in patient populations (21,28). In addition, the advantage of this global tractography method is that it enables the calculation of average weighted diffusivity measures for each tract, which allows for correlating fibertract integrity over the entire path length with disease-specific parameters (29).

Fibertracts metrics were extracted from within TRACULA and imported to SPSS version 22.0 (Armonk, NY; IBM Corp.) for statistical analysis. Group data were compared using two-tailed t-tests. A post-hoc correlation analysis (Pearson’s correlation) was conducted for tracts that showed significant group differences between migraine patients and healthy controls in order to investigate the relationship between DTI indices of fibertract integrity and number of years with migraine.

Results

There were no significant group differences as to age (migraineurs = 38.5, SD = 11.0, age range = 22–64 years; normal controls = 37.7, SD = 11.0, p = 0.83, age range: 23–60 years), state anxiety (migraineurs = 28.1, SD = 7.5; normal controls = 24.9, SD = 4.8; p = 0.12) or on measures of trait anxiety (migraineurs = 31.3, SD = 6.2, healthy controls = 29.5, SD = 5.1; p = 0.34) (see Table 2). Thirteen migraineurs (56.5%) experienced visual aura while the rest were without. For the migraine group, mean disease duration (number of years with migraine) was 18 years, with migraineurs suffering on average eight headaches per month. None of the study participants reported symptoms suggestive of depression as measured by the BDI-II (18). BDI-II scores were below 10 for all participants, which is within the normal range.

Table 2.

Participant characteristics of migraineurs and healthy controls.

	Migraineurs n = 23	Healthy controls n = 18	p value
Age, mean (SD)	38.5 (11.0)	37.7 (11.0)	.83
Sex (m/f)	8 /15	6/12	.92
State anxiety, mean (SD)	28.1 (7.5)	24.9 (4.8)	.12
Trait anxiety, mean (SD)	31.3 (6.2)	29.5 (5.1)	.34
Years with migraine, mean (SD)	18.0 (11.0)	n/a	n/a
Headache frequency, mean (SD)	7.9 (5.1)	n/a	n/a

SD: standard deviation; m: male; f: female; Headache frequency: number of headache days per month.

Motion parameters

Because of the sensitivity of DTI to movement, and the possibility of between-group differences in head movement (27), motion parameters were calculated for all participants. Parameters measured to calculate motion included average translation (mm) and rotation (degrees). Average translation during diffusion MRI was 0.77 mm (SD = 0.33) for migraineurs and 0.66 mm (SD = 0.19) for healthy controls. Average rotation in the scanner was 0.004 degrees (SD = 0.003) for migraineurs and 0.003 degrees (SD = 0.001) for healthy controls. There were no significant group differences for average translation (p = 0.19) or average rotation (p = 0.08). None of the participants included in this study exceeded 2 degrees rotational, and/or 2 mm translational movement and controls as well as migraineurs were well below the commonly recognized “cut-off” points for motion (27,30).

Group differences in MD, RD and FA for the anterior thalamic radiations, corticospinal tracts, and inferior longitudinal fasciculus tracts are shown in Table 3.

Table 3.

Diffusion characteristics for 3 major tracts in migraineurs and healthy controls.

	Migraineurs	Healthy controls	p value
Anterior thalamic radiations
L MD^a	0.762 (0.022)	0.745 (0.021)	0.014
R MD^a,b	0.773 (0.022)	0.760 (0.016)	0.029
L RD^a	0.612 (0.031)	0.590 (0.025)	0.021
R RD	0.618 (0.035)	0.560 (0.027)	0.069
L FA	0.347 (0.034)	0.362(0.032)	0.172
R FA	0.351 (0.042)	0.367 (0.033)	0.178
Corticospinal tract
L MD^a,b	0.738 (0.027)	0.720 (0.026)	0.029
R MD	0.743 (0.030)	0.730 (0.030)	0.186
L RD^a	0.580 (0.025)	0.560 (0.033)	0.017
R RD	0.583 (0.036)	0.569 (0.028)	0.174
L FA	0.394 (0.036)	0.406(0.045)	0.109
R FA	0.385 (0.042)	0.385 (0.028)	0.510
Inferior longitudinal fasciculus tract
L MD	0.774 (0.022)	0.761 (0.032)	0.138
R MD^a	0.780 (0.034)	0.754 (0.025)	0.011
L RD^a	0.577 (0.034)	0.553 (0.037)	0.043
R RD^a	0.566 (0.038)	0.541 (0.026)	0.024
L FA	0.436 (0.043)	0.457 (0.048)	0.152
R FA	0.454 (0.052)	0.467 (0.035)	0.379

AD: axial diffusivity; FA: fractional anisotropy; RD: radial diffusivity; L: left; R: right; m: male; f: female; SD: standard deviation. ^aStatistical significance at p < 0.05. ^bPositive correlation with number of years with migraine (p < 0.05). Mean diffusivity and radial diffusivity metrics are shown as 10⁻³ and FA is shown as the true value.

The migraineurs showed significantly higher bilateral MD (left MD p = 0.014; right MD p = 0.029) and higher left RD (p = 0.021) relative to healthy controls in the anterior thalamic radiations. Figure 1 shows a visual demonstration of the MD changes along the length of the bilateral anterior thalamic radiation fiber pathways (averaged over increments of 5 mm). No significant group differences in right RD or bilateral FA were found for the anterior thalamic radiations.

Figure 1.

The right and left anterior thalamic radiations (ATR) modeled on the anatomy of a single individual. Graphs show mean diffusivity (MD) for the right and left ATR in controls (blue) and migraine patients (red). Numbers on the X-axes indicate 5 mm segments along the tract (anterior-posterior direction) starting from the beginning of the tract in the frontal cortex.

Significant group differences in the corticospinal tract were identified for left MD (p = 0.029) and left RD (p = 0.017). No group differences were identified for right MD, right RD and bilateral FA in the corticospinal tract.

Significant group differences were seen for several diffusion metrics in the inferior longitudinal fasciculus tracts. Migraine patients showed significantly higher bilateral RD (right RD, p = 0.024; left RD p = 0.043) and right MD (p = 0.011) relative to healthy controls. There were no significant group differences in left MD and bilateral FA in the inferior longitudinal fasciculus tracts.

We researched whether the findings of altered white-matter tract integrity in the anterior thalamic radiations, corticospinal tracts and the inferior longitudinal fasciculus tracts were associated with number of years suffered with migraine. Because of the possibility that the number of years lived with migraine could be strongly associated with age, we first looked into whether an association existed between DTI metrics and age in healthy controls. As age is known to negatively affect structural integrity, we investigated the association between tract integrity and age in healthy controls.

There was no significant correlation between age and MD of the right anterior thalamic radiations (r = 0.350; p = 0.155). There was no correlation between left anterior thalamic radiations MD (r = 0.009; p = 0.972) and RD (r = 0.224; p = 0.371) with age. There was no significant correlation between age and left corticospinal tract MD (r = –0.093; p = 0.714) and between age and left corticospinal tract RD (r = −0.148; p = 0.558). There was no correlation between age and bilateral inferior longitudinal fasciculus RD (right: r = −0.066; p = 0.796; left: r = −0.126; p = 0.618) as well as between age and left inferior longitudinal fasciculus MD (r = −0.065; p = 0.797). Results indicated no correlations between DTI indices and age in healthy controls.

There was a significant positive correlation between number of years with migraine and MD of the right anterior thalamic radiations (r = 0.517; p = 0.012). There were no correlations between number of years with migraine and left MD (r = 0.222; p = 0.309) and left RD (r = 0.128; p = 0.562) in the anterior thalamic radiations. There was a positive correlation between number of years with migraine and left corticospinal tract MD (r = 0.468; p = 0.024) but a nonsignificant relationship between number of years with migraine and left corticospinal tract RD (r = 0.367; p = 0.085) (Figure 2). There was no correlation between number of years with migraine and bilateral inferior longitudinal fasciculus RD (right: r = 0.281; p = 0.194; left: r = 0.331; p = 0.149) and left inferior longitudinal fasciculus MD (r = 0.217; p = 0.321).

Figure 2.

For illustration purposes only, the average location of the anterior thalamic radiations (ATR), corticospinal tract (CST) and inferior longitudinal fasciculi (ILF) are shown on a template brain. Graphs indicate correlations between years with migraine and mean diffusivity (MD) in the right ATR and the left CST (p < 0.05).

An exploratory subgroup analysis between migraineurs with aura (n = 13) and migraineurs without aura (n = 10) showed no significant differences in MD, RD and FA for the anterior thalamic radiations, the corticospinal tracts and the inferior longitudinal fasciculus tracts.

Discussion

The main findings of this study are that migraine patients have tract-specific alterations in several fibertracts including the bilateral anterior thalamic radiations, the left corticospinal tract, and the inferior longitudinal fasciculus tracts. As shown in Figure 2, migraine patients demonstrate a positive correlation between years lived with migraine and increased MD in the right anterior thalamic radiations and the left corticospinal tract, perhaps indicating that the integrity of these fibertracts is mediated by migraine burden.

Although several studies to date have researched white-matter alterations in migraineurs compared to healthy controls (see Table 1), there is a lack of published studies investigating whole-tract alterations in migraine patients. The approach used in this study is based on global probabilistic tractography, which uses each individual’s own structural anatomy as estimation priors combined with a pre-labeled training set for the automatic reconstruction of several major fibertracts. This technique does not depend on the manual drawing of regions of interest, nor does it rely on specific manual thresholding for fibertract angles or length. As such, the automated path reconstruction prevents user-intervention errors and yields an automated procedure of assessing white-matter tract integrity that is suitable for evaluating large datasets.

Specifically, results indicate increased MD (bilateral anterior thalamic radiations, left corticospinal tract, right inferior longitudinal fasciculus) and RD (left anterior thalamic radiations, left corticospinal tract and bilateral inferior longitudinal fasciculus tracts) in migraine patients relative to healthy controls. No group differences in FA were observed for any of the tracts. Our findings of increased MD and RD are similar to two other migraine studies by Kara et al. (10) and Rocca et al. (17), which have also shown increased MD albeit for regions not examined in this study (red nucleus and optic radiations). Similarly, increased MD and RD have been reported in a study of migraineurs with depression for regions including the corpus callosum, superior longitudinal fasciculus and the corona radiate (12). In addition, patients with cluster headaches have increased MD and RD relative to healthy controls in regions including the frontal, parietal and occipital areas (14). Our findings that patients with migraine lack differences in FA are similar to several other DTI studies (9,10,13) but dissimilar to other migraine DTI studies in which decreased FA was identified in regions including the corpus callosum (12,14), optic radiations (17) as well as decreased FA for the frontal lobes, brain stem and cerebellum (7) and findings of increased FA in the bilateral thalami (9). It is of note that these regions largely differed from the fibertracts interrogated in the current analysis, making comparisons between studies difficult. (See Table 1 for a list of DTI findings in migraine). However the relatively cohesive findings of increased MD and RD in a variety of brain regions, compared to the diverse findings of increased, decreased and unaltered FA in migraine, could potentially indicate that MD and RD are more stable and sensitive measures relative to FA for identifying tract-integrity changes in migraine.

The interpretation of FA, MD and RD is profoundly complex and the specific biological underpinnings of individual diffusion metrics are still poorly understood. Although elevated MD has been indicated in prior studies as a potential indicator for edema, increased RD as suggestive of axonal loss or demyelination, and changes in FA that describe the degree of directional diffusion, as reflective of tissue health and organization (31), it remains to be investigated in what way different diffusion metrics relate to one another to define a disease. Therefore, the precise mechanism by which FA, RD and MD contribute to characterize migraine remains to be investigated.

The anterior thalamic radiations are bilateral fibertracts shown to modulate pain in healthy controls (32) and connect the prefrontal cortex with the thalamic region, which has a central role in the pathophysiology of the migraine disease process (33). For illustration purposes, Figure 1 shows the mean diffusivity differences over averaged 5 mm segments along the anterior thalamic radiations. These segment-by-segment differences indicate that most of the anterior thalamic radiations-tract differences between migraine patients and healthy controls arise from segments closer to the thalamic region, especially for the right anterior thalamic radiations. Interestingly, previous studies have indicated that the thalamus undergoes microstructural changes in migraine and show that the thalamus plays a key role in migraine pain processing (11,34). Yet, the precise way by which diffusion changes along the anterior thalamic radiations relate to thalamic microstructural alterations or contribute to a potential migraine tract profile (28) will need to be assessed in future studies. Our results of increased MD and RD of the anterior thalamic radiations are in line with findings by Yu et al., who have shown abnormal integrity in segments of the anterior thalamic radiations (anterior limb of the internal capsule) in migraine patients without aura (35). In healthy individuals, reduced FA in the anterior thalamic radiations was shown for individuals who have difficulty diverting attention away from a painful stimulus compared to people who are successfully able to redirect their attention in order to increase task performance (32). The current study did not show FA reductions but increased MD and RD, which indicate tract damage, albeit a potentially different cellular mechanism. Loss of anterior thalamic radiation tract integrity in migraineurs could be associated with migraineurs’ greater tendency to ruminate on pain symptomatology (36).

The corticospinal tract is a descending fibertract important for nociceptive perception (37). It originates in the primary (M1) and secondary motor cortex and somatosensory cortex and extends to the spinal cord (38). Resting-state functional imaging findings by Mainero and colleagues (4) demonstrate that migraine patients show altered functional connectivity between the periaqueductal gray networks and the precentral (motor) region and show hyper-responsitivity of the motor cortex and the underlying corticospinal tract using transcranial magnetic stimulation (39). Additionally, similar to our findings of altered diffusivity in the corticospinal tract, Rayan et al. have shown that patients suffering from chronic pain have increased left axial diffusivity within portions of the corticospinal tract (40) thus indicating compromised tract integrity. Future studies are needed to clarify the function of the corticospinal tract in migraine and pain disorders. The tract-specific changes for the left relative to the right corticospinal tract in migraine patients and in patients with chronic pain is an intriguing finding. It is currently not known whether the right and left corticospinal tracts have different roles in migraine and in pain disorders.

The inferior longitudinal fasciculus connects the anterior temporal region (temporal pole) to the occipital cortex by traversing along the inferior temporal cortex. The inferior longitudinal fasciculus connects regions important for the interpretation and perception of pain. Similar to findings described herein, Rayan et al. found increased axial diffusivity within portions of the inferior longitudinal fasciculus that correlated with pain symptomatology in patients with chronic pain (40), and Buckalev et al. found increased white matter hyperintensity in the anterior thalamic radiations and the inferior longitudinal fasciculus in patients with disabling chronic back pain (41). Indeed, the temporal pole is emerging as a pivotal region in the migraine pathology. Moulton et al. (42) showed that migraine patients have stronger temporal pole functional connectivity as well as hyperexcitability in the temporal pole during painful heat stimulation and results by Copolla et al. (43) indicated gray-matter loss in the left temporal pole in migraineurs between attacks. As such, these results could indicate that the temporal pole region is functionally and structurally altered in migraine and perhaps in other pain disorders.

Although our results indicate lack of integrity for several white-matter tracts, a unimodal approach based on DTI findings alone cannot explain the mechanistic cause for these alterations in migraine. However, our findings of correlations between tract integrity and years with migraine suggests that migraine chronicity affects white-matter health and that fibertract alterations assessed via DTI metrics could potentially represent a valuable disease-state marker. Multimodal study designs are needed that will elucidate how fibertract integrity parameters relate to brain structure (volume or cortical thickness) and functional imaging (task-related or resting-state) findings. Such studies that are able to combine brain structural and functional techniques will provide further insights into the migraine pathophysiology.

Summary

Migraine is associated with cortical and subcortical alterations of brain structure and function of areas associated with pain processing. Our study findings show abnormal fibertract integrity in the bilateral anterior thalamic radiations, corticospinal tracts and inferior longitudinal fasciculus tracts in migraineurs. Our results also indicate that years lived with migraine is positively associated with increased MD in the left corticospinal tract and the right anterior thalamic radiations, suggesting that a greater number of years with migraine negatively affects white matter fibertract health.

Clinical implications

Fibertract integrity for the bilateral anterior thalamic radiations, the left corticospinal tract, and the bilateral inferior longitudinal fasciculus tracts is compromised in migraine patients.

Years with migraine is positively correlated with loss of integrity in the right anterior thalamic radiations and the left corticospinal tract.

Footnotes

Funding

This study was supported by a grant from the National Institutes of Health (K23NS070891) to TJS.

Conflict of interest

None declared.

Acknowledgments

We would like to thank Judi Wieser, R.N., for her efforts in coordinating this study, and we express our gratitude to all participants for their commitment and time dedication to this study.

We state that both authors contributed significantly to the study design as well as the data acquisition and interpretation. Both authors were involved in writing and revising the manuscript. The published version of this manuscript was approved by both authors.

References

Chong

Dodick

Schlaggar

. Atypical age-related cortical thinning in episodic migraine. Cephalalgia 2014; 34: 1115–1124.

Schwedt

Chong

. Correlations between brain cortical thickness and cutaneous pain thresholds are atypical in adults with migraine. PloS One 2014; 9: e99791–e99791.

Maleki

Linnman

Brawn

. Her versus his migraine: Multiple sex differences in brain function and structure. Brain 2012; 135: 2546–2559.

Mainero

Boshyan

Hadjikhani

. Altered functional magnetic resonance imaging resting-state connectivity in periaqueductal gray networks in migraine. Ann Neurol 2011; 70: 838–845.

Rocca

Ceccarelli

Falini

. Brain gray matter changes in migraine patients with T2-visible lesions: A 3-T MRI study. Stroke 2006; 37: 1765–1770.

Zhao

Liu

Yan

. Abnormal brain activity changes in patients with migraine: A short-term longitudinal study. J Clin Neurol 2014; 10: 229–235.

Schmitz

Admiraal-Behloul

Arkink

. Attack frequency and disease duration as indicators for brain damage in migraine. Headache 2008; 48: 1044–1055.

Valfre

Rainero

Bergui

. Voxel-based morphometry reveals gray matter abnormalities in migraine. Headache 2008; 48: 109–117.

Coppola

Tinelli

Lepre

. Dynamic changes in thalamic microstructure of migraine without aura patients: A diffusion tensor magnetic resonance imaging study. Eur J Neurol 2014; 21: 287–e13.

10.

Kara

Kiyat Atamer

Onat

. DTI findings during spontaneous migraine attacks. Clin Neuroradiol 2013; 23: 31–36.

11.

Granziera

DaSilva

Snyder

. Anatomical alterations of the visual motion processing network in migraine with and without aura. PLoS Med 2006; 3: e402–e402.

12.

Yuan

Zhao

. White matter integrity affected by depressive symptoms in migraine without aura: A tract-based spatial statistics study. NMR Biomed 2013; 26: 1103–1112.

13.

Liu

Lan

. Migraine-related gray matter and white matter changes at a 1-year follow-up evaluation. J Pain 2013; 14: 1703–1708.

14.

Szabό

Kincses

Párdutz

. White matter disintegration in cluster headache. J Headache Pain 2013; 14: 64–64.

15.

Rocca

Colombo

Inglese

. A diffusion tensor magnetic resonance imaging study of brain tissue from patients with migraine. J Neurol Neurosurg Psychiatry 2003; 74: 501–503.

16.

Yuan

Qin

Liu

. Reduced fractional anisotropy of corpus callosum modulates inter-hemispheric resting state functional connectivity in migraine patients without aura. PloS One 2012; 7: e45476–e45476.

17.

Rocca

Pagani

Colombo

. Selective diffusion changes of the visual pathways in patients with migraine: A 3-T tractography study. Cephalalgia 2008; 28: 1061–1068.

18.

The International Classification of Headache Disorders: 2nd edition. Cephalalgia 2004; 24 (Suppl 1): 9–160.

19.

Beck

Steer

Ball

. Comparison of Beck Depression Inventories -IA and -II in psychiatric outpatients. J Pers Assess 1996; 67: 588–597.

20.

Spielberger

Gorsuch

Lushene

. Manual for the State-Trait Anxiety Inventory, Palo Alto, CA: Consulting Psychologists Press, 1983.

21.

Yendiki

Panneck

Srinivasan

. Automated probabilistic reconstruction of white-matter pathways in health and disease using an atlas of the underlying anatomy. Front Neuroinform 2011; 5: 23–23.

22.

Gronenschild

Habets

Jacobs

. The effects of FreeSurfer version, workstation type, and Macintosh operating system version on anatomical volume and cortical thickness measurements. PloS One 2012; 7: e38234–e38234.

23.

Dale

Fischl

Sereno

. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 1999; 9: 179–194.

24.

Segonne

Dale

Busa

. A hybrid approach to the skull stripping problem in MRI. Neuroimage 2004; 22: 1060–1075.

25.

Fischl

Salat

Busa

. Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron 2002; 33: 341–355.

26.

Fischl

Liu

Dale

. Automated manifold surgery: Constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE Trans Med Imaging 2001; 20: 70–80.

27.

Yendiki

Koldewyn

Kakunoori

. Spurious group differences due to head motion in a diffusion MRI study. Neuroimage 2013; 88C: 79–90.

28.

Sarica

Cerasa

Vasta

. Tractography in amyotrophic lateral sclerosis using a novel probabilistic tool: A study with tract-based reconstruction compared to voxel-based approach. J Neurosci Methods 2014; 224: 79–87.

29.

Carlson

Cha

Harmon-Jones

. Influence of the BDNF genotype on amygdalo-prefrontal white matter microstructure is linked to nonconscious attention bias to threat. Cereb Cortex 2014; 24: 2249–2257.

30.

Knaus

Silver

Kennedy

. Language laterality in autism spectrum disorder and typical controls: A functional, volumetric, and diffusion tensor MRI study. Brain Lang 2010; 112: 113–120.

31.

Alexander

Lee

Lazar

. Diffusion tensor imaging of the brain. Neurotherapeutics 2007; 4: 316–329.

32.

Erpelding

Davis

. Neural underpinnings of behavioural strategies that prioritize either cognitive task performance or pain. Pain 2013; 154: 2060–2071.

33.

Granziera

Daducci

Romascano

. Structural abnormalities in the thalamus of migraineurs with aura: A multiparametric study at 3 T. Hum Brain Mapp 2014; 35: 1461–1468.

34.

Kagan

Kainz

Burstein

. Hypothalamic and basal ganglia projections to the posterior thalamus: Possible role in modulation of migraine headache and photophobia. Neuroscience 2013; 248: 359–368.

35.

Yuan

Qin

. Axonal loss of white matter in migraine without aura: A tract-based spatial statistics study. Cephalalgia 2013; 33: 34–42.

36.

Holroyd

Drew

Cottrell

. Impaired functioning and quality of life in severe migraine: The role of catastrophizing and associated symptoms. Cephalalgia 2007; 27: 1156–1165.

37.

Tracey

Mantyh

. The cerebral signature for pain perception and its modulation. Neuron 2007; 55: 377–391.

38.

Jang

. The corticospinal tract from the viewpoint of brain rehabilitation. J Rehabil Med 2014; 46: 193–199.

39.

Brighina

Cosentino

Vigneri

. Abnormal facilitatory mechanisms in motor cortex of migraine with aura. Eur J Pain 2011; 15: 928–935.

40.

Rayhan

Stevens

Timbol

. Increased brain white matter axial diffusivity associated with fatigue, pain and hyperalgesia in Gulf War illness. PloS One 2013; 8: e58493–e58493.

41.

Buckalew

Haut

Aizenstein

. White matter hyperintensity burden and disability in older adults: Is chronic pain a contributor. PM R 2013; 5: 471–480; quiz 480.

42.

Moulton

Becerra

Maleki

. Painful heat reveals hyperexcitability of the temporal pole in interictal and ictal migraine states. Cereb Cortex 2011; 21: 435–448.

43.

Coppola G, Di Renzo A, Tinelli E, et al. Evidence for brain morphometric changes during the migraine cycle: A magnetic resonance-based morphometry study. Cephalalgia. 2015; 35: 783–791.

Migraine affects white-matter tract integrity: A diffusion-tensor imaging study

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Methods

Image acquisition

Data post-processing methods

Results

Motion parameters

Discussion

Summary

Clinical implications

Footnotes

Funding

Conflict of interest

Acknowledgments

References