Abstract
Objective
Most migraineurs develop cutaneous allodynia during migraines and many have cutaneous sensitization between attacks. Atypical pain modulation via the descending pain system may contribute to this sensitization and allodynia. The objective of this study was to test the hypothesis that compared to non-allodynic migraineurs, allodynic migraineurs have atypical periaqueductal gray (PAG) and nucleus cuneiformis (NCF) resting state functional connectivity (rs-fc) with other pain processing regions.
Design
Ten minutes resting-state BOLD data were collected from 38 adult migraineurs and 20 controls. Seed-based analyses compared whole-brain rs-fc with PAG and with NCF in migraineurs with severe ictal allodynia (n=8) to migraineurs with no ictal allodynia (n=8). Correlations between the strength of functional connections that differed between severely allodynic and non-allodynic migraineurs with allodynia severity were determined for all migraineurs (n=38). PAG and NCF rs-fc in all migraineurs was compared to rs-fc in controls.
Results
Migraineurs with severe allodynia had stronger PAG and NCF rs-fc to other brainstem, thalamic, insula and cerebellar regions that participate in discriminative pain processing, as well as to frontal and temporal regions implicated in higher-order pain modulation. Evidence that these rs-fc differences were specific for allodynia included: 1) strong correlations between some rs-fc strengths and allodynia severity among all migraineurs; 2) absence of overlap when comparing rs-fc differences in severely allodynic vs. non-allodynic migraineurs with those in all migraineurs vs. controls.
Conclusion
Atypical rs-fc of brainstem descending modulatory pain regions with other brainstem and higher-order pain modulating regions is associated with migraine-related allodynia.
Keywords: Migraine, Functional Connectivity, Functional Magnetic Resonance Imaging, Allodynia, Central Sensitization
Introduction
Cutaneous allodynia, a symptomatic manifestation of central sensitization, is the abnormal perception of normally non-noxious stimulation of the skin as being painful. [1] The majority of migraineurs develop cutaneous allodynia during migraine attacks. [1] Due to this allodynia, migraineurs may experience pain from light touch of the face or head, wearing earrings, eyeglasses, headbands, shaving one's face, and combing one's hair. Many migraineurs have persistent sensitization between migraine attacks (interictal), although often asymptomatic (sometimes called ‘pre-allodynia’). [2] Measurements of cutaneous pain thresholds confirm that pain thresholds are lower during migraine attacks compared to the interictal period, and that pain and pain tolerance thresholds are lower in interictal migraineurs than in healthy nonmigraine controls. [2-3] In addition to causing cutaneous allodynia, central sensitization may also reduce the effectiveness of migraine medication and may elevate the risk for development of more frequent migraine attacks. [4]
Central sensitization may develop in migraine due to atypical modulation of pain by the descending pain modulatory system, a system that primarily inhibits nociceptive transmission. [5-6] The descending pain system consists of several regions including the periaqueductal gray (PAG), nucleus cuneiformis (NCF), and rostral ventral medulla. [6-8] Prior functional imaging studies of experimental pain (e.g., heat-capsaicin model) have implicated the PAG and NCF in central sensitization. [7, 9] Furthermore, migraineurs exposed to painful stimulation have hypofunctional response of the descending pain system, suggestive of inadequate pain inhibition. [5]
Due to the clinical importance of central sensitization and allodynia in migraine and the likely role of the descending pain modulatory system in the development and/or maintenance of sensitization and allodynia in migraine, the current study further investigated the role of the descending pain modulatory system in migraine-related allodynia. Resting state functional connectivity magnetic resonance imaging (rs-fc MRI) was used to test the hypothesis that there is atypical interictal functional connectivity with two key regions of the descending pain modulatory system, the PAG and NCF, in migraineurs who have allodynia during migraine attacks.
Methods
Inclusion/Exclusion Criteria
38 subjects (ages 18-64 years) with migraine diagnosed according to International Classification of Headache Disorders II (ICHD-II) criteria and 20 healthy controls (ages 20-53 years) without migraine were recruited from the Washington University Department of Neurology and from the surrounding community. [10] All procedures were approved by Washington University's Human Research Protection Office. Subjects were excluded if they met ICHD-II criteria for medication overuse headache. Although the majority of subjects were not taking migraine prophylactic medication, use of medications that could be considered migraine prophylactic therapies was allowed as long as there were no changes in medications or their dosages within the 8 weeks prior to study participation. Potential subjects were excluded if they had any contraindication to MRI, had a prior brain injury, had a neurologic disorder other than migraine, had a psychiatric disorder other than anxiety or depression, or if they had any acute or chronic pain disorder other than migraine. Although control subjects had no history of migraine, they were not excluded if they had tension-type headache on 3 or fewer days per month.
Clinical Parameters
The following data were collected from all migraine subjects: 1) Number of years with migraine; 2) Headache frequency; 3) Migraine Disability Assessment Scale (MIDAS) score; 4) Beck Depression Inventory (BDI) score; 5) State-Trait Anxiety Inventory (STAI) scores; 6) Allodynia Symptom Checklist-12 (ASC-12) score. [11-13] The ASC-12 was used to determine the presence and severity of allodynia symptoms during the migraine attack. This validated, 12-item questionnaire quantifies subject perception of cutaneous allodynia symptoms yielding a score that is placed into one of 4 categories: 0-2 = no allodynia; 3-5 = mild allodynia; 6-8 = moderate allodynia; 9 or more = severe allodynia. [1, 14-15]
Imaging Protocol
Migraineurs were studied when they had been migraine free ≥48 hours and had not used migraine abortive medications for ≥48 hours. Migraineurs who experienced a migraine within 48 hours of their scheduled time were rescheduled for a later date. Controls were studied when they were in their usual state of good health. All structural and functional images were obtained on a Siemens MAGNETOM Trio 3T scanner (Erlangen, Germany) with total imaging matrix (TIM) technology using a 12-channel head matrix coil. Structural anatomic scans included a high-resolution T1-weighted sagittal magnetization-prepared rapid gradient echo (MP-RAGE) series (TR 2400ms, TE 1.13ms, 176 slices, 1.0mm^3 voxels) and a coarse T2-weighted turbo spin echo (TSE) series (TR 6150, TE 86.0, 36 axial slices, 1×1×4mm^3 voxels). Functional imaging was performed using a BOLD contrast-sensitive sequence (T2* evolution time = 25 ms, flip angle = 90°, resolution = 4×4×4 mm). Whole-brain EPI (echo planar imaging) volumes (MR frames) of 36 contiguous, 4mm thick axial slices were obtained every 2.5 seconds. BOLD data were collected via two 5-minute runs. During the rs-fc MRI scans, participants were instructed to keep their eyes closed, remain still, and not fall asleep.
Data Processing and Analysis
Following acquisition, fMRI BOLD data were preprocessed via standard methods used in our lab. [16-18] Briefly, all images from a single subject were combined into a 4-dimensional (x,y,z, time) time-series and adjusted for timing offsets using sinc interpolation. Images were adjusted for the slice intensity differences introduced by contiguous interleaved slice acquisition. Next, a 6-parameter rigid body realignment process was used to minimize movement-induced noise across all frames in all runs for each subject. Images were resliced by 3D cubic spline interpolation. Data were transformed into a common stereotactic space based on Talairach and Tournoux (1988) but using an in-house atlas composed of the average anatomy of 12 healthy young adults (ages 21-29 years) (see Lancaster et al., 1995; Snyder, 1996 for methods). [19-20] As part of the atlas transformation the data were resampled isotropically at 3 mm × 3 mm × 3 mm. Registration was accomplished via a 12-parameter affine warping of each individual's MP-RAGE to the atlas target, using difference image variance minimization as the objective function. The subject's T2-weighted image served as an intermediate target for transforming the BOLD images. The atlas-transformed images were checked against a reference average to ensure appropriate registration. Pre-processing for the rs-fc series was carried out in order to optimize the time-series data and to remove spurious variance. [21] These steps included removal of the linear trend and temporal band-pass filtering (.009 Hz<f< .08 Hz), Gaussian blur of 2 voxels FWHM, as well as regression of several “noise” parameters and their time-based derivatives including six motion parameters, a ventricular signal, a white matter signal, and a whole brain signal. The use of whole brain signal regression is an area of controversy because of the argument that doing so may produce spurious anti-correlations. However, evidence supporting its benefit for removal of the sub-millimeter movement related functional connectivity artifact has been published. [22-25] Work from our own laboratory substantiating this critical effect is currently under review for publication. This study used a volume censoring technique to identify and remove the aforementioned motion-related artifact that is not adequately addressed by frame realignment routines and movement parameter regression [22]. Briefly, data volumes (i.e., frames) with a frame-by-frame movement greater than 0.5 mm or a whole brain signal change greater than 0.5% were identified and eliminated.
Functional connectivity analyses used a seed-based/region of interest approach. Three mm diameter spheres were created around coordinates for the right PAG (Talairach 4, -36, -7) and the right NCF (Talairach 9, -29, -14). (Figure 1) Coordinates were chosen based upon our lab's prior resting functional connectivity work showing that these PAG and NCF regions were functionally connected to other pain matrix regions in migraine patients and based upon coordinates used in the published literature. [5, 9, 26] For each seed, a resting state time-series was extracted separately for each subject by computing the mean of the BOLD intensity of all voxels enclosed by the seed region boundaries at each MR frame (time-point). Correlations with this time-series were calculated for each voxel in the brain, then Fisher z transformed to produce a functional connectivity map for each seed in each subject.
Figure 1.
Periaqueductal gray (PAG) and nucleus cuneiformis (NCF) seed regions of interest. Seed regions were 3 mm diameter spheres centered on Talairach coordinates 4, −36, −7 for right PAG and Talairach coordinates 9, −29, −14 for right NCF.
One-sample t-tests (two-tailed) were performed on maps from all subjects (n = 58) to identify functional connections that significantly differed from zero (p ≤ .01). Two-sample t-tests were used to compare PAG and NCF resting state functional connectivity in migraineurs with severe ictal allodynia (ASC-12 score ≥9) to that of migraineurs without ictal allodynia (ASC-12 score ≤2). Regions were created from the results of these two-sample t-test images using an in-house peak-finding algorithm. This algorithm located peaks within contiguous voxels in each image and defined regions by first smoothing with a 1 mm kernel, then extracting peaks with a minimum distance of 10 mm from other peaks, a peak z-value of at least 2.0, and a minimum size of 6 voxels (3mm × 3mm × 3mm). The functional connectivity strength between each of the resulting regions and the seed region (either PAG or NCF) was calculated and two-sample t-tests (two-tailed) compared functional connectivity strengths among subject groups. If the absolute value of the functional connectivity strength between the seed region and another brain region was less than 0.1 in both subject groups (lack of significant functional connectivity in both subject groups), the functional connection was excluded from further analyses. Benjamini-Hochberg correction for multiple comparisons allowing for a false discovery rate of 2.5% was employed to identify functional connections that significantly differed between subject groups.
The strengths of functional connections that differed between migraineurs with severe allodynia and migraineurs with no allodynia were correlated with ictal allodynia scores in all 38 migraineurs using Pearson correlations. Since the 16 migraineurs (8 with severe allodynia and 8 with no allodynia) used to identify these functional connections were included in the sample of migraineurs used to determine correlations between functional connectivity strength and allodynia scores (n=38), this analysis is to be considered exploratory and p-values are not reported. Linear regression, stepwise entry, was employed to determine the set of functional connections that best explained the variance in ictal allodynia scores. In order to consider the potential effect of confounding variables on the correlations between functional connection strength and ictal allodynia scores, Pearson correlations between the strength of functional connections and number of years with migraine, headache frequency, age, state anxiety, trait anxiety, and depression scores were calculated.
The validity of PAG and NCF rs-fc differences between migraineurs with severe allodynia and migraineurs without allodynia and the specificity of findings for the presence of allodynia were further investigated by comparing PAG and NCF rs-fc strength between all migraineurs (n=38) and a group of non-migraine controls (n=20). T-tests (2-sample) were used to compare migraineurs to controls using the rs-fc strengths of those functional connections that differed in migraineurs with severe allodynia compared to migraineurs without allodynia. Benjamini-Hochberg correction for multiple comparisons allowing for a false discovery rate of 2.5% was employed to identify functional connections that significantly differed between subject groups. Voxels with rs-fc correlations to PAG or NCF that differed both between migraineurs vs. controls and in migraineurs with severe allodynia vs. migraineurs with no allodynia were identified.
Results
Subject Characteristics (Table 1)
Table 1.
Subject Characteristics.
| Subject Group | Age (years) | Sex | Headache Frequency (days/month) | Years with Migraine | ASC-12 | Allodynia Category | State Anxiety | Trait Anxiety | BDI |
|---|---|---|---|---|---|---|---|---|---|
| Migraine (n=38) | 32 (±11) | F=32 M=6 |
15 (±8) | 14 (±8) | 6 (±5) | None = 8 Mild = 12 Moderate = 10 Severe = 8 |
35 (±12) | 37 (±11) | 6 (±8) |
| Migraine - Severe Ictal Allodynia (n=8) | 32 (±8) | F=6 M=2 |
20 (±11) | 13 (±8) | 13 (±5) | Severe = 8 | 39 (±10) | 37 (±16) | 6 (±11) |
| Migraine -No Ictal Allodynia (n=8) | 33 (±16) | F=6 M=2 |
14 (±8) | 12 (±10) | 1 (±1) | None =8 | 35 (±15) | 37 (±11) | 2 (±4) |
| Control (n=20) | 34 (±10) | F=15 M=5 |
NA | NA | NA | NA | 26 (±7) | 32 (±11) | 4 (±6) |
Numbers are means followed by standard deviation in parenthesis (except for sex and allodynia category which are counts). ASC-12 scores are indicative of cutaneous allodynia symptoms during a migraine attack. State and trait anxiety scores are from the State-Trait Anxiety Inventory (STAI). Mean state and trait anxiety scores among migraineurs are within 1 standard deviation of the normal mean (mean scores in the general population are 36.4 ± 10.6). [38] Depression scores are from the Beck Depression Inventory (BDI). Average BDI scores are consistent with no depression (1-10: normal; 11-16: mild mood disturbance; 17-20: borderline clinical depression; 21-30: moderate depression; 31-40: severe depression; over 40: extreme depression). Anxiety and depression scores are available from 18/20 control subjects. F = female; M = male; ASC-12 = allodynia symptom checklist 12.
Thirty-eight migraine subjects and 20 controls were included in this study. Table 1 shows subject age and sex, headache frequency, number of years with migraine, ASC-12 scores, category of allodynia severity, STAI scores, and BDI scores. All but 8 of the 38 migraineurs (79%) reported having at least mild symptoms of cutaneous allodynia during migraine attacks, including 8 with severe, 10 with moderate, and 12 with mild allodynia. Eight of 38 (21%) migraine subjects were taking migraine prophylactic medications, including 3 of the 8 with severe ictal allodynia and 2 of the 8 with no ictal allodynia.
PAG and NCF Functional Connectivity Differs in Migraineurs with Severe Ictal Allodynia Compared to Migraineurs without Ictal Allodynia
Using the PAG as a seed region, comparison of rs-fc in migraineurs with severe ictal allodynia to migraineurs with no ictal allodynia yielded 19 functional connections that significantly differed between these migraine subgroups. (Figure 2) Functional connections that were stronger in subjects with severe allodynia than those with no allodynia (stronger positive or stronger negative BOLD temporal correlations) included PAG with: pons, thalamus, cerebellum, precuneus, posterior insula, inferior temporal cortex, and inferior and superior frontal cortex. Functional connections with the PAG that were weaker in subjects with severe allodynia (weaker positive or weaker negative BOLD temporal correlations) included: middle and superior frontal regions.
Figure 2.
Functional connectivity to periaqueductal gray (PAG) differs in migraineurs with severe allodynia compared with migraineurs with no allodynia. The strength of 19 functional connections with the PAG differed in migraine subjects with severe ictal allodynia (red squares on scatterplot) compared with migraineurs with no ictal allodynia (blue triangles on scatterplot). Average functional connectivity strength is demonstrated with open red squares (migraineurs with severe ictal allodynia) and open blue triangles (migraineurs with no ictal allodynia). Voxel locations of each region included in the scatterplot and the location of the PAG seed are shown on the brain slices. Red coloration of voxels indicates that PAG functional connectivity was more positive in migraineurs with severe ictal allodynia compared with migraineurs with no ictal allodynia. Blue coloration of voxels indicates that PAG functional connectivity was more negative in migraineurs with severe ictal allodynia compared with migraineurs with no ictal allodynia. Axial slices are shown with the left hemisphere on the left side. BOLD = blood-oxygen-level-dependent; MD = medial dorsal.
NCF rs-fc in migraineurs with severe ictal allodynia and in migraineurs with no ictal allodynia significantly differed for 30 functional connections. (Figure 3) NCF functional connections that were stronger in migraineurs with severe ictal allodynia included: dorsal pons, midbrain, ventral medulla, cerebellum, thalamus, precuneus, inferior and middle frontal cortex, superior temporal cortex, occipital cortex, and inferior and superior parietal cortex. Functional connections with NCF that were weaker in migraineurs with severe ictal allodynia included regions in: middle and superior temporal cortex and occipital cortex.
Figure 3.
Functional connectivity to nucleus cuneiformis (NCF) differs in migraineurs with severe allodynia compared with migraineurs with no allodynia. The strength of 30 functional connections with the NCF differed in migraine subjects with severe ictal allodynia (red squares on scatterplot) compared with migraineurs with no ictal allodynia (blue triangles on scatterplot). Average functional connectivity strength is demonstrated with open red squares (migraineurs with severe ictal allodynia) and open blue triangles (migraineurs with no ictal allodynia). Voxel locations of each region included in the scatterplot and the location of the NCF seed are shown on the brain slices. Red coloration of voxels indicates that NCF functional connectivity was more positive in migraineurs with severe ictal allodynia compared with migraineurs with no allodynia. Blue coloration of voxels indicates that NCF functional connectivity was more negative in migraineurs with severe ictal allodynia compared with migraineurs with no ictal allodynia. Axial slices are shown with the left hemisphere on the left side. BOLD = blood-oxygen-level-dependent; VPM = ventral posterior medial.
Correlations between PAG rs-fc and NCF rs-fc with ictal allodynia (ASC-12) scores among all 38 migraine subjects are shown in Tables 2 and 3. Linear regression analysis yielded a model consisting of functional connections between PAG and medial dorsal thalamus, precuneus, and superior temporal cortex that best accounted for the variance in ictal allodynia scores (r2 = .512, p <.001). Functional connectivity strength of the NCF with occipital, cerebellum, precuneus, and temporal cortex best accounted for the variance in ictal allodynia scores (r2 = .667, p < .001). There were no significant correlations between the strength of rs-fc with PAG or NCF and subject age, number of years with migraine, migraine frequency, state anxiety, trait anxiety, or depression scores.
Table 2.
Correlations between PAG functional connectivity strength and allodynia severity scores.
| Region Name | Region # | Coordinates | Correlation | Slope |
|---|---|---|---|---|
| Left Precuneus | 7 | −5, −75, 37 | .517 | 16.45 |
| MD Thalamus | 5 | 1, −21, 7 | .504 | 11.01 |
| Right Inferior Temporal 1 | 18 | 41, −65, −22 | −.474 | −13.45 |
| Right Inferior Cerebellum | 2 | 14, −54, −57 | .462 | 12.84 |
| Ventral Medulla | 4 | −7, −33, −44 | .418 | 8.59 |
| Right Superior Temporal | 12 | 56, −39, 20 | −.398 | −12.12 |
| Left Ventral Pons | 11 | −9, −26, −29 | .394 | 8.55 |
| Right Superior Frontal 2 | 17 | 18, 48, 32 | −.390 | −13.64 |
| Dorsal Pons | 1 | 5, −37, −26 | .389 | 11.45 |
| Left Inferior Temporal | 9 | −38, −14, - | .375 | 10.87 |
| Left MD Thalamus | 3 | −9, −21, 17 | .367 | 8.98 |
| Right Superior Frontal 1 | 14 | 10, 15, 57 | −.341 | −10.76 |
| Left Superior Frontal | 15 | −32, 42, 31 | −.330 | −12.94 |
| Right Inferior Temporal 2 | 19 | 47, −71, 2 | −.310 | −10.28 |
| Cerebellum - Posterior | 6 | 0, −61, −45 | .293 | 8.93 |
| Left Middle Frontal | 13 | −30, −9, 58 | −.282 | −10.17 |
| Right Posterior Insula | 8 | 25, −23, 14 | .277 | 9.67 |
| Right Ventral Pons | 10 | 8, −22, −31 | .245 | 6.98 |
| Left Inferior Frontal | 16 | −44, 19, 7 | −.218 | −8.98 |
Correlations between the strengths of functional connections that differed between migraineurs with no allodynia and migraineurs with severe allodynia and ASC-12 scores were calculated. P-values are not reported since the 16 subjects used to determine rs-fc differences (migraineurs with severe allodynia vs. migraineurs without allodynia) were also included in the sample of 38 migraineurs used to calculate these correlations. Talairach coordinates. Region number refers to the number assigned to that region in Figure 2. Slope = slope of the linear regression line (regression coefficient). MD = medial dorsal
Table 3.
Correlations between NCF functional connectivity strength and allodynia severity scores.
| Region Name | Region # | Coordinates | Correlation | Slope |
|---|---|---|---|---|
| Left Occipital | 27 | −31, −83, 33 | −.599 | −17.50 |
| Left Middle Frontal 2 | 26 | −39, 33, 35 | −.516 | −16.98 |
| Right Inferior Frontal 1 | 6 | 43, 29, 4 | .454 | 14.83 |
| Left Cerebellum – Anterior Lobe 2 | 11 | −15, −55, −21 | .449 | 12.80 |
| Left Cerebellum – Anterior Lobe - Medial | 3 | −10, −46, −38 | .442 | 12.41 |
| Right Occipital | 28 | 30, −81, 37 | −.441 | −17.05 |
| Right Cerebellum - Posterior Lobe 4 | 23 | 39, −63, −39 | .440 | 12.66 |
| Left Posterior Middle Temporal | 18 | −50, −56, 13 | .436 | 10.16 |
| Right Superior Temporal 1 | 4 | 37, −32, 5 | .417 | 14.21 |
| Left Precuneus | 19 | −7, −60, 36 | .413 | 10.46 |
| Left Superior Parietal | 25 | −14, −63, 55 | −.402 | −12.44 |
| Right Cerebellum - Posterior Lobe 3 | 22 | 17, −68, −23 | .396 | 11.82 |
| Right Cerebellum – Anterior Lobe - Lateral | 5 | 47, −56, −44 | .390 | 9.87 |
| Right Precuneus | 12 | 17, −54, 33 | .389 | 14.20 |
| Left Cerebellum – Anterior Lobe 1 | 1 | −6, −58, −12 | .387 | 8.93 |
| Right Cerebellum – Anterior Lobe | 2 | 6, −59, −14 | .385 | 8.28 |
| Right Pulvinar/VPM | 10 | 13, −28, 6 | .385 | 12.01 |
| Right Inferior Parietal | 30 | 52, −30, 31 | −.381 | −11.50 |
| Right Inferior Frontal 2 | 16 | 48, 25, 13 | .368 | 12.42 |
| Right Cerebellum – Posterior Lobe 1 | 15 | 31, −64, −28 | .366 | 11.07 |
| Right Cerebellum – Anterior Lobe – Medial 1 | 9 | 11, −48, −36 | .364 | 8.94 |
| Ventral Medulla | 17 | 2, −40, −50 | .342 | 7.63 |
| Right Superior Temporal 2 | 7 | 43, −48, 10 | .336 | 12.29 |
| Left Anterior Middle Temporal | 29 | −31, 2, −41 | −.311 | −8.36 |
| Left Middle Frontal 1 | 20 | −26, 57, 13 | .306 | 10.96 |
| Right Cerebellum – Anterior Lobe – Medial 2 | 14 | 11, −45, −20 | .299 | 5.42 |
| Left Pulvinar/VPM | 8 | −2, −22, 5 | .291 | 6.59 |
| Right Cerebellum - Posterior Lobe 2 | 21 | 31, −54, −42 | .289 | 8.44 |
| Dorsal Pons | 13 | 8, −35, −32 | .265 | 5.98 |
| Dorsal Midbrain | 24 | 6, −35, −12 | .260 | 5.40 |
Correlations between the strengths of functional connections that differed between migraineurs with no allodynia and migraineurs with severe allodynia and ASC-12 scores were calculated. P-values are not reported since the 16 subjects used to determine rs-fc differences (migraineurs with severe allodynia vs. migraineurs without allodynia) were also included in the sample of 38 migraineurs used to calculate these correlations. Talairach coordinates. Region number refers to the number assigned to that region in Figure 3. Slope = slope of the linear regression line (regression coefficient). VPM = ventral posterior medial
Migraineurs and non-migraine controls were compared for rs-fc strength of the 19 PAG functional connections and 30 NCF functional connections that differed between migraineurs with severe allodynia and migraineurs with no allodynia. There were no significant differences in rs-fc between controls and migraineurs for these 49 functional connections. Compared to controls, whole-brain rs-fc to PAG and NCF in migraineurs significantly differed for 26 functional connections (14 to PAG, 12 to NCF). [Supplementary Figures 1 and 2] There was no substantial overlap between rs-fc differences found when comparing migraineurs with severe allodynia to migraineurs with no allodynia with those differences found when comparing migraineurs to controls. The only overlap present was for PAG to a few voxels in inferior temporal cortex [Supplementary Figure 3].
Discussion
The major finding in this study is the demonstration of differences in rs-fc with PAG and NCF in interictal migraineurs who experience severe allodynia during migraine attacks compared to migraineurs without ictal allodynia during migraine attacks. Correlations between the strengths of several of these functional connections with the severity of ictal allodynia symptoms among our entire cohort of migraine subjects and an absence of similar differences between all migraineurs and non-migraine controls suggest that the observed differences in rs-fc are directly related to the presence of ictal allodynia. Overall, our findings further implicate the PAG and NCF, two key regions of the brainstem descending pain modulatory system, as participants in the development and/or maintenance of central sensitization in migraine.
Central Sensitization and Allodynia in Migraine
Central sensitization is a condition in which neurons have lower activation thresholds, increased responsiveness to afferent inputs, increased spontaneous activity, and enlarged receptive fields. [7, 8, 27] Trigeminal sensitization in migraine leads to cutaneous hypersensitivity, cutaneous allodynia, referral of pain beyond trigeminal innervated locations, and may predispose the migraineur to future migraine attacks. [28] In addition, central sensitization may lead to poorer response to abortive migraine medications and may increase the risk of transforming from less frequent to very frequent migraine attacks (transformation from episodic migraine to chronic migraine). [1, 4, 15, 29] Evidence for central sensitization is found in ≈75% of migraineurs during attacks and sensitization persists between attacks in a proportion of migraineurs. [2, 3, 14] In a population study of >11,000 migraineurs, 2/3 of respondents completing the ASC-12 had scores consistent with the presence of ictal allodynia. [1, 15] Severe allodynia was present in ≈1/5 of all migraineurs. [1] Thus, central sensitization and cutaneous allodynia are common in migraine patients, are responsible for several symptoms of migraine, may reduce the efficacy of migraine treatment, and may contribute to the worsening of migraine patterns. [1-4] Research leading to mechanistic descriptions of migraine-related sensitization and allodynia may eventually lead to treatments that more effectively reduce sensitization and its clinical consequences.
Descending Modulation of Pain and Central Sensitization
The descending modulatory pain system is responsible for modulating spinal cord and trigeminal nociceptive transmission. The PAG and NCF are key regions of the descending modulatory pain system, serving as the major sources of input to rostral ventral medulla, a region responsible for the final output from this pain modulating system. [5, 9, 30-31] These brainstem regions play prominent roles in modulating perceived pain intensity and are anatomically connected to each other as well as to higher-order pain processing regions in the frontal lobes, thalamus, cingulate cortex, amygdala, insula, and hypothalamus. [6, 32] Regions functionally connected to PAG, according to prior fMRI studies, include: midbrain tegmentum, substantia nigra, raphe nucleus, striatum, globus pallidus, hypothalamus, thalamus, cerebellum, anterior cingulate cortex, rostral ventromedial medulla, insula, frontal cortices, post-central gyrus, superior temporal gyrus, and inferior parietal lobule. [33-34] Although inhibition of nociceptive transmission is the predominant role of the brainstem descending pain modulatory system, the PAG, NCF, and rostral ventral medulla also facilitate nociception. [6, 9, 35-36] A balance between pain facilitation and inhibition via the descending pain system is hypothesized to be necessary for appropriate pain processing. [6] Low levels of pain inhibition, high levels of pain facilitation, or an imbalance of these two opposing processes could participate in the development or maintenance of sensitization and could contribute to the production of chronic pain.
Prior investigations have implicated the brainstem descending pain modulatory system in sensitization and allodynia. Functional imaging studies identified an inverse relationship between activity in the ventrolateral periaqueductal gray and pain intensity. [7, 37] Electrical stimulation of the ventrolateral PAG results in analgesia and reductions in allodynia. [38-39] Functional MRI of humans using punctate mechanical stimulation in an area of secondary hyperalgesia (heat/capsaicin sensitization model) showed significant increases in brainstem activation localized to two midbrain regions consistent with PAG and NCF. [9] Migraine studies have implicated PAG and NCF in migraine-related sensitization. An fMRI study of 12 interictal migraineurs measured brainstem activity in response to painful cutaneous heat stimulation. [5] Compared to controls, migraineurs had less pain-induced activation of NCF, suggesting less pain inhibition in migraineurs. A study comparing interictal PAG rs-fc in 5 migraineurs with ictal allodynia to 5 migraineurs with no ictal allodynia found the migraineurs with allodynia to have weaker PAG rs-fc to prefrontal regions, anterior cingulate, and anterior insula, findings interpreted by the investigators as further evidence that pain modulatory systems might be involved in the development of allodynia. [40]
PAG and NCF Functional Connectivity Differs in Migraineurs with Severe Allodynia
Migraineurs with severe ictal allodynia have stronger PAG and NCF functional connectivity with other brainstem regions including a region consistent with the location of the rostral ventral medulla. There are several possible explanations for these stronger PAG and NCF functional connections in migraineurs with severe allodynia: 1) frequent co-activations of descending pain modulatory system regions leads to stronger rs-fc amongst these regions; 2) stronger rs-fc develops as a compensatory response to the increased pain of allodynia, allowing for greater pain inhibition by the descending modulatory system; 3) stronger rs-fc amongst regions of the descending modulatory system results in increased facilitation of pain and thus predisposes the individual to the development of allodynia. Longitudinal studies of rs-fc and allodynia within individual migraineurs may help to differentiate amongst these proposed explanations.
Migraineurs with severe ictal allodynia also had stronger rs-fc to other regions that primarily participate in sensory-discriminative pain (i.e. intensity and location of pain) including: thalamus, cerebellum, and posterior insula (PAG only). Rs-fc connections to “higher order” pain processing regions showed a mixed picture with generally stronger connections in migraineurs with severe allodynia, although weaker connections were found for a few functional connections. These higher order pain-modulating regions predominantly participate in affective (i.e. emotional response to pain, fear, motivation to stop the painful stimulus) and cognitive (e.g. pain memory, attention, expectation) pain processing. Since these regions may be responsible for mediating activity of brainstem pain modulating regions (“top-down modulation”), atypical rsfc between higher order pain-modulating regions and brainstem regions identified in this study may be indicative of aberrant higher order control over the brainstem descending pain modulatory system. [6, 41-42]
There was very little overlap between rs-fc that differed between migraineurs with severe allodynia and those with no allodynia and rs-fc that differed between migraineurs and controls. Absence of significant overlap suggests that the rs-fc differences found between migraineurs with severe allodynia and those with no allodynia are specifically related to allodynia, and not attributable to migraine. Correlations between ASC-12 scores and several PAG and NCF functional connection strengths and an absence of correlations between rs-fc strengths and years with migraine further support this assertion. Our findings, and the fact that the majority of migraineurs develop allodynia during migraine attacks, suggest that it is important for future migraine studies investigating the descending pain modulatory system to account for potential effects of allodynia on study findings, differentiating effects of allodynia from potential effects of migraine. Future studies will also investigate similarities and differences in PAG and NCF rs-fc in migraine-related allodynia to allodynia associated with other diseases (e.g. fibromyalgia).
Study Limitations
The ASC-12 requires subjects to recall ictal allodynia symptoms and estimate their frequency during their most severe headaches. This data collection method is vulnerable to recall bias. Although prospective reporting of allodynia symptoms would avoid recall bias, it would require data collection over a prolonged time-period in order to assess usual migraine symptoms and avoid temporal sampling bias. The ASC-12 is a standard and validated tool assessing cutaneous allodynia in migraine. [1, 15] A few subjects were taking medications used for migraine prophylaxis (8/38 migraine subjects including 3/8 subjects with severe allodynia and 2/8 subjects with no allodynia). Although we are not aware of any studies that have specifically investigated the effects of migraine prophylactic medications on rs-fc in migraineurs, there are studies showing that rs-fc can change within individuals who initiate medications. [43-44] Although it is often difficult to determine whether these changes in rs-fc are directly attributable to medication effects or to improvements in the clinical condition for which the medication was prescribed, it is likely that medications have direct influences on rs-fc. [45] It is unlikely that medication use is driving our results, since a minority of subjects was taking prophylactic medications and their use was relatively balanced among the severe allodynia and no allodynia cohorts. Furthermore, examination of scatterplots graphing rs-fc strengths by subject revealed that subjects taking prophylactic medications did not consistently have the strongest or weakest rs-fc amongst all subjects (i.e. always stronger or always weaker rs-fc strength compared to the group mean). The relatively low spatial resolution of fMRI makes it difficult to definitively determine that our PAG ROI was entirely within the ventrolateral subdivision of the PAG, a subdivision of the PAG that has been implicated to participate in transmission of trigeminovascular nociception. [46-48] However, we maximized the likelihood that the PAG ROI was within the ventrolateral PAG by choosing coordinates based upon prior studies demonstrating functionally connectivity to other pain matrix regions, activation of the PAG in response to painful stimuli, visualizing the ROIs on our atlas, and limiting the ROI to a 3 mm diameter sphere. Furthermore, although we imaged 38 migraine subjects, only the 8 with severe allodynia and the 8 with no allodynia were included in the primary analysis. Inclusion of relatively small numbers of subjects in the primary analysis is a substantial limitation of this study. Results need to be validated in similar studies with larger sample sizes. Given small sample sizes, we were not able to determine potential rs-fc differences between men and women with migraine. One published migraine study has suggested that such differences might exist. [49] Correlations between ASC-12 scores and rs-fc strength were calculated including the subjects with no allodynia and severe allodynia. Thus, the correlations are considered exploratory and should be validated or refuted in future studies using new cohorts of migraineurs with varying levels of allodynia.
Conclusions
Study findings further implicate PAG and NCF, key regions of the descending pain modulatory system, as regions involved in migraine-related allodynia. Compared to migraineurs without allodynia, migraineurs with severe allodynia have atypical PAG and NCF rs-fc to other regions that modulate sensory-discriminative pain processing and with higher order pain modulating regions. Additional studies are necessary to determine if atypical PAG and NCF rs-fc predispose the migraineur to developing allodynia, or if atypical rs-fc results from allodynia. Since allodynia is common in migraine, results in increased pain, may predispose to the development of more frequent migraine attacks, and might reduce the efficacy of migraine medications, methods of treating and preventing allodynia are warranted. Treatments that “normalize” atypical PAG and NCF rs-fc might be successful in this regard.
Supplementary Material
Acknowledgements
This work was supported by the National Institutes of Health [K23NS070891] and the National Headache Foundation.
Footnotes
The authors report no conflicts of interest related to this manuscript.
References
- 1.Bigal ME, Ashina S, Burstein R, Reed ML, Buse D, Serrano D, Lipton RB. Prevalence and characteristics of allodynia in headache sufferers. Neurology. 2008;70:1525–1533. doi: 10.1212/01.wnl.0000310645.31020.b1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schwedt TJ, Krauss MJ, Frey K, Gereau RW., IV Episodic and chronic migraineurs are hypersensitive to thermal stimuli between migraine attacks. Cephalalgia. 2011;31:6–12. doi: 10.1177/0333102410365108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Burstein R, Yarnitsky D, Goor-Aryeh I, Ransil BJ, Bajwa ZH. An association between migraine and cutaneous allodynia. Annals of Neurology. 2000;47:614–624. [PubMed] [Google Scholar]
- 4.Burstein R, Collins B, Jakubowski M. Defeating migraine pain with triptans: a race against the development of cutaneous allodynia. Annals of Neurology. 2004;55:19–26. doi: 10.1002/ana.10786. [DOI] [PubMed] [Google Scholar]
- 5.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. doi: 10.1371/journal.pone.0003799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bingel U, Tracey I. Imaging CNS modulation of pain in humans. Physiology. 2008;23:371–380. doi: 10.1152/physiol.00024.2008. [DOI] [PubMed] [Google Scholar]
- 7.Mainero C, Zhang W-T, Kumar A, Rosen B, Sorensen AG. Mapping the spinal and supraspinal pathways of dynamic mechanical allodynia in the human trigeminal system using cardiac-gated fMRI. Neuroimage. 2007;35:1201–1210. doi: 10.1016/j.neuroimage.2007.01.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lee MC, Zambreanu L, Menon DK, Tracey I. Identifying brain activity specifically related to the maintenance and perceptual consequence of central sensitization in humans. Journal of Neuroscience. 2008;28:11642–11649. doi: 10.1523/JNEUROSCI.2638-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zambreanu L, Wise RG, Brooks JC, Iannetti GD, Tracey I. A role for the brainstem in central sensitisation in humans. Evidence from functional magnetic resonance imaging. Pain. 2005;114:397–407. doi: 10.1016/j.pain.2005.01.005. [DOI] [PubMed] [Google Scholar]
- 10.International Classification of Headache Disorders Committee The international classification of headache disorders, 2nd edition. Cephalalgia. 2004;24(Suppl 1):1–160. doi: 10.1111/j.1468-2982.2003.00824.x. [DOI] [PubMed] [Google Scholar]
- 11.Stewart WF, Lipton RB, Kolodner KB, Sawyer J, Lee C, Liberman JN. Validity of the Migraine Disability Assessment (MIDAS) score in comparison to a diary-based measure in a population sample of migraine sufferers. Pain. 2000;88:41–52. doi: 10.1016/S0304-3959(00)00305-5. [DOI] [PubMed] [Google Scholar]
- 12.Moran PJ, Mohr DC. The validity of Beck Depression Inventory and Hamilton Rating Scale for depression items in the assessment of depression among patients with multiple sclerosis. J Behav Med. 2005;28:35–41. doi: 10.1007/s10865-005-2561-0. [DOI] [PubMed] [Google Scholar]
- 13.Kennedy BL, Schwab JJ, Morris RL, Beldia G. Assessment of state and trait anxiety in subjects with anxiety and depressive disorders. Psychiatric Quarterly. 2001;72:263–276. doi: 10.1023/a:1010305200087. [DOI] [PubMed] [Google Scholar]
- 14.Jakubowski M, Silberstein SD, Ashkenazi A, Burstein R. Can allodynic migraine patients be identified interictally using a questionnaire? Neurology. 2005;65:1419–1422. doi: 10.1212/01.wnl.0000183358.53939.38. [DOI] [PubMed] [Google Scholar]
- 15.Lipton RB, Bigal ME, Ashina S, Burstein R, Silberstein S, Reed ML, Serrano D, Stewart WF. Cutaneous allodynia in the migraine population. Annals of Neurology. 2008;63:148–158. doi: 10.1002/ana.21211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Miezin FM, Maccotta L, Ollinger JM, Petersen SE, Buckner RL. Characterizing the hemodynamic response: effects of presentation rate, sampling procedure, and the possibility of ordering brain activity based on relative timing. Neuroimage. 2000;11:735–739. doi: 10.1006/nimg.2000.0568. [DOI] [PubMed] [Google Scholar]
- 17.Schlaggar BL, Brown TT, Lugar HM, Visscher KM, Miezin FM, Petersen SE. Functional nueroanatomical differences between adults and school-age children in the processing of single words. Science. 2002;296:1476–1479. doi: 10.1126/science.1069464. [DOI] [PubMed] [Google Scholar]
- 18.Brown TT, Lugar HM, Coalson RS, Miezin FM, Petersen SE, Schlaggar BL. Developmental changes in human cerebral functional organization for word generation. Cerebral Cortex. 2005;15:275–290. doi: 10.1093/cercor/bhh129. [DOI] [PubMed] [Google Scholar]
- 19.Lancaster JL, Glass TG, Lankipalli BR, Downs H, Mayberg H, Fox PT. A modality-independent approach to spatial normalization of tomographic images of the human brain. Human Brain Mapping. 1995;3:209–223. [Google Scholar]
- 20.Snyder AZ. Difference image versus ratio image error function forms in PET-PET realignment. In: Myer R, Cunningham VJ, Bailey DL, Jones T, editors. Quantification of Brain Function Using PET. Academic Press; San Diego, Ca: 1996. pp. 131–137. [Google Scholar]
- 21.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;102:9673–9678. doi: 10.1073/pnas.0504136102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage. 2012;59:2142–2154. doi: 10.1016/j.neuroimage.2011.10.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Satterthwaite TD, Elliott MA, Gerraty RT, Ruparel K, Loughead J, Calkins ME, Eickhoff SB, Hakonarson H, Gur RC, Gur RE, Wolf DH. An improved framework for confound regression and filtering for control of motion artifact in preprocessing of resting-state functional connectivity data. Neuroimage. 2013;64:240–256. doi: 10.1016/j.neuroimage.2012.08.052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Satterthwaite TD, Wolf DH, Loughead J, Ruparel K, Elliott MA, Hakonarson H, Gur RC, Gur RE. Impact of in-scanner head motion on multiple measures of functional connectivity: relevance for studies of neurodevelopment in youth. Neuroimage. 2012;60:623–632. doi: 10.1016/j.neuroimage.2011.12.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Van Dijk KRA, Sabuncu MR, Buckner RL. The influence of head motion on intrinsic functional connectivity MRI. Neuroimage. 2012;59:431–438. doi: 10.1016/j.neuroimage.2011.07.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Moulton EA, Becerra L, Maleki N, Pendse G, Tully S, Hargreaves R, Burstein R, Borsook D. Painful heat reveals hyperexcitability of the temporal pole in interictal and ictal migraine states. Cerebral Cortex. 2011;21:435–448. doi: 10.1093/cercor/bhq109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Burstein R, Yamamura H, Malick A, Strassman AM. Chemical stimulation of the intracranial dura induces enhanced responses to facial stimulation in brain stem trigeminal neurons. Journal of Neurophysiology. 1998;79:964–982. doi: 10.1152/jn.1998.79.2.964. [DOI] [PubMed] [Google Scholar]
- 28.Bartsch T, Goadsby PJ. The trigeminocervical complex and migraine: current concepts and synthesis. Current Pain and Headache Reports. 2003;7:371–376. doi: 10.1007/s11916-003-0036-y. [DOI] [PubMed] [Google Scholar]
- 29.Bigal ME, Lipton RB. Modifiable risk factors for migraine progression. Headache. 2006;46:1334–1343. doi: 10.1111/j.1526-4610.2006.00577.x. [DOI] [PubMed] [Google Scholar]
- 30.Bausbaum AI, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Ann Rev Neurosci. 1984;7:309–338. doi: 10.1146/annurev.ne.07.030184.001521. [DOI] [PubMed] [Google Scholar]
- 31.Behbehani MM, Zemlan FP. Response of nucleus raphe magnus neurons to electrical stimulation of nucleus cuneiformis: role of acetylcholine. Brain Research. 1986;369:110–118. doi: 10.1016/0006-8993(86)90518-4. [DOI] [PubMed] [Google Scholar]
- 32.Hadjipavlou G, Dunckley P, Behrens TE, Tracey I. Determining anatomical connectivities between cortical and brainstem pain processing regions in humans: a diffusion tensor imaging study in healthy controls. Pain. 2006;123:169–178. doi: 10.1016/j.pain.2006.02.027. [DOI] [PubMed] [Google Scholar]
- 33.Kong J, Tu P-C, Zyloney C, Su T-P. Intrinsic functional connectivity of the periaqueductal gray, a resting fMRI study. Behavioural Brain Research. 2010;211:215–219. doi: 10.1016/j.bbr.2010.03.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Linnman C, Beucke JC, Jensen KB, Gollub RL, Kong J. Sex similarities and differences in pain-related periaqueductal gray connectivity. Pain. 2012;153:444–454. doi: 10.1016/j.pain.2011.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Fields HL, Bry J, Hentall I, Zorman G. The activity of neurons in the rostral ventral medulla of the rat during withdrawal from noxious heat. Journal of Neuroscience. 1983;3:2545–2552. doi: 10.1523/JNEUROSCI.03-12-02545.1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Haws CM, Williamson AM, Fields HL. Putative nociceptive modulatory neurons in the dorsolateral pontomesencephalic reticular formation. Brain Research. 1989;483:272–282. doi: 10.1016/0006-8993(89)90171-6. [DOI] [PubMed] [Google Scholar]
- 37.Lee BH, Park SH, Won R, Park YG, Sohn JH. Antiallodynic effects produced by stimulation of the periaqueductal gray matter in a rat model of neuropathic pain. Neuroscience Letters. 2000;291:29–32. doi: 10.1016/s0304-3940(00)01375-6. [DOI] [PubMed] [Google Scholar]
- 38.Hosobuchi Y, Adams JE, Linchitz R. Pain relief by electrical stimulation of the central gray matter in humans and reversal by naloxone. Science. 1977;197:183–186. doi: 10.1126/science.301658. [DOI] [PubMed] [Google Scholar]
- 39.Pertovaara A, Kontinen VK, Kalso EA. Chronic spinal nerve ligation induces changes in response characteristics of nociceptive spinal dorsal horn neurons and in their descending regulation originating in the periaqueductal gray in the rat. Experimental Neurology. 1997;147:428–436. doi: 10.1006/exnr.1997.6555. [DOI] [PubMed] [Google Scholar]
- 40.Mainero C, Boshyan J, Hadjikhani N. Altered functional MRI resting-state connectivity in periaqueductal gray networks in migraine. Annals of Neurology. 2011;70:838–845. doi: 10.1002/ana.22537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Lorenz J, Minoshima S, Casey KL. Keeping pain out the mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain. 2003;126:1079–1091. doi: 10.1093/brain/awg102. [DOI] [PubMed] [Google Scholar]
- 42.Valet M, Sprenger T, Boecker H, Willoch F, Rummeny E, Conrad B, Erhard P, Tolle TR. Distraction modulates connectivity of the cingulo-frontal cortex and the midbrain during pain: an fMRI analysis. Pain. 2004;109:399–408. doi: 10.1016/j.pain.2004.02.033. [DOI] [PubMed] [Google Scholar]
- 43.Lui S, Li T, Deng W, Jiang L, Wu Q, Tang H. Short-term effects of antipsychotic treatment on cerebral function in drug-naïve first-episode schizophrenia revealed by “resting state” functional magnetic resonance imaging. Arch Gen Psychiatry. 2010;67:783–792. doi: 10.1001/archgenpsychiatry.2010.84. [DOI] [PubMed] [Google Scholar]
- 44.Konova AB, Moeller SJ, Tomasi D, Volkow ND, Goldstein RZ. Effects of methylphenidate on resting-state functional connectivity of the mesocorticolimbic dopamine pathways in cocaine addiction. JAMA Pscyhiatry. 2013;70:857–868. doi: 10.1001/jamapsychiatry.2013.1129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.McCabe C, Mishor Z. Antidepressant medications reduce subcortical-cortical resting-state functional connectivity in healthy volunteers. Neuroimage. 2011;57:1317–1323. doi: 10.1016/j.neuroimage.2011.05.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Hoskin KL, Bulmer DCE, Lasalandra M. Fos expression in the midbrain periaqueductal grey after trigeminovascular stimulation. J Anat. 2001;197:29–35. doi: 10.1046/j.1469-7580.2001.19810029.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Keay KA, Bandler R. Vascular head pain selectively activates ventrolateral periaqueductal gray in the cat. Neurosci Lett. 1998;245:58–60. doi: 10.1016/s0304-3940(98)00168-2. [DOI] [PubMed] [Google Scholar]
- 48.Knight YE, Goadsby PJ. The periaqueductal grey matter modulates trigeminovascular input: a role in migraine? Neuroscience. 2001;106:793–800. doi: 10.1016/s0306-4522(01)00303-7. [DOI] [PubMed] [Google Scholar]
- 49.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;135:2546–2559. doi: 10.1093/brain/aws175. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.