Journal Logo

Research Paper

Intradural artery dilation during experimentally induced migraine attacks

Christensen, Casper E.a; Younis, Samairaa; Lindberg, Ulrichb; de Koning, Patrickc; Tolnai, Danield; Paulson, Olaf B.e; Larsson, Henrik B.W.b; Amin, Faisal M.a; Ashina, Messouda,*

Author Information
doi: 10.1097/j.pain.0000000000002008

1. Introduction

Migraine is a complex neurological disease with attacks comprising elaborate interplay of the trigeminovascular system and deep brain structures.6,8 An interesting trait of migraine is the ability to trigger attacks endogenously (eg, stress, menstruation, changes in sleep patterns etc.) or exogenously (alcohol, tobacco, pharmacological agents etc).5,30,31 This inducibility enables researchers to examine migraine biomarkers in vivo in controlled settings.5 The dominant feature of a migraine attack is headache, but the exact mechanisms of pain remain unknown.8,17 Although migraine headache presumably comes from activation of meningeal nociceptors,8,19 we are hindered by an inability to study this activation in humans in vivo. However, the dural vasculature—including the middle meningeal artery (MMA)—is affected by neuropeptide release from trigeminal afferents and can thereby serve as a proxy for nociceptor activation.7,20,24,27 Magnetic resonance angiography (MRA) studies examined the extracranial part of MMA assuming that the extracranial segment mirrors the intradural compartment, and thus reflects changes in the meningeal afferents during migraine attacks.1,4,20,32 To the best of our knowledge, no studies have been performed directly on the intradural segments of MMA in vivo. Two migraine models are of particular interest for angiographic purposes: Calcitonin gene-related peptide (CGRP) is known for inducing migraine attacks with concomitant pain side dilation of extracranial MMA and the middle cerebral artery (MCA)4; and sildenafil is a potent migraine trigger that induces attacks without changes in MCA or the superficial temporal artery (STA), recorded using ultrasonography.23

Here, we investigate the intradural MMA and the MCA during CGRP- and sildenafil-induced attacks of migraine without aura in a double-blind crossover study. We hypothesized that intradural MMA and the MCA dilate during provoked migraine attacks.

2. Methods

2.1. Patient recruitment

We recruited patients with migraine without aura through advertisements at hospitals and educational institutions as well as a Danish recruitment website (www.forsoegsperson.dk). Male and female migraine patients were eligible for inclusion if they had migraine without aura according to the classification criteria of the International Headache Society version 3-beta (ICHD-3 beta).16 Patients had to be 18 to 50 years of age and experience attacks of migraine at least once every other month. Exclusion criteria were: inconsistent headache laterality, chronic migraine, any other primary headache disorder (apart from tension-type headache), daily medication intake (apart from oral contraceptives), pregnant or breastfeeding females, daily smoking, a history of serious somatic or psychiatric disease, and hypotension or hypertension (systolic blood pressure >150 or <90 mm Hg and/or diastolic blood pressure >90 or <50 mm Hg). Furthermore, patients were excluded if there were contraindications for magnetic resonance imaging (MRI) (ie, ferromagnetic implants, recent surgical procedures, claustrophobia etc.) and/or if there was evidence of any other condition deemed incompatible with participation by the investigators.

All patients underwent a routine medical examination, and pregnancy testing was performed for patients of child-bearing potential. The study was conducted at Rigshospitalet Glostrup in Denmark from September 2017 to December 2018.

2.2. Standard protocol approvals, registrations, and patient consents

All patients provided written informed consent in accordance with the Declaration of Helsinki, and the study was approved by the ethics committee of the Capital Region of Denmark (H-15019063). The study is part of a larger study that is registered at ClinicalTrials.gov (NCT03143465), and other parts of the study have been and/or will be published elsewhere.

2.3. Study design

Patients reported to the clinic headache-free for at least 48 hours. Patients were fasting for 4 hours before study start, and coffee, tea, alcohol, cocoa, and tobacco were not allowed for 12 hours up until study start. First, a baseline MRI scan was performed after which the patients were relocated to a laboratory for the following migraine provocation protocol. On 2 separate study days, patients were allocated to receive a tablet of 100 mg sildenafil (TEVA pharmaceutical industries Ltd, Petah Tikva, Israel) or CGRP (Tocris Bioscience, Bristol, United Kingdom) as an IV infusion (1.5 μg/minute for 20 minutes). Drug administration was conducted in double-blind fashion, which was achieved by administering both a tablet and an infusion on each day with only one compound containing active substance. After drug administration, patients rested in the supine position for 90 minutes while being monitored with blood pressure, blood oxygenation, respiratory rate, and heart rate recordings along with a headache interview every 10 minutes. After the 90 minutes had passed, monitoring was limited to headache interviews every 30 minutes. Six hours after drug administration, patients underwent a second MRI scan similar to the baseline scan and were discharged hereafter. During the second MRI, vital parameter monitoring was resumed. A study design flowchart is depicted in Figure 1.

Figure 1.
Figure 1.:
Study design. CGRP, calcitonin gene-related peptide; MRI, magnetic resonance imaging.

2.4. Data acquisition

Vital parameters were monitored on a Veris monitor (Medrad, Warrendale, PA). Headache characteristics interviews were performed by investigators with experience in headache diagnostics using a standardized interview based on the diagnostic criteria for migraine without aura.16

All MRI scans were performed on the same 3.0 T Philips Achieva dStream MRI Scanner (Philips Medical Systems, the Netherlands) using a 32-channel phased-array head coil. Three-dimensional time-of-flight MRA was conducted for vessel imaging. A scout MRA was performed with field of view (FOV) 200 × 200 × 150 mm3, acquired matrix size 200 × 134, acquired voxel resolution 1.00 × 1.50 × 2.00 mm3, reconstructed resolution 0.39 × 0.39 × 1.00 mm3, repetition time (TR) 23 ms, echo time (TE) 3.5 milliseconds, flip angle 18°, SENSE p reduction 2, 2 chunks, and a duration of 2 minutes 46 seconds. Two subsequent MRAs were performed for visualization of MMA, and STA (MMA scan) and MCA (MCA scan) with the following parameters: MMA scan had FOV 200 × 200 × 36.75 mm3, acquired matrix size 800 × 570, acquired voxel resolution 0.25 × 0.35 × 0.70 mm3, reconstructed voxel resolution 0.20 × 0.20 × 0.35 mm3, TR 23 milliseconds, TE 3.5 milliseconds, flip angle 18°, SENSE p reduction 2.5, 3 chunks, and a duration of 13 minutes 06 seconds. Middle cerebral artery scan used FOV 200 × 200 × 12.25 mm3, acquired matrix size 784 × 571, acquired voxel resolution 0.26 × 0.35 × 0.70 mm3, reconstructed voxel resolution 0.20 × 0.20 × 0.35 mm3, TR 23 milliseconds, TE 3.5 milliseconds, flip angle 18°, SENSE p reduction 2.5, 1 chunk, and a duration of 2 minutes 59 seconds.

2.5. Data analysis

Magnetic resonance angiography data were transferred to a separate workstation and analyzed using LKEB LAVA vessel contour detection software.21 The software allows the user to define start- and end-points of a vessel segment for contour analysis; then, it detects a pathline of the vessel and fits a 3D model to the underlying image data to allow for circumference readings along the centerline of the chosen segment. If slices along the centerline were distorted or circumference was otherwise immeasurable, said slice was discarded. The workflow enabled the operator to choose the same segment of every vessel within subject across scans to provide reliable measurements of change over time and still be blinded to patient, scan state, and pain side. Ultimately, a 5-mm segment is chosen comprising up to 34 slices and the mean circumference is computed for that segment at that time point. We analyzed the intradural segment of the MMA, identified as the innermost part covered by the FOV, approximately 20 mm after MMA enters the skull through the foramen spinosum. Furthermore, we analyzed STA right after branching from the external carotid artery, and we analyzed the M1 segment of MCA. Headache data were used to determine whether patients developed migraine attacks based on the following criteria: Attacks fulfilling either (1) or (2):

  • (1) Headache meeting criteria C and D for migraine without aura according to ICHD-3 beta.16 C. Headache has at least 2 of the following characteristics: Unilateral location; pulsating quality; moderate or severe pain intensity; aggravation by activity (coughing during in-hospital phase); D. During headache, at least one of the following: Nausea and/or vomiting; photophobia and phonophobia.
  • (2) Headache described as mimicking the patient's usual migraine attack and treated with acute migraine medication.

If patients experienced pain on both sides of the head, we defined pain side depending on which the patient reported the most severe headache for side-to-side analyses. If no such distinction could be made, headache was defined as bilateral.

2.6. Statistical analyses

Headache characteristics are reported for participants included in the final analysis, and changes in mean arterial pressure and heart rate in those participants are presented descriptively as relative changes from baseline.

Our primary end-point was change in arterial circumference of the intradural MMA during attacks of migraine compared to baseline within subjects who suffered migraine attacks after drug administration across drugs and sides. Second, we analyzed changes in arterial circumference during migraine in the STA and MCA. Circumference change was evaluated with linear mixed-effects models with scan state as fixed effects and subject with sides nested as random effects. Secondary analyses of drug and pain side specificity for arterial dilation were performed descriptively as percentage change from baseline dependent on pain side. Analyses were performed using R (version 3.5.2) with lme4 package (version 1.1-20), and P-values are reported as two-tailed with a 5% significance level.

2.7. Data availability

Data can be made available upon reasonable request to the corresponding author.

3. Results

Thirty-four patients with migraine without aura were recruited. Eight patients did not complete both study days resulting in 60 completed experiments. Forty experiments resulted in migraine attacks at scan 2. One of these was scanned outside the time window (6 hours after drug administration) and was excluded from the final analysis, which left 39 remaining (20 after CGRP and 19 after sildenafil) (Fig. 2).

Figure 2.
Figure 2.:
Experiment completion flowchart. CGRP, calcitonin gene-related peptide.

3.1. Intradural arterial changes during attacks of migraine

Circumference of the intradural segment of MMA, across both drugs and both sides, was increased by 0.11 mm (P = 0.005) during migraine attacks compared to baseline (baseline intercept: 3.18 mm ± 0.07 [SE]). The dilation corresponds to a relative change of 3.6% [95% CI: 1.4%-5.7%]. We found similar dilation in the 2 interventions: 3.4% [95% CI: 0.3%-6.6%] for CGRP and 3.7% [95% CI: 0.8%-6.6%] for sildenafil (Fig. 3). Explorative analysis revealed intradural MMA dilation in patients reporting unilateral—but not bilateral—pain during migraine attacks (Fig. 4).

Figure 3.
Figure 3.:
Dilation of cranial arteries during migraine. Dilatory patterns depicted as mean ± SE in each of the 3 vascular segments. Black lines represent mean dilation across drugs and sides. Red and blue are dilation within drug and across sides, respectively. MMA, middle meningeal artery; MCA, middle cerebral artery; STA, superficial temporal artery.
Figure 4.
Figure 4.:
Dilation of cranial arteries by headache location. Dilatory patterns depicted as mean ± SE in the 3 vessel segments divided into pain side, nonpain side, and bilateral pain headache reports (calcitonin gene-related peptide and sildenafil combined). MMA, middle meningeal artery; MCA, middle cerebral artery; STA, superficial temporal artery. MMA pain side 6.3% dilation (n = 28), nonpain side 3.7% dilation (n = 28), and bilateral pain 0.0% change (n = 11). MCA pain side 9.3% dilation (n = 28), nonpain side 8.8% dilation (n = 28), and bilateral pain 10.6% change (n = 11). STA pain side 3.2% dilation (n = 28), nonpain side 2.4% dilation (n = 28), and bilateral pain 0.5% change (n = 11).

3.2. Intracerebral and extracranial arterial dilation during migraine attacks

We found a mean dilation of MCA of 9.4% [95% CI: 7.1%-11.7%] and a 2.3% dilation [95% CI: 0.2%-4.4%] of the STA (Fig. 3).

3.3. Attack characteristics

Individual attack characteristics for the 39 attacks included in the final analysis is depicted in Table 1. The 2 interventions yielded similar attack characteristics; however, median time to onset of migraine was 180 minutes after CGRP and 270 minutes after sildenafil. Because all subjects included in the analysis were scanned before and 6 hours after drug administration, median time from onset of migraine to scan was 3 hours (range 0-5.5 hours) for CGRP and 1.5 hours (range 0-4.7 hours) for sildenafil.

Table 1 - Clinical characteristics of headache and associated symptoms after CGRP and sildenafil.
ID Drug Time to peak headache Headache characteristics Associated symptoms Mimics usual migraine Migraine attack onset Treatment (time)/efficacy
1 CGRP
Sildenafil
1 h
7 h
Bilat(left)/5/Throb/+
Bilat(left)/5/Throb/+
+/+/+
+/+/+
Yes
Yes
40 min
5 h
NS
Suma 50 mg (10 h)/No; I + P (23 h)/NA*
2 Sildenafil 8 h Bilat(left)/9/Throb+Pres/+ +/+/+ Yes 4.5 h Suma 100 mg (7 h)/No; NS (8 h)/No; NS (9 h)/Yes
3 Sildenafil 7 h Right/8/Throb/+ −/+/+ Yes 80 min NS
4 CGRP

Sildenafil
80 min

8 h
Bilat(left)/4/Throb+Pres/+

Bilat(right)/8/Throb+Pres/+
−/+/+

−/+/+
Yes

Yes
2.5 h

6 h
None

CP × 2 (9 h)/Yes
5 CGRP 7 h Bilat(left)/7/Throb/+ +/+/+ Yes 3 h Suma 50 mg (8 h)/No; suma 50 mg (14 h)/NS; suma 50 mg (22 h)/Yes
6 CGRP 10 h Bilat(right)/3/Throb/+ +/+/+ No 3.5 h None
7 CGRP 7 h Bilat(right)/5/Throb/+ +/+/+ Yes 6 h 2 AC + P (8 h)/Yes
8 CGRP
Sildenafil
70 min
10 h
Bilat/4/Throb/+
Bilat/7/Throb/+
+/+/+ Yes
Yes
70 min
2.5 h
None
I (11 h)/No
9 CGRP
Sildenafil
7 h
4.5 h
Bilat(right)/6/Throb/+
Bilat(right)/7/Throb/+
+/+/+
+/+/+
Yes
Yes
4.5 h
2.5 h
I + P (8 h)/Yes; 2 AC (12 h)/Yes
10 CGRP 6 h Bilat(right)/7/Throb+Pres/+ +/+/+ Yes 70 min Suma s.c. + O (6 h)/Yes
11 CGRP
Sildenafil
4.5 h
8 h
Bilat/4/Throb+Pres/+
Bilat/7/Throb/+
−/+/+
−/+/+
No
Yes
6 h
5.5 h
2 AC (7 h)/Yes
I + P (7 h)/No; I (8 h)/Yes
12 CGRP
Sildenafil
5.5 h
6 h
Left/8/Throb/+
Left/2/Pres/+
+/+/+
−/+/−
Yes
Yes
3 h
6 h
2 AC + O (7 h)/No; I + P (9 h)/No; 2 AC (12 h)/Yes, AC (23)/No
1.5 AC (9 h)/Yes
13 CGRP

Sildenafil
7 h

7 h
Bilat(left)/7/Throb/+

Bilat(right)/9/Throb/+
+/+/+

+/+/+
Yes

Yes
30 min

5 h
Suma 50 mg (7 h)/No; suma 50 mg (9 h)/Yes
Suma s.c. (7 h)/Yes
14 Sildenafil 7 h Bilat(left)/10/Pres/+ +/+/+ Yes 5.5 h O + Diclofenac (8 h)/Yes
15 CGRP

Sildenafil
5.5 h

5 h
Bilat/8/Throb/+

Bilat/7/Pres/+
+/+/+

+/+/+
Yes

Yes
2.5 h

4.5 h
Suma 25 mg (7 h)/No; suma 25 mg (10 h)/Yes
Suma 25 mg (7 h)/Yes
16 CGRP
Sildenafil
5 h
4 h
Bilat/6/Throb/+
Bilat/8/Throb+Pres/+
−/+/+
+/+/+
Yes
Yes
5.5 h
3 h
AC (7 h)/Yes
AC + O (7 h)/Yes
17 CGRP
Sildenafil
6 h
6.5 h
Bilat(left)/5/Throb/+
Left/9/Throb+Pres/+
+/+/+
+/+/+
Yes
Yes
5.5 h
2.5 h
AC (8 h)/NS; 2 AC (17 h)/Yes
Suma s.c. + O (7 h)/Yes
18 CGRP 7 h Left/4/Pres/+ +/+/− Yes 5 h AC (8 h)/No
19 CGRP

Sildenafil
6.5 h

6.5 h
Left/6/Throb/+

Left/8/Throb/+
+/+/+

+/+/+
Yes

Yes
3 h

3 h
Suma 50 mg (7 h)/No; suma 50 mg (10 h)/NS
None
20 CGRP 6 h Bilat/10/Throb+Pres/+ +/−/− Yes 3.5 h Suma s.c. (7 h)/Yes
21 CGRP
Sildenafil
8 h
6 h
Left/5/Throb/+
Left/4/Throb/+
+/+/+
−/+/+
Yes
Yes
5.5 h
6 h
None
None
22 CGRP
Sildenafil
7 h
4 h
Bilat/6/Throb/+
Bilat/6/Throb/+
−/+/+
+/+/+
Yes
Yes
50 min
3 h
I + P (7 h)/NA; P (8 h)/Yes
I + P (7 h)/NA; AC (10 h)/No
23 Sildenafil 9 h Bilat(left)/8/Pres/+ +/+/+ Yes 3 h Suma 50 mg (7 h)/No
24 Sildenafil 8 h Bilat(left)/4/Pres/+ −/+/+ Yes 6 h None
25 CGRP

Sildenafil
7 h

6 h
Bilat(right)/6/Throb+Pres/+

Bilat(right)/5/Throb/+
−/+/+

+/+/+
Yes

Yes
50 min

4 h
Suma 50 mg + I (7 h)/Yes

Suma 50 mg + I (8 h)/Yes
Migraine attack frequency: Days/Months. Headache characteristics: Localization at time of scan (parentheses denote most affected side)/Intensity/Quality/Aggravation. Associated symptoms: Nausea/Photophobia/Phonophobia. The criteria for a migraine-like attack are described in “Methods.” Treatment efficacy: ≥ 50% decrease of headache intensity within 2 hours. Bilat: Bilateral. Uni: Unilateral. Thr: Throbbing. Pr: Pressing. NS: not specified (missing data). NR: Not reported. Spon: Spontaneous migraine attack. Diclofenac: Diclofenac 100 mg (suppository). I: Ibuprofen 400 mg. Suma s.c.: Sumatriptan 6 mg (subcutaneous injection). CP: Codeine 30.6 mg + Paracetamol 500 mg. P: Paracetamol 1 g. Suma: Sumatriptan. AC: Aspirin 500 mg + Caffeine 50 mg. O: Ondansetron 16 mg (suppository).
*Does not fulfill the associated symptoms criteria, but the attack is reported to mimic the patient's usual attack.
No data from >24 hours.
Shifted from left to right at 7 hours.
CGRP, calcitonin gene-related peptide.

3.4. Vital parameters

Relative changes to mean arterial blood pressure and heart rate after drug administration are depicted in Figure 5 showing immediate drop in blood pressure after CGRP with a concomitant increase in heart rate.

Figure 5.
Figure 5.:
Vital parameters after drug administration. Relative changes in mean arterial pressure and heart rate after administration of each intervention. CGRP, calcitonin gene-related peptide; MAP, mean arterial pressure.

4. Discussion

We present evidence of intradural MMA dilation during migraine attacks without aura in patients provoked with CGRP and sildenafil. In addition, we demonstrate that both CGRP- and sildenafil-induced attacks are associated with dilation of the cerebral and meningeal arteries. Collectively, our data provide further evidence supporting an association between meningeal artery dilation and unilateral—but not bilateral—migraine attacks.

4.1. Meningeal vasodilation and migraine

The dura mater—and correspondingly the MMA—is densely innervated by trigeminal fibers with projections to higher-order neurons in the brainstem and spinal cord.13,26,33 Activation of trigeminal fibers leads to release of various neuropeptides (eg, CGRP), which in turn acts as vascular modulators on the smooth muscle cells of the meningeal arteries in a positive feedback loop.7,14,27 We suggest that dilation of the intradural MMA in this study represents this feedback loop and is caused by vasoactive neuropeptide release from sensory fibers in the cranial meninges during migraine. In vascular smooth muscle and endothelial tissue, the cellular actions of CGRP and sildenafil promote vasodilation,7,10,18 but it is unknown which cells are responsible for migraine induction.5 With both sildenafil and CGRP being vasoactive, one could speculate that a drug effect per se could influence our results. Recently, we reported time differences in vasodilation with CGRP and sildenafil within the first 2 hours after administration,10,11 which is likely due to their pharmacokinetic profiles (CGRP has a half-life of ∼10 minutes and sildenafil 4.1 hours).7,28 However, it is unlikely that the vascular differences reported here are influenced by pharmacokinetics, as those patients who reported bilateral pain did not exhibit sustained dilation of the MMA. Furthermore, neither CGRP nor sildenafil dilates the MCA in healthy volunteers.10,22 We are not familiar with any studies that reported MR angiographic data 6 hours after administration of sildenafil or CGRP in healthy volunteers for comparison. Because of the difference in vasoreactivity in healthy volunteers together with the fact that—aside from time to onset—migraine attacks were similar between the 2 drugs in patients (Table 1), we are confident that dilation of the intracranial vasculature—in this study—is a response to migraine rather than a direct drug effect. This is in accordance with recent studies that have shifted the focus of vasodilation from being the driver of migraine pain to a second-order response to trigeminal activation.20,27 The apparent difference between sildenafil- and CGRP-induced attacks on MCA dilation (Fig. 3) may reflect differences in timing. Median time since onset of attack until scan was different: 3hours for CGRP and 1.5 hours for sildenafil, meaning that patients were arguably further along in their progressing migraine attack during the second scan on the CGRP day than on the sildenafil day.

Unilateral migraine attacks drive the intradural dilation (Fig. 4). This corresponds to previous findings of the pain side being more dilated than the nonpain side.4,20 We propose that our findings may reflect differences in trigeminovascular activation in different subphenotypes of migraine—ie, bilateral migraine headache is unlikely of meningeal origin as unilateral migraine headache. This aligns with studies from the 1980s that reported different vascular properties of migraine in different phenotypes based on whether compression of extracranial arteries would alleviate pain.12 Vasoactive intestinal peptide dilates the cranial arteries without causing severe headaches or migraine,2,29 which suggests that vasodilation per se is not the source of head pain. Instead, the intradural and intracerebral dilation, we report, is a response to ongoing activation of trigeminal fibers throughout the course of a migraine attack.

4.2. Vasodilation is time-dependent

Different patterns of vasodilation have been reported in previous human migraine studies ranging from no dilation of MMA1,32 and MCA32 to substantial dilation of one and/or both.1,2,4,20 A possible factor in vasodilation is the timing of attack scans across studies because vascular changes during migraine seem to be affected by time from onset of migraine to examination.35Figure 6 outlines 4 studies—including the present—and their findings concerning dilation of MMA and MCA. The 4 studies encompass a range of times from onset of migraine to scan from immediate scanning to scans performed 21 hours after onset.1,4Figure 6 further shows a proposed model of the time course of vascular reactions to migraine. It seems apparent that unilateral attacks of migraine are associated with both intracerebral and meningeal vasodilation and that pain side dilation precedes nonpain side dilation. This corresponds to clinical reports from patients experiencing unilateral pain in the beginning of their attacks and a more diffuse bilateral pain in later phases.9 In later phases of migraine, evidence suggests that meningeal activation recedes while cerebral arteries are still affected (Fig. 6D). Clinical observations point to migraine pain laterality as a feature that is also influenced by time. That is, pain localization changes within patient within attack over time, which adds to the intricacies of vasodilatory patterns in migraine. To further investigate this time-dependency of dilatory reactions in migraine, it would be interesting to conduct numerous consecutive scans all the way from drug administration into later phases of migraine, perhaps using a faster MRA protocol to further increase temporal resolution.

Figure 6.
Figure 6.:
Proposed model of vasoreactivity in migraine. From onset of migraine and during the course of unilateral attacks, we propose that the pain side dilates followed by the nonpain side in middle meningeal artery (MMA) and middle cerebral artery (MCA). Vasodilation as reported in 3 previous studies1,4,20 as well as the current study during migraine is inserted at approximated time points. Axes are arbitrary with y-axis describing relative dilation from baseline and x-axis describing time passed from onset of attack.

The previous reports that sildenafil-induced attacks were not accompanied by arterial dilation might also be explained by timing because measurements were stopped after 180 minutes and most patients did not yet have a migraine attack at that time,23 which corresponds well to our findings, namely that median time to onset of attack after sildenafil was 270 minutes.

4.3. Strengths and limitations

Our study draws from a line of experience in MRA analyses with high test–retest reliability and minimal operator bias.3,11 Along with the use of well-established provocation regimes,5 we are confident in our findings. Attacks yielded by our 2 study drugs were quite similar with respect to headache and associated symptoms and met our predefined criteria for provoked migraine attacks (Table 1). Two participants reported headache that did not mimic their usual attacks, but still fulfilled IHS criteria (Table 1). Migraine attack characteristics exhibit high interindividual and intraindividual heterogeneity, and we suggest that this variability causes responses to the “mimics” question to yield high positive predictive values but low negative predictive values.25,36 The purpose of our study was to investigate the pathophysiological mechanism of the overarching disease process, ie, migraine, irrespective of individual perceptions of attack variation as long as they meet the currently validated IHS criteria along with previously defined experimental migraine criteria.15,34 In future studies, it would be relevant to address the discussion of whether the “mimics” criterion should in fact be considered crucial to the provoked attack definition, and maybe be an obligate criterion.

As per our current conviction, we are confident that the attacks reported in this study are indeed migraine attacks—although they may not be identical to the recollection of “usual” attacks of a numbered few patients. As such, we believe that our model correctly infers pathophysiological changes during attacks in a migraine population.

Limitations include that there was no placebo day for comparison, as well as a heterogenous patient group with different pain localization properties. A common limitation of MRA studies is the poor temporal resolution, where the proposed time-dependent dilatory responses are difficult to ascertain with just 2 measurements per study day. Magnetic resonance angiography studies with more sequential scans are time-consuming and expensive to conduct; however, our study suggests sequential investigations to be a necessity in the vascular study of migraine.

5. Conclusion

The novel findings of this study are the dilation of intradural MMA as well as MCA during migraine after both CGRP and sildenafil. Vasodilation as cause or consequence of migraine remains to be clarified, but we propose that the intradural vasculature is affected by activation of trigeminal afferents during unilateral migraine attacks. Furthermore, the effect that trigeminal nociceptors have on the vasculature during migraine seems to unreel in a time-dependent manner, causing dilation patterns to differ between onset and later phases of migraine.

Conflict of interest statement

C.E. Christensen has received personal fees for lecturing from Teva and serves as consultant for Teva. F.M. Amin has received personal fees and/or honoraria for lecturing from Teva, Eli Lilly, and Novartis. F.M. Amin is principal investigator for a Novartis Phase IV trial and member of advisory boards for Eli Lilly and Novartis. M. Ashina is a consultant or scientific advisor for Alder, Allergan, Amgen, Eli Lilly, Novartis, and Teva, and principal investigator for Alder, Amgen, ElectroCore, Novartis, and Teva trials. M. Ashina has no ownership interest and does not hold stock in any pharmaceutical company. M. Ashina serves as coeditor of the Journal of Headache and Pain, associate editor of Cephalalgia, and associate editor of Headache. The remaining authors have no conflicts of interest to declare.

Statistical analyses conducted by C.E. Christensen are assisted by the department of Biostatistics, University of Copenhagen.

Acknowledgements

The authors thank medical students Thomas Søborg, Nikolaj Toft, and Marius Lendal, laboratory technicians Lene Elkjær and Winnie Grønning, and Drs Anne Luise Vollesen and Nita Wienholtz for assistance with data collection, data handling, and drug administration.

The study was funded by grants from the Lundbeck foundation (R155-2014-171 and R249-2017-1608). The funding bodies had no influence on study design, conduct, participant inclusion, nor collection or interpretation of data or preparation, review, approval of, or decision to submit the manuscript.

Supplemental video content

Video content associated with this article can be found online at http://links.lww.com/PAIN/B128.

References

[1]. Amin FM, Asghar MS, Hougaard A, Hansen AE, Larsen VA, de Koning PJ, Larsson HB, Olesen J, Ashina M. Magnetic resonance angiography of intracranial and extracranial arteries in patients with spontaneous migraine without aura: a cross-sectional study. Lancet Neurol 2013;12:454–61.
[2]. Amin FM, Hougaard A, Schytz HW, Asghar MS, Lundholm E, Parvaiz AI, de Koning PJH, Andersen MR, Larsson HBW, Fahrenkrug J, Olesen J, Ashina M. Investigation of the pathophysiological mechanisms of migraine attacks induced by pituitary adenylate cyclase-activating polypeptide-38. Brain 2014;137:779–94.
[3]. Amin FM, Lundholm E, Hougaard A, Arngrim N, Wiinberg L, de Koning PJ, Larsson HB, Ashina M. Measurement precision and biological variation of cranial arteries using automated analysis of 3 T magnetic resonance angiography. J Headache Pain 2014;15:25.
[4]. Asghar MS, Hansen AE, Amin FM, van der Geest RJ, Koningvan der P, Larsson HBW, Olesen J, Ashina M. Evidence for a vascular factor in migraine. Ann Neurol 2011;69:635–45.
[5]. Ashina M, Hansen JM, á Dunga BO, Olesen J. Human models of migraine—short-term pain for long-term gain. Nat Rev Neurol 2017;13:713–24.
[6]. Ashina M, Hansen JM, Do TP, Melo-Carrillo A, Burstein R, Moskowitz MA. Migraine and the trigeminovascular system—40 years and counting. Lancet Neurol 2019;18:795–804.
[7]. Brain SD, Grant AD. Vascular actions of calcitonin gene-related peptide and adrenomedullin. Physiol Rev 2004;84:903–34.
[8]. Burstein R, Noseda R, Borsook D. Migraine: multiple processes, complex pathophysiology. J Neurosci 2015;35:6619–29.
[9]. Chakravarty A, Mukherjee A, Roy D. Migraine pain location in adult patients from eastern India. Ann Indian Acad Neurol 2008;11:98–102.
[10]. Christensen CE, Amin FM, Younis S, Lindberg U, de Koning P, Petersen ET, Paulson OB, Larsson HBW, Ashina M. Sildenafil and calcitonin gene-related peptide dilate intradural arteries: a 3T MR angiography study in healthy volunteers. Cephalalgia 2019;39:264–73.
[11]. Christensen CE, Younis S, Lindberg U, Boer VO, de Koning P, Petersen ET, Paulson OB, Larsson HBW, Amin FM, Ashina M. Ultra-high field MR angiography in human migraine models: a 3.0 T/7.0 T comparison study. J Headache Pain 2019;20:48.
[12]. Drummond PD, Lance JW. Extracranial vascular changes and the source of pain in migraine headache. Ann Neurol 1983;13:32–7.
[13]. Edvinsson L, Gulbenkian S, Barroso CP, Cunha e Sá M, Polak JM, Mortensen A, Jørgensen L, Jansen-Olesen I. Innervation of the human middle meningeal artery: immunohistochemistry, ultrastructure, and role of endothelium for vasomotility. Peptides 1998;19:1213–25.
[14]. Edvinsson L. The trigeminovascular pathway: role of CGRP and CGRP receptors in migraine. Headache J Head Face Pain 2017;57:47–55.
[15]. Guo S, Olesen J, Ashina M. Phosphodiesterase 3 inhibitor cilostazol induces migraine-like attacks via cyclic AMP increase. Brain 2014;137:2951–9.
[16]. Headache Classification Committee of the International Headache Society. The international classification of headache disorders, 3rd edition (beta version). Cephalalgia 2013;33:629–808.
[17]. Headache Classification Committee of the International Headache Society (IHS). The international classification of headache disorders, 3rd edition. Cephalalgia 2018;38:1–211.
[18]. Jackson G, Benjamin N, Jackson N, Allen MJ. Effects of sildenafil citrate on human hemodynamics. Am J Cardiol 1999;83:13–20.
[19]. Jacobs B, Dussor G. Neurovascular contributions to migraine: moving beyond vasodilation. Neuroscience 2016;338:130–44.
[20]. Khan S, Amin FM, Christensen CE, Ghanizada H, Younis S, Olinger ACR, de Koning PJH, Larsson HBW, Ashina M. Meningeal contribution to migraine pain: a magnetic resonance angiography study. Brain 2019;142:93–102.
[21]. van ’t Klooster R, de Koning PJH, Dehnavi RA, Tamsma JT, de Roos A, Reiber JHC, van der Geest RJ. Automatic lumen and outer wall segmentation of the carotid artery using deformable three-dimensional models in MR angiography and vessel wall images. J Magn Reson Imaging 2012;35:156–65.
[22]. Kruuse C, Thomsen LL, Jacobsen TB, Olesen J. The phosphodiesterase 5 inhibitor sildenafil has no effect on cerebral blood flow or blood velocity, but nevertheless induces headache in healthy subjects. J Cereb Blood Flow Metab 2002;22:1124–31.
[23]. Kruuse C, Thomsen LL, Birk S, Olesen J. Migraine can be induced by sildenafil without changes in middle cerebral artery diameter. Brain 2003;126:241–7.
[24]. Kurosawa M, Messlinger K, Pawlak M, Schmidt RF. Increase of meningeal blood flow after electrical stimulation of rat dura mater encephali: mediation by calcitonin gene-related peptide. Br J Pharmacol 1995;114:1397–402.
[25]. Linde M, Mellberg A, Dahlöf C. The natural course of migraine attacks. A prospective analysis of untreated attacks compared with attacks treated with a triptan. Cephalalgia 2006;26:712–21.
[26]. Liu Y, Broman J, Edvinsson L. Central projections of the sensory innervation of the rat middle meningeal artery. Brain Res 2008;1208:103–10.
[27]. Mason BN, Russo AF. Vascular contributions to migraine: time to revisit? Front Cel Neurosci 2018;12:233.
[28]. Nichols DJ, Muirhead GJ, Harness JA. Pharmacokinetics of sildenafil after single oral doses in healthy male subjects: absolute bioavailability, food effects and dose proportionality. Br J Clin Pharmacol 2002;53:5S–12S.
[29]. Rahmann A, Wienecke T, Hansen J, Fahrenkrug J, Olesen J, Ashina M. Vasoactive intestinal peptide causes marked cephalic vasodilation, but does not induce migraine. Cephalalgia 2008;28:226–36.
[30]. Rasmussen BK, Olesen J. Migraine with aura and migraine without aura: an epidemiological study. Cephalalgia 1992;12:221–8.
[31]. Russell M, Rasmussen B, Fenger K, Olesen J. Migraine without aura and migraine with aura are distinct clinical entities: a study of four hundred and eighty-four male and female migraineurs from the general population. Cephalalgia 1996;16:239–45.
[32]. Schoonman GG, Ferrari MD. Migraine headache is not associated with cerebral or meningeal vasodilatationa 3T magnetic resonance angiography study. Brain 2009;132:2192–200.
[33]. Schueler M, Neuhuber WL, De Col R, Messlinger K. Innervation of rat and human dura mater and pericranial tissues in the parieto-temporal region by meningeal afferents. Headache 2014;54:996–1009.
[34]. Schytz HW, Birk S, Wienecke T, Kruuse C, Olesen J, Ashina M. PACAP38 induces migraine-like attacks in patients with migraine without aura. Brain 2009;132:16–25.
[35]. Shayestagul NA, Christensen CE, Amin FM, Ashina S, Ashina M. Measurement of blood flow velocity in the middle cerebral artery during spontaneous migraine attacks: a systematic review. Headache J Head Face Pain 2017;57:852–61.
[36]. Viana M, Sances G, Ghiotto N, Guaschino E, Allena M, Nappi G, Goadsby PJ, Tassorelli C. Variability of the characteristics of a migraine attack within patients. Cephalalgia 2016;36:825–30.
Keywords:

Migraine; Headache; Calcitonin gene-related peptide; Sildenafil; Neuroimaging; MRI

Supplemental Digital Content

© 2020 International Association for the Study of Pain