Abstract
INTRODUCTION
T here are multidirectional interactions between migraine, vascular structure and function, and vascular disease. Migraine is associated with an increased risk for several vascular disorders, including ischaemic stroke, coronary artery disease and cervical arterial dissection (1–5). There is increased odds for livedo reticularis in migraineurs and migraine is a component of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (6, 7). Altered vascular structure and function in migraineurs may play a causal role in the pathophysiology of migraine. Several lines of evidence for endothelial dysfunction have been identified in migraineurs.
The vascular endothelium serves multiple functions, which when altered could contribute to migraine pathophysiology or which may serve as a common mechanistic pathway to migraine and vascular disease. The endothelium is involved in control of vascular tone, haemostasis, fibrinolysis, inflammation and vascular permeability. Endothelial cells contribute to the control of blood flow and blood pressure by secretion of vasodilators (nitric oxide and prostacyclin) and vasoconstrictors (endothelin and platelet activating factor) (8). Changes in vascular diameter are identified during migraine and abnormalities in control of vascular tone are hypothesized to play a role in migraine pathophysiology. The endothelium also serves a role in blood haemostasis and fibrinolysis (9). In its normal state the endothelium is antithrombotic, inhibiting clotting and platelet adhesion. However, it becomes prothrombotic in the presence of endothelial injury and inflammation. Endothelial dysfunction, resulting in alteration of the antithrombotic and prothrombotic balance, may contribute to the increased risk of cardiovascular and cerebrovascular disease in migraineurs.
This manuscript highlights some of the most important recent findings with regard to endothelial dysfunction and migraine and gives examples of several paths by which different types of endothelial dysfunction may associate with migraine. Abstracted manuscripts have investigated vascular wall structure, control of vascular tone, endothelial dependent haemostasis, endothelial maintenance, and the genetics of endothelial nitric oxide synthesis.
ARTERIAL STRUCTURE AND FUNCTION AND MIGRAINE
Yetkin E, Ozisik H, Ozcan C, Aksoy Y, Turhan H. Decreased endothelium-dependent vasodilatation in patients with migraine: a new aspect to vascular pathophysiology of migraine. Coronary Artery Disease 2006; 17:29–33
Objective:
To compare flow-mediated dilatation of the brachial artery in migraine without aura subjects with flow-mediated dilatation in non-migraine controls.
Methods:
Flow-mediated dilatation of the brachial arteries of 45 migraine without aura subjects was compared with age- and gender-matched controls. Migraineurs were studied when headache free. Arterial ultrasound examination was performed with subjects at rest and following 4.5 min of forearm tourniquet application at 300 mm Hg. Flow-mediated dilatation was determined by calculating the change in post-tourniquet diameter as a percentage of the baseline diameter.
Results:
Flow-mediated dilatation of the brachial artery in migraine without aura subjects was lower than that of controls (8.02 ± 4.095% vs. 10.72 ± 3.52%, P = 0.001).
Conclusions:
Migraine without aura subjects have a reduction in the normal vasodilatory response to shear stress, measured as flow-mediated dilatation, compared with non-migraineurs.
Yetkin E, Ozisik H, Ozcan C, Aksoy Y, Turhan H. Increased dilator response to nitrate and decreased flow-mediated dilatation in migraineurs. Headache 2007; 47:104–10
Objective:
To investigate endothelium-dependent and endothelium-independent vasodilatation in migraineurs.
Methods:
Flow-mediated and nitrate-mediated dilatation of the brachial artery was studied in 24 migraineurs and 26 controls matched for age and gender. Flow-mediated dilatation and nitrate dilatation were defined as the change in post-stimulus (forearm tourniquet and 0.4 mg sublingual nitroglycerin, respectively) diameter of the brachial artery as a percentage of the baseline diameter. All subjects were studied when headache free.
Results:
Flow-mediated dilatation in migraineurs was lower than in controls (7.6 ± 3.7% vs. 10.4 ± 3.5%, P = 0.008). Nitrate-mediated dilatation was greater in migraineurs than in controls (25% vs. 14%, P < 0.001).
Conclusions:
Migraineurs have increased nitrate-mediated brachial dilatation but decreased endothelium-dependent vasodilatation. Migraineurs may have a systemic abnormality in control of vascular tone.
COMMENTARY
These studies provide evidence for abnormal control of vascular tone in migraineurs. Flow-mediated dilatation is a measure of the blood vessels' ability to dilate in response to shear stress. Migraineurs have decreased dilatation of the brachial artery following release of tourniquet. This suggests that migraine is associated with systemic abnormalities in the control of vascular tone. Presence of this abnormality when headache-free serves as evidence that altered control of vascular tone is a persistent phenomenon, not one only associated with an acute migraine attack. However, these studies are unable to determine the direction of the relationship between migraine and dysregulation of vessel tone. Possibilities include: (i) repeated migraine attacks cause dysfunction in endothelial control of vascular tone; (i) dysfunctional endothelial control of vessel tone contributes to the production of migraine headaches; and (iii) migraine and dysfunctional control of vascular tone co-exist in a non-causal manner.
Vanmolkot FH, Van Bortel LM, de Hoon JN. Altered arterial function in migraine of recent onset. Neurology 2007; 68:1563–70
Objective:
To study arterial structure and function in subjects with recent onset of migraine.
Methods:
Fifty episodic migraine subjects with or without aura, 18–35 years of age, and without cardiovascular disease and cardiovascular risk factors, were enrolled. Subjects had migraine for the preceding 1–6 years. Migraine subjects were compared with 50 controls matched for age, gender, smoking status and hormonal contraceptive use. Measures of vascular structure and function included: brachial and aortic blood pressure (BP), structural and mechanical arterial wall properties of the common carotid artery, brachial artery and common femoral artery (diameter, distension, cross-sectional compliance, distensibility coefficient, and intima-media thickness), aortic augmentation index, aortic pulse wave velocity, and flow-mediated dilatation of the brachial artery. Migraine subjects were studied when headache free for at least 72 h and more than 24 h from their next migraine headache.
Results:
Two-thirds of the migraine subjects had at least occasional migraine with aura. Median headache frequency was 1.6 attacks per month, and mediantime since diagnosis of migraine was 2.4 years. Compared with controls migraineurs had higher brachial pressures (systolic BP 111 ± 8 mmHg vs. 108 ± 9 mmHg, P = 0.06; diastolic BP 67 ± 7 mmHg vs. 64 ± 5 mmHg, P = 0.005; mean arterial pressure 84 ± 7 mmHg vs. 81 ± 6 mmHg, P = 0.01). Migraineurs also had higher aortic pressures (systolic BP 96 ± 7 mmHg vs. 91 ± 7 mmHg, P = 0.004; diastolic BP 68 ± 7 mmHg vs. 64 ± 6 mmHg, P = 0.003; mean arterial pressure 81 ± 7 mmHg vs. 76 ± 6 mmHg, P = 0.002). Migraine subjects had smaller diameter brachial (4.82 ± 0.93 mm vs. 5.39 ± 0.89 mm, P = 0.01), common femoral (11.8 ± 1.9 mm vs. 12.7 ± 1.5 mm, P = 0.06) and superficial temporal arteries (1.94 ± 0.42 mm vs. 2.10 ± 0.39 mm, P = 0.052). The cross-sectional compliance of the brachial (0.30 ± 0.17 mm2/kPa vs. 0.37 ± 0.19 mm2/kPa, P = 0.02) and common femoral arteries (1.19 ± 0.55 mm2/kPa vs. 1.42 ± 0.59 mm2/kPa, P = 0.04) was less in the migraine group. Aortic augmentation index was increased in migraineurs (4 ± 10% vs. −1 ± 11%, P = 0.02). After adjustment for potential confounders, aortic augmentation index correlated independently with the presence of migraine (P = 0.04; model R = 0.66; F statistic = 14.2, P < 0.001). Flow-mediated dilatation of the brachial artery normalized to peak shear rate was lower in migraine subjects compared with controls (29 ± 15 vs. 37 ± 15 10−3% sec, P = 0.006).
Conclusions:
Migraine is associated with alterations in arterial structure and function. Identification of these alterations in young migraine subjects with a short history of migraine suggests that these alterations are not the result of repeated migraine attacks and may be related to the pathophysiology of migraine.
COMMENTARY
This study investigating subjects with recent onset of infrequent migraine helps to address, but does not conclusively answer, questions regarding the direction of the relationship between vascular abnormalities and migraine. Investigators had previously identified an interictal increase in arterial stiffness at the brachial artery in episodic migraineurs compared with non-migraine controls (10). Migraine subjects had smaller distension and decreased compliance. Investigators theorized that migraine is associated with increased arterial tone and/or altered arterial wall structure, resulting in increased arterial stiffness. However, the effect of repeated use of vasoactive migraine medications and repetitive attacks of migraine on arterial stiffness was queried. The study outlined above was performed in order to further delineate the direction of a possible cause-and-effect relationship between arterial alterations and migraine. Identification of vascular abnormalities in young migraine subjects with a short migraine history, infrequent attacks, and less frequent use of vasoactive medications, suggests that the arterial abnormalities are not the result of migraine, but are associated with migraine pathophysiology. However, longitudinal studies with repeated measurements over time would most adequately address the cause-and-effect question.
HAEMOSTASIS AND MIGRAINE
Tietjen GE, Al-Qasmi MM, Athanas K, Utley C, Herial NA. Altered haemostasis in migraineurs studied with a dynamic flow system. Thrombosis Research 2007; 119:217–22
Objective:
To evaluate haemostasis in migraine subjects, migraineurs with livedo reticularis, and to investigate the association of platelet haemostasis with von Willebrand factor.
Methods:
Fifty-four migraine subjects (19 episodic with aura, 20 episodic without aura, and 15 transformed migraine) and 30 controls were studied. Blood was drawn at least 7 days following the last migraine attack and at least 72 h after the last use of non-steroidal anti-inflammatory medication. Von Willebrand factor activity and antigen, platelet haemostasis time, clotting time and collagen-induced thrombus formation time were determined.
Results:
Twenty-one of the 54 migraine subjects had livedo reticularis. Compared with controls, episodic migraineurs had higher mean von Willebrand factor activity (142.7 ± 60.8 vs. 103.4 ± 35.7, P < 0.01) and antigen (132.1 ± 45.9 vs. 104.5 ± 41.4, P < 0.05). A similar pattern was detected when comparing migraine with aura with controls and migraine without aura with controls. Episodic migraine subjects had shorter platelet haemostasis time (3.8 ± 1.8 min vs. 4.6 ± 1.6 min, P = 0.03), clotting time (23.5 ± 4.4 min vs. 25.5 ± 4.1 min, P = 0.03) and collagen-induced thrombus formation time (4.6 ± 1.1 min vs. 5.1 ± 0.9 min, P = 0.04) compared with controls. No differences were found between transformed migraine subjects and controls. Increased von Willebrand factor activity (155 ± 59 vs. 103 ± 36, P = 0.001) and antigen (141 ± 43 vs. 104 ± 42, P = 0.001) were found in migraine subjects with livedo reticularis as compared with controls and between migraine subjects with livedo reticularis compared with migraine subjects without livedo reticularis (activity 155 ± 59 vs. 123 ± 49, P < 0.05;antigen 141 ± 43 vs. 117 ± 40, P < 0.05). There was decreased platelet haemostasis time in migraine subjects with livedo reticularis compared with controls (3.67 ± 1.6 min vs. 4.63 ± 1.6 min, P = 0.03) and an inverse correlation between von Willebrand factor activity and platelet haemostasis time (r =−0.51, P = 0.01) in migraine subjects with livedo reticularis.
Conclusions:
Migraine, especially migraine with livedo reticularis, is associated with endothelial dysfunction and hypercoagulability.
COMMENTARY
Tietjen and colleagues investigated haemostatic function in migraineurs with and without livedo reticularis and compared them with non-migraine controls. Prior studies have suggested associations between migraine, livedo reticularis, abnormal haemostatic function and stroke. The prevalence of livedo reticularis, a condition due to constriction of small and medium arteries at the dermis to subcutis border, is higher in those with migraine (odds ratio of 2.3; 95% confidence interval 1.08–4.71 in women) (6). Livedo reticularis is associated with increased odds for stroke, coronary artery disease and deep venous thrombosis (6, 11). Migraine is also frequently present in patients with Sneddon's syndrome, a disorder comprising livedo reticularis and ischaemic stroke (12). Given these associations, investigators correctly hypothesized that migraine subjects with livedo reticularis may have greater manifestations of endothelial dysfunction and altered haemostasis than migraine subjects without livedo reticularis.
The decreased platelet haemostasis time, clotting time and collagen-induced thrombus formation time identified in episodic migraineurs in this study suggest abnormalities of primary and secondary haemostasis. Furthermore, elevated von Willebrand factor serves as a marker of endothelial dysfunction in episodic migraineurs. Von Willebrand factor functions in initiating primary haemostasis by supporting platelet adhesion and aggregation and carries clotting factor VIII (9). Thus, increased von Willebrand factor activity is associated with greater degrees of haemostasis. Prior investigations have detected interictal elevations in von Willebrand factor antigen and activity in migraineurs with and without stroke compared with non-migraine controls (13). Von Willebrand factor antigen, but not activity, has been shown to increase further during a migraine attack compared with the interictal state (14).
In this study, migraineurs with livedo reticularis had more pronounced abnormalities of haemostasis than those without livedo reticularis. This supports the notion that migraine is associated with abnormalities in haemostasis and that the presence of livedo reticularis magnifies these abnormalities. These observations may contribute to the explanation for an increased risk of cardiovascular and cerebrovascular disease in migraine, a risk that is increased further in the presence of livedo reticularis.
ENDOTHELIAL MAINTENANCE AND MIGRAINE
Lee S-T, Chu K, Jung K-H, Kim D-H, Kim E-H, Choe VN, Kim J-H, Im W-S, Kang L, Park J-E, Park H-J, Park H-K, Song E-C, Lee S-K, Kim M, Roh J-K. Decreased number and function of endothelial progenitor cells in patients with migraine. Neurology 2008; 70:1510–17
Objective:
To compare the number and function of endothelial progenitor cells, biomarkers for vascular function and the risk of cardiovascular disease, in migraineurs to controls.
Methods:
Endothelial progenitor cell colony-forming unit levels, migratory capacity and senescence were evaluated in 67 migraineurs without aura, 25 migraineurs with aura, 74 subjects with tension-type headache, and 37 non-headache controls. Subjects were free from cardiovascular and cerebrovascular disease. Framingham risk factor scores were calculated for all subjects.
Results:
Consistent with prior literature, lower endothelial progenitor cell counts in controls were associated with higher Framingham risk scores (r =−0.475, P = 0.005). Migraineurs with aura had lower endothelial progenitor cell colony-forming unit counts (8.6 ± 10.1) than migraineurs without aura (20.4 ± 22.2, P = 0.001) and subjects with tension-type headache (47.8 ± 24.3, P < 0.001). Migraineurs without aura had lower endothelial progenitor cell colony counts than subjects with tension type headache (P < 0.001). Endothelial progenitor cell colony counts in subjects with tension-type headache did not differ from counts in non-headache controls (55.6 ± 23.0, P = 0.138). Endothelial progenitor cell migratory capacity was reduced (P < 0.05) and cellular senescence levels were increased (P < 0.05) in migraineurs compared with non-headache and tension-type headache controls.
Conclusions:
Migraine is associated with reductions in the number and function of endothelial progenitor cells, serving as a marker for dysfunctional endothelium. Endothelial dysfunction may explain the increased risk of cardiovascular disease and stroke in this patient group.
COMMENTARY
Investigators studied the concentration and function of endothelial progenitor cells in migraineurs. Endothelial progenitor cells normally assist in the maintenance of the endothelium by replacing injured cells. Endothelial progenitor cells serve as a marker for endothelial repair capacity, with the number of circulating endothelial progenitor cells being inversely correlated with the risk of cardiovascular disease and stroke (15). Based upon the observation that migraine is associated with an increased risk of cerebrovascular and cardiovascular disease, investigators correctly hypothesized that migraineurs have reduced numbers and function of endothelial progenitor cells. Impaired ability for endothelial repair in migraineurs may account for the increased risk of vascular diseases associated with migraine. At this time it is not clear if migraine leads to a reduced capacity for endothelial repair, if reduced ability for endothelial repair leads to migraine, or if the two co-exist in a non-causal manner (16).
NITRIC OXIDE SYNTHASE AND MIGRAINE
Toriello M, Oterino A, Pascual J, Castillo J, Colás R, Alonso-Arranz A, Ruiz-Alegría C, Quintela E, Montón F, Ruiz-Lavilla N. Lack of association of endothelial nitric oxide synthase polymorphisms and migraine. Headache 2008; 48:1115–9
Objective:
To investigate for an association between two endothelial nitric oxide synthase (eNOS) gene polymorphisms, Glu268Asp amino acid exchange (rs1799983) and −786T>C (rs1800779), and migraine.
Methods:
One hundred and forty-nine migraineurs without aura, 188 migraineurs with aura and 341 healthy controls were studied. Genotypic and allelic distributions were determined via real-time polymerase chain reaction (PCR) assay. The χ2 method was used to detect genotype distribution deviation from the Hardy-Weinberg equilibrium. Haplotype frequencies were estimated and individual odds ratios were calculated.
Results:
There were no significant differences in genotype or allele distribution for studied eNOS polymorphisms between all migraine subjects and controls. In addition, there were no differences between migraine with aura subjects and controls. No specific haplotype modified the risk for migraine.
Conclusions:
There is not a genetic association between the eNOS gene and migraine.
COMMENTARY
Nitric oxide serves several functions that may contribute to migraine. Nitric oxide mediates cerebral blood flow and arterial diameter, and is involved in nociceptive processing (17). Nitric oxide has been theorized to play an important role in the vasodilation detected during migraine headache. Platelet nitric oxide levels increase during the pain of migraine and nitric oxide increases calcitonin gene-related peptide promoter activity in trigeminal neurons (18, 19). The observation that nitrates can trigger migraine headaches in migraineurs and that blockade of nitric oxide synthase (NOS) is effective in aborting migraines further supports an association between nitric oxide function and migraine pathophysiology (20).
Nitric oxide is produced following oxidation of L-arginine, a process dependent upon isoforms of NOS, including endothelial NOS (eNOS) (21). Several common polymorphisms of the eNOS gene have been described. The Asp298 polymorphism investigated in the study described above had previously been found to be associated with an increased risk of cardiovascular and cerebrovascular diseases (22, 23). The Asp298 variant results in decreased eNOS activity (24). −786T>C causes reduced eNOS mRNA and decreased nitric oxide production, and is associated with elevated risk for cardiovascular disease (25).
A prior study of the eNOS Glu298Asp polymorphism in migraineurs found that subjects with eNOS AspAsp homozygosity (found in 14.7% of subjects) carried an increased risk of migraine with aura (26). Homozygotes had an odds ratio for migraine with aura of 3.02 compared with migraine without aura (95% confidence interval 1.21–7.51, P = 0.02), and a 2.21 odds ratio of migraine with aura compared with no migraine (95% confidence interval 1.00–5.04, P = 0.05). These data were suggestive of a role of nitric oxide in migraine pathophysiology and investigators theorized that eNOS AspAsp homozygosity may contribute to the explanation for an increased risk of vascular disease in migraine patients. However, results from the study abstracted above contradict these findings. Possible explanations for this discrepancy include the use of different statistical methods (e.g. correction for multiple comparisons) and study power. Of note, the study abstracted above used a sample sufficient to detect odds ratios of 1.7 or greater for migraine and odds ratios of 2.0 or greater for migraine with aura. Future studies will help to clarify whether eNOS gene polymorphisms serve as risk factors for migraine.