Sage Journals: Discover world-class research

Abstract

Aim

Migraine has been associated with stroke as well as with several non-atherosclerotic vascular conditions leading to discussions about the potential role of endothelium in the etiopathogenesis of migraine and migraine-associated stroke. We present a systematic review of the literature on vascular biomarkers in migraine, including those suggesting endothelial activation and damage.

Methods

We conducted a systematic literature search from 1990 to 2013 using multiple research databases with the keywords “migraine,” “headache,” “vascular,” and “biomarkers.” We used selected inclusion and exclusion criteria to create a final pool of studies for this review.

Results

The literature search identified a total of 639 citations of which 129 were included in our review. The final pool of clinical- and population-based studies assessed the level of various biomarkers (e.g. inflammatory, prothrombotic, endothelial activation, endothelial repair) in migraineurs of varying ages, gender, and demographic characteristics. Although for each biomarker there is at least one study suggesting an association with migraine, in many cases the quality of evidence is poor and there are conflicting studies showing no relationship. The results were, therefore, in each case inconclusive.

Conclusion

This systematic review indicated that in migraine populations there are a number of positive vascular biomarker studies, including some involving novel biomarkers such as endothelial microparticles and endothelial precursor cells. These lend insight into possible pathophysiological mechanisms by which migraine may be associated with stroke. More high-quality studies are needed to establish whether a true association between promising vascular biomarkers and migraine exists.

Keywords

Endothelium migraine biomarkers vascular markers headache systematic review

Introduction

Migraine is a prevalent neurovascular brain disorder, estimated to affect 13% of the population, with between 20% and 40% of them experiencing migraine with aura (MA) (1). Over the past two decades, MA has been increasingly recognized as an important risk factor for ischemic stroke, particularly in young women (2). More recently, the migraine-stroke link has been documented to be strongest in those without conventional cardiovascular risk profiles (3). The association of migraine with a number of non-atherosclerotic vascular conditions such as Raynaud’s phenomenon (4), variant angina (5), pre-eclampsia (6), and livedo reticularis (7), has, however, fueled speculation for a pivotal role of the endothelium in migraine pathogenesis and in migraine-associated stroke (8).

The endothelium is an active metabolic and endocrine organ that produces a wide variety of substances and expresses a multitude of receptors. When healthy, the endothelium is able to maintain a favorable balance between vasodilation and constriction, thrombolysis and thrombosis, and proliferation and apoptosis (9,10). Endothelial dysfunction (ED) is characterized by impaired vascular reactivity, as evidenced by decreased flow-mediated dilation (endothelial dependent) of the cerebral and systemic vasculature. Another component of ED is endothelial activation (or perturbation), a process brought about by inflammation that can lead to platelet activation, aggregation, coagulation, and clot formation, as well as suppression of anticoagulant substances (11). Indirect evidence of activation includes identification of circulating pro-inflammatory and prothrombotic biomarkers. A novel, more direct indicator of endothelial activation is the endothelial microparticle (EMP). Microparticles are submicron vesicles released from the cell membrane into the circulation in response to various stimuli, the nature of which (e.g. activation, injury, apoptosis) can be discerned from their protein composition (12). The quantity of circulating EMP correlates with the extent of ED (12,13). An additional biomarker of endothelial dysfunction is the endothelial precursor (or progenitor) cell, which is released from the bone marrow into the circulation in order to repair and regenerate damaged endothelium (14).

The objective of this review was to investigate within the migraine population whether there are serological biomarkers of vascular risk that may shed light on the pathogenesis of migraine or the mechanisms by which migraine is associated with stroke. Ideally, vascular biomarkers allow for identification of individuals at risk for stroke, aid in diagnosis of MA or of stroke, facilitate following disease activity, and measure response to treatment. The focus of this review was on indicators of endothelial activation, both indirect (biomarkers of inflammation, and coagulation) and direct (EMPs), as well as indicators of endothelial damage and repair, including endothelial precursor cells (EPCs).

Methods

We conducted a comprehensive review of the literature. A systematic review of PubMed, ERIC, Google Scholar, and Academic Search Premier databases was carried out to identify peer-reviewed, original scientific contributions looking at the association of migraine/headaches with vascular biomarkers. A Boolean search strategy was employed using the following keywords: “migraine,” “headache,” “biomarkers,” and “vascular.” The order of keywords was changed in repeated searches across databases to extract the final pool of relevant studies. An initial pool of 639 articles was identified. Further filtration using predetermined criteria was conducted to include a final list of 129 studies. The inclusion criteria were: studies published in English, original clinical- or population-based studies with primary data, and published between 1990 and 2014 (older articles were used to provide a historical context or when recent literature was not available). Our decision to set the earlier limit at 1990 was based on the fact that a uniform definition of migraine was developed by the International Headache Society and published in 1988. Studies published in other languages, review articles, and commentaries/editorials were eliminated. References from articles selected for this review were also cross-checked. Two independent investigators (GT, JK) reviewed the final pool of studies to reach a consensus on inclusion of relevant articles. Discrepancies were sorted out with discussion.

Results

In this review of the relationship of migraine and vascular biomarkers we examined 18 studies of lipids, 13 of C-reactive protein (CRP), 12 of platelets (all but three of which were from before 1990), 10 of tumor necrosis factor alpha (TNF-α), eight of endothelin-1 (ET-1), eight of homocysteine, six of interleukin-6 (IL-6), five of von Willebrand factor (vWF) and of fibrinogen, four of interleukin-1 (IL-1), interleukin-10 (IL-10), and adiponectin, three of endothelial precursor cells, three of transforming growth factor beta 1 (TGF-β1), and two of tissue plasminogen activator (tPA), asymmetric dimethylarginine (ADMA), and one newly published study of EMPs.

Biomarkers of inflammation

CRP

CRP is an acute-phase reactant, which is predominantly released from the liver and adipose tissues in response to inflammatory stress (15). It is involved in host immunity, including activation of the complement system, and enhancement of the endothelial expression of matrix metalloproteinases (16). In addition, by its interaction with Fc receptors, CRP produces pro-inflammatory cytokines (17). Elevated in a variety of disease states, CRP has also correlated with the risk of developing cardiovascular disease (18 –20). A number of studies have looked at CRP in migraine, with conflicting results (Table 1). In four case-control studies, elevated CRP was associated with migraine (21 –24). In one study the subtype with higher CRP levels was migraine without aura (MO) (21), in two others, MA (22,23), and in the fourth, neither (24). In the general population-based Cerebral Abnormalities in Migraine, an Epidemiological Risk Analysis (CAMERA) study, CRP was elevated in migraine, particularly in women and in those with aura (25). Furthermore, levels of CRP correlated with aura frequency and duration. Large population-based studies have also demonstrated that CRP is elevated in patients with headache in children and adolescents (26) and in older individuals (27).Three additional case-control studies found increased CRP levels in migraineurs compared to controls. Differences were not significant, although the small sample sizes could be a potential limitation (28 –30). Two other small studies found no significant relationship between CRP and migraine (31,32). In the very large population-based Reykjavik study, however, CRP levels were not increased among migraine sufferers compared with non-migraineurs (33), although there was a trend (p = 0.08) for an increase in young adult women with MO.

Table 1.

Association of high-sensitivity C-reactive protein with migraine.

Study author/year	Sample size/study design	Sample characteristics	Observations
Vanmolkot and de Hoon, 2007 (21)	50 migraineurs +50 controls. Case-control study.	Migraineurs (32 MA, 18 MO), 78% female, matched to controls for gender, age, smoking status, and current hormonal contraceptive use. Inclusion criteria: 18–35 yo, BMI 18–30 kg/m², diagnosis by ICHD criteria. Exclusion: CV disease, hypertension, diabetes, hypercholesterolemia, pregnancy or lactation, and daily use of drugs.	Compared to controls: Higher CRP in migraine; adjusted p = 0.045. Higher CRP in MO; adjusted p = 0.003.
Tietjen et al., 2009 (22)	125 migraineurs +50 controls. Case-control study.	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education, and income. Mean frequency 12 ± 9 days/month. Mean age 37 ± 9 years. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: Higher hs-CRP in migraine, OR 4.71 (1.48–14.97). Higher hs-CRP in MA, OR 7.99 (2.32–27.61). Higher hs-CRP in MO, OR 2.63, (0.73–9.45).
Güzel et al., 2013 (23)	51 migraineurs +27 controls. Case-control study.	Migraineurs (27 MA, 24 MO) matched to controls for demographic characteristics, marital status, and socioeconomic status. ROC curve results used.	Compared to controls: Higher hs-CRP in migraineurs, p < 0.05.
Hamed et al., 2010 (24)	63 headache patients +35 controls. Case-control study.	Adults with headache (14 MA, 24 MO). Inclusion: 18–40 yo; diagnosis by ICHD criteria. Exclusion: Smoking, alcoholism, pregnancy, diabetes, medication use, and chronic illness.	Compared to controls: Higher hs-CRP in migraine, p = 0.01. No difference in the levels between MA and MO.
Tietjen et al., 2011 (25)	283 migraineurs +134 controls. Population study (CAMERA study).	Migraineurs (155 MA, 128 MO), similar to controls for age, education, hypertension, tobacco, diabetes, cholesterol, or contraceptive use. Inclusion: 30–60 yo, diagnosis by ICHD criteria.	Compared to controls: Higher hs-CRP in migraine, p = 0.03. Higher hs-CPR in MA, p = 0.02. Not higher in MO.
Nelson et al., 2010 (26)	2295 headache cases +9475 controls. Population study (NHANES).	Children (4–11 yo) and adolescents (12–19 yo) with severe headache. Inclusion: 4–19 yo; diagnosis of headache by self-report.	Compared to non-headache controls: Higher hs-CRP in headache, and in girls vs boys. More children with headache were in the top quintile for CRP value (OR 1.51, CI = 1.24–1.83).
Kurth et al., 2008 (27)	5087 migraineurs +22539 controls. Population study (Women’s Health Study).	Migraineurs (70.5% with active migraine, of whom 39.7% have MA). Inclusion: Women > 45 yo; headaches diagnosed by validated questionnaire.	Compared to women with no migraine: Higher odds for elevated hs-CRP for any migraine, active migraine (OR 1.14, 95% CI 1.02, 1.27), and prior migraine.
Guldiken et al., 2011 (28)	50 migraineurs +25 controls. Case-control study.	Migraineurs (23 MA, 27 MO), mean age in years 38.0 ± 8.6. Inclusion: Female; diagnosis by ICHD criteria. Exclusion: Pregnancy, disorders, chronic disease, drugs affecting prolactin prolactin levels, and chronic diseases.	Compared to controls: No difference in hs-CRP in migraine. No difference in hs-CRP between MA and MO.
Guldiken et al., 2008 (29)	60 migraineurs +25 controls. Case-control study.	Migraineurs (46 females), mean age 38.3 ± 10.2 years. Inclusion: Diagnosis by ICHD criteria. Exclusion: CV or metabolic diseases, medications altering glucose metabolism.	Compared to controls: No difference in hs-CRP in migraine.
Bernecker et al., 2011 (30)	48 migraineurs +48 controls. Case-control study.	Migraineurs, mean age 37 ± 11 years. Inclusion: All Caucasian females; diagnosis by ICHD criteria. Exclusion: Diabetes, depression, thyroidal dysfunction, metabolic and CV diseases, chronic illnesses, infection, pregnancy and nursing.	Compared to controls: No difference in hs-CRP in migraine.
Gruber et al., 2010 (31)	60 migraineurs +76 controls. Case-control study.	Migraineurs (43 normal weight, 17 obese); controls (61 normal weight, 15 obese), matched for age. Inclusion: Diagnosis by ICHD criteria.	Compared to normal-weight controls: No difference in hs-CRP in normal- weight migraine. Higher hs-CRP in obese migraineurs (p = 0.02).
Silva et al., 2007 (32)	50 migraineurs +25 controls. Case-control study.	Migraineurs (25 MA and 25 MO, 44 female), matched for age and gender. Inclusion: Diagnosis by ICHD criteria.	Compared to controls: No difference in hs-CRP in migraine. No difference in hs-CRP between MA and MO.
Gudmundsson et al., 2009 (33)	272 migraineurs +6653 controls. Population-based cohort study.	Migraineurs matched for age and sex. Inclusion: Diagnosis by ICHD criteria. Exclusion: Without MI. Multivariate adjustments: Age, BMI, cholesterol, smoking status, education, current hormone use, DM, systolic BP, antihypertensive therapy.	Compared to controls: No difference in hs-CRP in migraine. Borderline higher hs-CRP in migraine for women with MO, 19–34 yo (p = 0.08). Lower hs-CRP in migraine for women, 60–81 yo (p = 0.007).

MA: migraine with aura; MO: migraine without aura; BMI: body mass index; hs-CRP: high-sensitivity C-reactive protein; ICHD: International Classification of Headache Disorders; CV: cardiovascular; MI: myocardial infarction; DM: diabetes mellitus; BP: blood pressure; yo: years old; OR: odds ratio; CI: confidence interval; NHANES: National Health and Nutrition Examination Survey; CAMERA: Cerebral Abnormalities in Migraine, an Epidemiological Risk Analysis; ROC: receiver-operating characteristic.

Cytokines

Cytokines are small proteins, produced by most cells in the body, that possess multiple biologic activities that promote cell-cell interaction. Neuropeptides, such as calcitonin gene-related peptide (CGRP), involved in migraine can stimulate the release of pro-inflammatory cytokines, some of which are established to contribute to both the risk of and the outcome from cerebrovascular disease (34). In addition, cytokines have been found to have pain-mediating functions as well as immunological functions (34 –36). Understandably, they have been the subject of numerous clinical studies in migraine (Table 2).

Table 2.

Cytokines in migraine.

Study author/year	Sample size/study design	Sample characteristics	Observations
Tumor necrosis factor alpha (TNF-α)
Tietjen et al., 2010 (37)	125 migraineurs +50 controls. Case-control study.	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education, and income. Mean frequency 12 ± 9 days/month. Mean age 37 ± 9 years. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: Higher TNF-α in migraine (OR 11.87; 3.50–30.20), correlated with headache frequency.
Perini et al., 2005 (38)	25 migraineurs +18 controls. Case-control study.	25 migraine patients (22 females, 4 MA), during and outside attacks. IHS criteria for diagnosis. Exclusion: Preventive meds, inflammatory, infectious, immune disease, drug use.	Compared to controls: Higher TNF-α levels during migraine attack. Lower TNF-α levels outside attack.
Fidan et al., 2006 (39)	25 migraineurs +25 controls. Case-control study.	Migraineurs (7 MA, 18 MO) were assessed during and between headaches, matched for age (mean = 40 years) and gender (80% female). IHS criteria used for diagnosis.	Compared to controls: No difference in TNF-α in migraine. No difference during and between attacks.
Uzar et al., 2011 (40)	64 migraineurs +34 controls. Case-control study.	No difference between migraine and controls for BMI, BP, age, sex, serum glucose and lipids. ICHD criteria were used for diagnosis.	Compared to controls: No difference in TNF-α levels in migraine or comparing ictal to interictal values.
Gergont et al., 2005 (41)	61 headache patients +28 controls. Case-control study.	Participants included children (mean age 13.5 years) with migraine (16 MA, 14 MO) and TTH. IHS criteria used for diagnosis.	Compared to controls: No differences in migraine, ETTH.
Sarchielli et al., 2006 (42)	7 MO patients. Observational study.	3 males and 4 females, mean age 33.3 ± 7.4 years, duration 9.7 ± 3.4 years. ICHD criteria used for diagnosis. Exclusion: Smoking; renal and CV diseases, hypertension, epilepsy, pregnancy.	Transitory increases in the levels of TNF-α in the internal jugular blood of MO patients during the early hours of migraine attacks (peaking at 1 hour and declining after 6 hours).
van Hilten et al., 1991 (43)	20 migraineurs. Observational study.	Migraineurs (6 MA, 14 MO), mean age of 45 years, range 27–56 years. Ictal-interictal study.	Compared to interictal values: No differences in TNF-α levels during migraine attack.
Rozen and Swidan, 2007 (44)	38 headache patients. Observational study.	20 new daily persistent headache (NDPH), 16 chronic migraine, 2 post-traumatic.	Elevated CSF TNF-α levels: 19 of 20 NDPH patients. 16 of 16 chronic migraine patients.
Bernecker et al., 2011 (30)	48 migraineurs + 48 controls. Case-control study.	Migraineurs, mean age 37 ± 11 yrs. Inclusion: All Caucasian females; diagnosis by ICHD criteria Exclusion: Diabetes, depression, thyroidal dysfunction, metabolic and cardiovascular diseases, chronic illnesses, infection, pregnancy and nursing.	Compared to controls: No difference in TNF-α in migraine.
Bø et al., 2009 (45)	87 headache patients +20 controls. Case-control study.	39 ETTH (21 females), 34 MO (22 females), 24 MA (21 females), 10 cervicogenic headache (5 females). Age range 16–77 years. ICHD criteria used for diagnosis.	Elevated CSF TNF-α levels: 1 of 20 controls, 2 of 10 cervicogenic headache, 4 of 30 ETTH, 3 of 25 MO, 7 of 20 MA.
Interleukin-6 (IL-6)
Kocer et al., 2010 (35)	42 TTH patients and 37 controls.	Headache patients: 20 ETTH, 22 CTTH. ICHD criteria used for diagnosis.	Compared to controls: Higher IL-6 level in headache. No differences between ETTH and CTTH.
Tietjen et al., 2010 (37)	125 migraineurs +50 controls. Case-control study.	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education and income. Mean frequency 12 ± 9 days/month. Mean age 37 ± 9 years. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: Higher likelihood if elevated IL-6 level in migraine (OR 5.03; 95% CI, 2.01–12.70).
Fidan et al., 2006 (39)	25 migraineurs +25 controls. Case-control study.	Migraineurs (7 MA, 18 MO) were assessed during and between headaches, matched for age (mean = 40 years) and gender (80% female). IHS criteria used for diagnosis.	Compared to controls: Higher IL-6 levels during and between migraine attack. Compared to interictal period: Higher IL-6 levels during migraine attack.
Uzar et al., 2011 (40)	64 migraineurs +34 controls. Case-control study.	No difference between migraine and controls for BMI, BP, age, sex, serum glucose, and lipids. ICHD criteria were used for diagnosis.	Compared to controls: Higher IL-6 levels in migraine (3.1 ±1.44 vs. 2.40 ±0.22 pg/ml, p = 0.007).
Perini et al., 2005 (38)	25 migraineurs +18 controls. Case-control study.	25 migraine patients (22 females, 4 MA), during and outside attacks. IHS criteria for diagnosis. Exclusion: Preventive meds, inflammatory, infectious, immune disease, drug use.	Compared to controls: No differences in IL-6 in migraine. Compared to interictal period: No differences in IL-6 during attack.
Sarchielli et al., 2006 (42)	7 MO patients. Observational study.	3 males and 4 females, mean age 33.3 ± 7.4 years, duration 9.7 ± 3.4 years. ICHD criteria used for diagnosis. Exclusion: Smoking, renal and CV diseases, hypertension, epilepsy, pregnancy.	Transitory increases in the levels of IL-6 in the internal jugular blood of MO patients during the early hours of migraine attacks as compared to later stages (p < 0.003, peaking at 1 hour and declining after 6 hours).
Interleukin-1 (IL-1)
Uzar et al., 2011 (40)	64 migraineurs +34 controls. Case-control study.	No difference between migraine and controls for BMI, BP, age, sex, serum glucose and lipids. ICHD criteria were used for diagnosis.	Compared to controls: Higher IL-1 levels in migraine (5.73 ±1.44 vs. 4.90 ±1.40 pg/ml, p = 0.006).
Perini et al., 2005 (38)	25 migraineurs +18 controls. Case-control study.	25 migraine patients (22 females, 4 MA), during and outside attacks. IHS criteria for diagnosis. Exclusion: Preventive meds, inflammatory, infectious, immune disease, drug use.	Compared to interictal period: Higher IL-1 levels during attack.
Fidan et al., 2006 (39)	25 migraineurs +25 controls. Case-control study.	Migraineurs (7 MA, 18 MO) were assessed during and between headaches, matched for age (mean = 40 years) and gender (80% female). IHS criteria used for diagnosis.	Compared to controls: No difference in IL-1 levels in migraine. Compared to interictal period: No difference in Il-1 levels during attacks.
van Hilten et al., 1991 (43)	20 migraineurs. Observational study.	Migraineurs (6 MA, 14 MO), mean age of 45 years, range 27–56 years. Ictal-interictal study.	Compared to interictal period: No difference in IL-1 levels during attacks.
Interleukin-10 (IL-10)
Perini et al., 2005 (38)	25 migraineurs +18 controls. Case-control study.	25 migraine patients (22 females, four MA), during and outside attacks. IHS criteria for diagnosis. Exclusion: Preventive meds, inflammatory, infectious, immune disease, drug use.	Compared to controls: Higher IL-10 levels during migraine attack compared to interictal period: Higher IL-10 levels during migraine attack.
Fidan et al., 2006 (39)	25 migraineurs +25 controls. Case-control study.	Migraineurs (7 MA, 18 MO) were assessed during and between headaches, matched for age (mean = 40 years) and gender (80% female). IHS criteria used for diagnosis.	Compared to controls: No differences in IL-10 levels in migraine. In migraineurs: IL-10 levels increased during attacks (p < 0.05).
Munno et al., 2001 (47)	23 migraineurs +23 controls. Case-control study.	23 MO patients (14 females) for blood sampling during active headache periods. Patients age from 19–58 years (mean age of 38.5 years). Controls matched for age and sex. Patients with allergies excluded. IHS criteria used for diagnosis. Diagnosis was made according to IHS criteria.	Compared to controls: Higher IL-10 levels in migraine (p < 0.01). After sumatriptan given to 10 patients: IL-10 decreased in 2 patients, undetectable in the others.
Uzar et al., 2011 (40)	64 migraineurs +34 controls. Case-control study.	Mean age of the migraine patients was 35.4 ± 11.5 years (17–56 years). Study group consisted of 45 women and 19 men. The Control group comprised of 24 women and 10 men. Mean age of the control group was 34.7 ± 11.7 years (17–55 years). No difference was seen in groups based on BMI, BP, age, sex, serum glucose and lipids. ICHD criteria were used for diagnosis.	Compared to controls: Lower levels of IL-10 in migraine (p = 0.007).
Transforming growth factor-β1 (TGF-β1beta)
Transforming growth factor-β1 (TGF-β1)
Tietjen et al., 2010 (37)	125 migraineurs +50 controls. Case-control study.	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education, and income. Mean frequency 12 ± 9 days/month. Mean age 37 ± 9 years. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: Higher TGF-β1 in migraine (OR 12.06; CI 8.01, 20.70). Correlated with headache frequency p < 0.001).
Ishizaki et al., 2005 (48)	80 headache patients +58 controls. Case-control study.	Headache patients (68 migraineurs and 12 ETTH). IHS criteria used for diagnosis. Migraineurs (23 MA, 45 MO). Mean monthly frequencies: 2.3 MA, 4.0 MO. Exclusion: Lupus or sarcoidosis.	Compared to controls: Higher TGF-β1 levels in migraine. No differences between MA and MO. No correlatetion with headache frequency or duration of illness.
Güzel et al., 2013 (23)	51 migraineurs + 27 controls. Case-control study.	Migraineurs (27 MA, 24 MO) matched to controls for demographic characteristics, marital status, and socioeconomic status. ROC curve results used.	Compared to controls: Higher TGF-β1 in migraineurs, p < 0.05. Significant difference between MA > MO.

MA: migraine with aura; MO: migraine without aura; BMI: body mass index; hs-CRP: high-sensitivity C-reactive protein; ICHD: International Classification of Headache Disorders; IHS: International Headache Society; CV: cardiovascular; MI: myocardial infarction; CTTH: chronic tension-type headache; ETTH: episodic tension-type headache; NDPH: new daily persistent headache; DM: diabetes mellitus; CSF: cerebrospinal fluid; BP: blood pressure; yo: years old; OR: odds ratio; CI: confidence interval.

TNF-α

TNF-α is a pro-inflammatory cytokine with a primary role in the regulation of immune cells, while also being involved in coagulation, lipid metabolism, cell proliferation, and apoptosis.

Interictal studies

TNF-α was elevated in premenopausal female migraineurs compared to controls matched for sex, age, and vascular risk factors (37). In the migraine cohort, the levels correlated with headache frequency, body mass index (BMI), and high-sensitivity CRP. These results are in contradiction to negative, smaller case-control studies in adults (30,38 –40) and children (41).

Ictal studies

Some studies (38,42), but not all (39,43), have also demonstrated an ictal increase in TNF-α levels compared to the baseline state. This putative ictal rise implicates a role for inflammation in the acute migraine attack, with cytokine release secondary to the release of sensory neuropeptides from activated trigeminal endings. An uncontrolled study in chronic daily headache sufferers found elevated cerebrospinal fluid (CSF) TNF-a levels, mostly in conjunction with normal serum TNF-α levels (44). A different study of individuals with migraine or tension-type headache (TTH) compared to headache-free controls, showed elevated ictal CSF TNF-α levels compared to the levels in the control group (45).

IL-6

IL-6 is a cytokine produced at the site of inflammation or tissue damage and a major inducer of the acute phase reaction. It also plays a role in the transition from acute to chronic inflammation (30).

Interictal studies

The levels were elevated in four migraine interictal case-control studies (35,37,39,40), including one of young women, in whom the IL-6 levels correlated with headache frequency, BMI, and high-sensitivity CRP (37).

Ictal studies

A transient increase of IL-6 from internal jugular blood samples was observed during the first two hours after MO attack onset, compared with the time of catheter insertion (42).

Interleukin-1β (IL-1β)

IL-1β is an important mediator of response to infection, inflammation, and immunologic conditions. IL-1β, secreted by monocytes and macrophages is the main source of circulating IL-1β.

Interictal studies

Migraine patients had a higher serum IL-1β levels than healthy subjects in one study (40), but not another (39).

Ictal studies

Circulating levels of IL-1β during migraine attacks were significantly higher in comparison to their levels outside attacks during one study (38), but were not elevated in two other studies (39,43).

IL-10

IL-10 is an anti-inflammatory cytokine and works via inhibiting synthesis of pro-inflammatory cytokines, such as IL-1β, IL-6, and TNF-α (46).

Interictal

In one case-control study, migraine patients had lower interictal IL-10 levels than healthy controls (38), and in another there was no difference (39).

Ictal

In two ictal interictal studies of migraineurs, circulating levels during attacks were higher than levels outside attacks (38,47).

Other interleukins

Other families of interleukins (e.g. IL-2, IL-4, and IL-5) have also been studied in relation to migraine etiopathogenesis. IL-2, IL-4, and IL-5 are responsible for B and T immune cell activation and proliferation and have been found to be lower in migraineurs during ictal measurements (36,39,47).

TGF-β1

TGF-β1 is a multifunctional pro-inflammatory cytokine that controls cell proliferation, motility, differentiation, and apoptosis. It plays a role in blood vessel formation and immune system function.

Interictal studies

Case-control evaluations of TGF-β1in migraine showed elevation compared to controls in three studies (37,23,48). In one of the studies, the cytokine levels in the cases (premenopausal women with migraine) correlated with headache frequency (37). In a second study there were differences between patients with MA compared to those with MO (23). In the third study, TGF-β1 levels did not correlate with headache frequency, nor were there differences in levels between individuals with MA and MO (48). Lending support to the associations between TGF-β1 and migraine is the finding that genetic variants in the TGF-β1 inflammatory pathway also appear to be associated with migraine (49).

Adiponectin

Adiponectin is an anti-inflammatory cytokine (adipocytokine) secreted by the adipose tissue, which is involved in regulation of glucose homeostasis and other metabolic processes, such as diabetes mellitus type 2, cardiovascular disease, and neurodegenerative disorders. Adiponectin, which is inversely correlated with obesity and BMI, reduces expression of pro-inflammatory cytokines, such as TNF-a, and in brain endothelial cells, IL-6 (50). In headache, there have been three cross-sectional studies of adiponectin (37,51,52). In the initial study, after adjustment for waist-to-hip ratio and BMI, the concentration of three subtypes of adiponectin (total adiponectin, high-molecular weight (HMW), middle-molecular weight (MMW)) was higher in chronic daily headache (CDH) sufferers than in those with episodic migraine (EM) and in the controls (51). The second study was a case-control investigation restricted to premenopausal women with migraine and matched controls (37). In both adjusted and unadjusted analyses, there were no significant difference in total adiponectin concentration between migraineurs and controls. Similarly, a third study including men and women with (n = 50) and without (n = 74) migraine, also showed no significant difference in total adiponectin levels (52). In a study of individuals with EM treated acutely with Treximet® (vs placebo), total adiponectin concentrations were decreased at 30, 60 and 120 minutes compared to the value at onset, in both adjusted and unadjusted analyses (53). In non-responders total adiponectin (unadjusted and adjusted) levels did not significantly change over time. In all participants, increases in the high- to low-molecular weight (HMW:LMW) adiponectin ratio were associated with an increase in pain severity, and this ratio increased in non-responders and decreased in responders after treatment as compared to onset in adjusted analyses.

Lipids

The biological link between hyperlipidemia and migraine, and the implications for the relationship of migraine and stroke, has been an area of considerable research interest (Table 3). Elevated cholesterol is a risk factor for vascular disease and there is basic science evidence that cortical spreading depression (CSD), the putative correlate of MA, can be triggered by a bolus of cholesterol crystals in mice (54). Multiple studies, many of them large and population based, and spanning a range of ages, have demonstrated an elevation of total cholesterol in individuals with migraine compared to controls (3,27,31,55 –60). In some series, including the Epidemiology of Vascular Ageing (EVA) study and the Genetic Epidemiology of Migraine (GEM) study, risk of elevated total cholesterol was exclusively found in the MA subgroup (55,56). In the American Migraine Prevalence and Prevention (AMPP) study, the risk of elevated total cholesterol was highest in MA, but was also elevated in MO compared to controls (57). In contrast, the Women’s Health Study (WHS) demonstrated an increase of total cholesterol in migraine, but levels did not differ based on aura status (27,59). In addition to total cholesterol, other lipid subtypes have been associated with migraine, including elevations of triglycerides, low-density lipoprotein cholesterol (LDL-c), oxidized LDL-c (the most atherogenic form), as well as decreases in anti-inflammatory high-density lipoprotein cholesterol (HDL)-c (29 –31,52,55,59 –62). Of note, several studies found no significant differences in lipid profiles of migraineurs and controls, including studies of children and of the elderly (21,24,26,29,32,63 –67).

Table 3.

Association of migraine with lipids.

Study author/year	Sample size/study design	Sample characteristics	Observations
Rist et al., 2011 (55)	230 headache patients +925 controls. Observational study.	Headache (64 non-migraine headache, 166 had migraine, including 23 MA). Average age 69 years. Headaches diagnosed by ICHD criteria.	Compared to lowest tertile of cholesterol: Increased risk of MA in second (OR 4.67; 0.99, 21.97) and third (OR 5.97; 1.29–27.61) tertiles of TC. Elevated triglycerides increased risk of MA (OR 4.42; 1.32–14.77).
Scher et al., 2005 (56)	620 migraineurs +5135 controls. Population study.	Migraineurs (31% MA, 64% MO, and 5% unclassified). Diagnosed by ICHD criteria.	Compared to controls: Higher TC in MA (OR 1.43; 0.97–2.1). Higher TC: HDL ratio OR 1.64; 1.1– 2.4).
Bigal et al., 2010 (57)	6102 migraineurs +5243 controls. Population study.	Headache diagnosis using a validated mailed questionnaire. Migraineurs were more likely to be women (80.3% vs 53%, p < 0.001) and have had lower incomes.	Compared to controls: Increased odds of high cholesterol in migraine (32.7% vs 25.6%, OR 1.4, 95% CI 1.3–1.5). Risk was highest in MA, but also elevated in MO.
Gilad et al., 2014 (58)	163 migraineurs. Observational study.	Migraineurs (0% MA; 145 females, mean age 59.4 ± 6.6 years, range 55–88 years).	Compared to controls: Higher total cholesterol (TC) in MA (OR 1.43; 0.97–2.1). Higher TC: HDL ratio (OR 1.64; 1.1–2.4). 48% MA patients had high lipids compared to 23% MO patients (p < 0.001).
Kurth et al., 2008 (27)	5087 migraineurs +22,539 controls. Population study (Women’s Health Study).	Migraineurs (70.5% with active migraine, of whom 39.7% have MA). Inclusion: Women >45 yo; headaches. Diagnosed by validated questionnaire.	Compared with no migraine history: Higher odds for elevated total cholesterol (OR 1.09; 1.01, 1.18). Higher odds for elevated non-HDL-c (OR 1.14; 1.05–1.23), for women with any migraine. No differences based on MA status and migraine frequency.
Kurth et al., 2008 (3)	27,519 participants. Women’s Health Study.	3577 (13.0%) active migraine, of whom 1418 (39.6%) reported MA. ICHD-II criteria used for diagnosis. Exclusion: Non-Caucasian, CVD, angina, cancer, and chronic illness.	Compared to controls (TC = 5.46; HDL-c = 1.39): Higher mean total cholesterol in active MA (5.52), active MO (5.51), and prior migraine (5.56) (p < 0.001). Lower mean HDL-c in active MA (1.37), active migraine MO (1.38), and prior migraine (1.36) (p < 0.001).
Monastero et al., 2008 (59)	151 migraineurs +1658 controls. Population study.	Migraineurs were younger (mean ± SD: 68.3 ± 8.6 vs. 73.4 ± 10.1, p = 0.002), and high % females (79.5% vs. 52.9%, p < 0.0001). Diagnosed by ICHD criteria. Exclusion: Dementia, stroke, mental retardation, and severe psychosis.	Compared to controls: Higher levels of TC (223.4±39.8 vs. 212.6 ± 40.7, p = 0.003) and LDL-c (146.6 ± 36.2 vs. 136.8 ± 37.9, p = 0.004) in migraine, higher odds (OR 1.6; 1.1, 2.3) for elevated TC in migraine, especially in elderly males (OR = 3.8; 1.4, 9.9).
Gruber et al., 2010 (31)	60 migraineurs +76 controls. Case-control study.	Normal-weight controls (61, 42 females), normal-weight migraineurs (43, 38 females). Obese controls (15, 6 females), obese migraineurs (17, 11 females). Diagnosed by ICHD criteria.	Compared to normal-weight controls: Increased cholesterol, LDL-c and oxidized LDL-c (p < 0.05) in normal-weight migraineurs. Highest vs. lowest quartile of LDL-c, OR for migraine 7.93 (95% CI: 2.23–28.15; p = 0.001).
Saberi et al., 2011 (60)	102 migraineurs +103 controls. Case-control study.	Migraineurs (84 women, 18 men); mean age 34.9 ± 11.8 years. Controls (79 women, 24 men); mean age 32.8 ± 5.7 years). Diagnosed by ICHD criteria. Multivariate analysis.	Compared to controls: Higher odds of hypertriglyceridemia (OR 3.8; CI: 1.8–7.9), hyper-cholesterolemia (OR 5.0; CI 2.2–11.4), and low HDL-c (OR 0.28; CI 0.13–0.6) in migraine.
Winsvold et al., 2011 (61)	20,768 headache patients +27,945 controls. Population study.	Total of 48,713 subjects (age ≥ 20 years) included: no headache (27,945), non-migrainous headache (12,517), MO (2820), and MA (585). ICHD-II criteria used for diagnosis.	Compared to no-headache group: Lower HDL-c in migraine. No difference in TC levels.
Stam et al., 2013 (62)	360 migraineurs +617 controls. Population study.	Migraineurs (209 MO 209, 151 MA). Migraineurs more likely to be women, smokers, lower education. ICHD-II criteria used for diagnosis.	Compared to controls: Lower HDL-c in migraine. No differences were found for other atherosclerosis parameters.
Bernecker et al., 2011 (52)	50 migraineurs +74 controls. Case-control study.	Total of 124 participants with 50 persons people (42 women) with migraine (MA = 20). Exclusion criteria: BMI ≥ 30, CDH, depression, diabetes, thyroidal dysfunction, metabolic and CV diseases, chronic illnesses, infectious diseases, elevated hs-CRP.	Compared to controls: Increased oxidized LDL in migraineurs. No differences in cholesterol, HDL, LDL, triglycerides, Lp(a).
Nelson et al., 2010 (26)	2295 headache patients +9475 controls. Population study.	Children (4–11 yo) and adolescents (12–19 yo) with severe headache. Inclusion: 4–19 yo; diagnosis of headache by self-report. The prevalence of headache rose with increasing age. Highest headache prevalence, 27.4%, in girls aged 16–18 years.	Compared to controls: No difference in TC, HDL-c, non-HDL-c and LDL-c in headache.
Benseñor et al., 2011 (63)	165 migraineurs +1285 controls. Longitudinal study.	Participants were 65–102 years of age, with; low-income elderly in Brazil. Migraineurs were younger than non-migraineurs (mean age 70.6 vs 72.1 years; p = 0.001).	Compared to controls: No difference regarding lipid profile after adjustment for age and sex.
Jahromi et al., 2013 (64)	320 migraineurs +1190 controls. Cross-sectional study.	All the participants (n = 1510) were females and obese with mean age of 38.5 ± 12.6 years. IHS criteria used for diagnosis.	Compared to controls: No differences cholesterol, LDL-c, HDL-c in migraine.
Harandi et al., 2013 (65)	347 migraineurs +267 controls. Observational study.	Participants were 18–55 years old. Migraineurs without aura, higher % females, higher age, BP, and lower blood sugar. ICHD-II criteria used for diagnosis.	Compared to controls: No difference for LDL, HDL in migraine.
Fava et al., 2014 (66)	166 migraineurs +83 controls. Case-control study.	All participant were female. Migraineurs (83 EM, 83 CM). Migraine diagnosed by ICHD criteria.	Compared to controls: No difference in lipids for EM. No difference in lipids for CM.
Rockett et al., 2013 (67)	30 migraineurs +29 controls. Case-control study.	Migraineurs (15 normal weight and 15 obese). Controls (15 normal weight and 14 obese).	Compared to normal-weight controls and migraineurs: Lower HDL-c in obese migraineurs.

MA: migraine with aura; MO: migraine without aura; BMI: body mass index; hs-CRP: high-sensitivity C-reactive protein; ICHD: International Classification of Headache Disorders; IHS: International Headache Society; CV: cardiovascular; MI: myocardial infarction; DM: diabetes mellitus; LDL: low-density lipoprotein; HDL-c: high-density lipoprotein cholesterol; EM: episodic migraine; CM: chronic migraine; Lp(a): lipoprotein a; TC: total cholesterol; CSF: cerebrospinal fluid; BP: blood pressure; yo: years old; OR: odds ratio; CI: confidence interval.

Biomarkers of thrombosis and coagulation

Platelet activation

Serotonin is an important regulator of vasoactivity and pain sensitization. Platelets carry the majority of blood serotonin, and through aggregation are stimulated to release it into the circulation. Platelets are activated to aggregate by exposure to injured endothelium and conditions of high shear stress. The platelet has been an early popular target for speculation as to its role in migraine pathogenesis (68), and later for conjecture as serving as a link between migraine and stroke (69). In studies of platelet function in migraine there have been reports of increased platelet activation during and between attacks (70 –75), as well as only between the attacks (68,76), or only during attacks (77,78). Studies have also shown a decrease, rather than increase, in platelet aggregation during or between migraine attacks (79 –81). The lack of uniformity in results may, in part, be related to variations in experimental methods and devices. Nonetheless, it has obscured the role of platelets in migraine. A recent report that aspirin is efficacious in prevention of MA rather than MO (82) is interesting to consider in the context of the link between MA and stroke. Ischemia is a recognized experimental technique for inducing CSD, the neuronal process widely recognized to be the physiological template of aura. A platelet clot could, thus, potentially cause cortical ischemia, thereby generating CSD/aura in susceptible individuals, and in rare cases, migrainous infarction.

vWF

vWF, widely accepted as an important plasma biomarker of ED in diseased populations (83), activates platelet glycoprotein IIb/IIIa receptors, causing platelet adhesion and aggregation (84). This procoagulatory protein has been associated with each major cardiovascular risk factor (85,86), as well with stroke in clinic and community-based longitudinal studies (87,88). The literature on vWF in migraine is sparse (Table 4) (22,89 –91). In one study, migraineurs with prior stroke had higher vWF levels than did migraineurs without stroke, and both groups had higher vWF levels than controls (90). In a larger case-control study evaluating young female migraineurs (22), the mean interictal vWF activity values in migraineurs was similar to those found in the two previous studies (90,91). The hypothesis that elevated vWF in migraine is indicative of endothelial activation is supported by correlations of vWF with the inflammatory biomarker CRP. Ictal studies of migraine have documented increased vWF levels during the course of a migraine attack (89). Interestingly, low vWF levels, in patients with von Willebrand disease, (vWD) have also been associated with migraine (92).

Table 4.

Coagulation biomarkers in migraine.

Study author/year	Sample size/study design	Sample characteristics	Observations
von Willebrand factor (vWF)
Cesar et al., 1995 (89)	17 migraineurs +25 controls. Case-control study.	Migraineurs all MO. IHS criteria for migraine diagnosis.	Ictal rise in vWF antigen from 72.4 ± 29 U/dl in the headache-free period to 130.2 ± 75 U/dl during the attack (p < 0.01).
Tietjen et al., 2007 (91)	54 migraineurs +30 controls. Case-control study.	Migraineurs (19 MA, 20 MO, 15 transformed), 94% female. Controls matched for age, but were more educated. ICHD criteria for diagnosis.	Compared to controls: Higher mean vWF activity and antigen levels in migraine, especially those with livedo reticularis.
Tietjen et al., 2001 (90)	74 migraineurs +35 controls. Case-control study.	Migraineurs included 11 individuals with prior stroke. IHS criteria used for migraine diagnosis.	Compared to controls: Higher vWF antigen and activity in migraine, especially with prior stroke.
Tietjen et al., 2009 (22)	125 migraineurs +50 controls. Case-control study.	Participants all premenopausal women, mean age 37 ± 9 years. Migraineurs (61 MA, 64 MO) matched for age, race, education, income. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: Higher adjusted odds for elevated vWF activity (OR 6.5; 1.94, 21.83) in MA, (OR 4.6; 1.37, 15.38) in MO.
Kirtava et al., 2003 (92)	102 vWD patients +88 controls. Case-control study.	Cases (Mean age= 39 ± 11.9 yrs, range = 18–70). Controls (Mean age 38 ± 11.2 yrs, range = 18–65). Participants mostly white, married, with some college education.	Compared to controls: More migraine in patients with vWD (41% vs. 13%, p < 0.0001).
Fibrinogen
Tietjen et al., 2011 (25)	283 migraineurs +134 controls. Population study (CAMERA Study).	Migraineurs (155 MA, 128 MO), similar to controls for age, education, hypertension, tobacco, diabetes, cholesterol, or contraceptive use. Inclusion: 30–60 yo, diagnosis by ICHD criteria.	Compared to controls: Higher adjusted odds for elevated fibrinogen in migraine (OR 1.89; 1.20–2.95), and especially MA (OR 1.97; 1.19–3.27).
Walkowiak et al., 1989 (102)	30 migraineurs +20 controls. Case-control study.	Migraineurs (n = 30) and control donors (n = 24). Methods: Binding of fibrinogen to platelets washed from blood.	Compared to controls (18.75 mcg/10 (8) platelets): Increased fibrinogen (37.26 mcg/10 (8) platelets) in migraine (p < 0.001).
Yucel et al., 2013 (103)	59 migraineurs +30 controls. Case-control study.	Migraineurs (49 MA, 52 females) and controls (26 females) matched for age. Exclusion: Chronic disease, renal disease, pregnancy, infectious disease, thyroid, and cardiovascular dysfunction.	Compared to controls: Higher mean levels of fibrinogen in migraine. No difference of ictal vs interictal values.
Kurth et al., 2008 (27)	5087 migraineurs +22,539 controls. Population study.	Migraineurs (70.5% with active migraine, of whom 39.7% have MA). Inclusion: Women > 45 yo; headaches diagnosed by validated questionnaire.	Compared to controls: No difference in fibrinogen levels for migraine, active MA, active MO, prior migraine.
Bianchi et al., 1996 (101)	17 migraineurs +11controls. Case-control study.	Matched for age and gender. Observations were made during headache-free periods.	Compared to controls: Lower fibrinogen levels in migraine.
Tissue plasminogen activator (tPA)
Tietjen et al., 2009 (22)	125 migraineurs +50 controls. Case-control study	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education and income. Inclusion: 18–50 yo; premenopausal, diagnosed by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis, anti-platelet drugs.	Compared to controls: Higher tPA antigen levels in migraine.
Bianchi et al., 1996 (101)	17 migraineurs +11 controls. Case-control study.	The study participants were matched for age and gender. Observations were made during headache-free periods.	Compared to controls: Lower levels of tPA in migraine.

MA: migraine with aura; MO: migraine without aura; BMI: body mass index; hs-CRP: high-sensitivity C-reactive protein; ICHD: International Classification of Headache Disorders; IHS: International Headache Society; CV: cardiovascular; MI: myocardial infarction; DM: diabetes mellitus; LDL: low-density lipoprotein; HDL-c: high-density lipoprotein cholesterol; EM: episodic migraine; CM: chronic migraine; Lp(a): lipoprotein a; tPA: tissue plasminogen activator; CSF: cerebrospinal fluid; BP: blood pressure; yo: years old; OR: odds ratio; CI: confidence interval; vWD: von Willebrand Disease.

Fibrinogen

Fibrinogen is a plasma glycoprotein produced by hepatocytes whose major function is as a precursor to fibrin during blood clot formation. As a bridge between platelets, fibrinogen is involved in primary hemostasis and platelet aggregation, as well as leukocyte-endothelial cell interaction. It is also the main determinant of whole blood and plasma viscosity (93). Elevated fibrinogen levels induce a state of hypercoagulability, trigger endothelial injury, and aggravate cerebral hypoperfusion (94,95). Fibrinogen may also damage blood vessel walls by causing smooth muscle proliferation and migration (96). In some, but not all, epidemiological studies fibrinogen has been linked to heart disease and stroke, including in young and middle-aged adults (97 –99). Plasma fibrinogen levels have also been associated with the risk of silent cerebrovascular lesions (100). There are only a few studies of fibrinogen in migraine (Table 4). Although the WHS, which included women >45 years of age, found no differences in fibrinogen levels between migraineurs and controls (27), one study reported slightly lower plasma levels in migraineurs (101), whereas three others have reported an increase of fibrinogen in migraineurs compared to controls (25,102,103). Additionally, data from the CAMERA general population-based study revealed that the elevated fibrinogen levels were most closely associated with MA in women with a high frequency of attacks (25). The analysis controlled for use of estrogen-containing oral contraceptives (OCP), which have been linked to elevated levels of fibrinogen.

tPA

tPA is a serine protease on endothelial cells that catalyzes the conversion of plasminogen to plasmin, the enzyme involved in clot breakdown. Elevated tPA antigen reflects decreased fibrinolysis. In a population-based case-control study of young (15- to 44-year-old) women with stroke, elevated tPA antigen levels were an independent marker of increased stroke risk (Table 4) (104). We found tPA antigen levels to be elevated in the plasma of young female migraineurs between attacks as compared with headache-free controls (22). A case-control study of 17 individuals with MO reported reduced tPA levels (101).

Homocysteine

Homocysteine is a metabolite of the amino acid methionine and is associated with increased risk for premature vascular disease and thrombosis. In five interictal case-control studies, homocysteine was elevated in migraine or in headache compared to controls (Table 5) (26,105 –108). In one study, levels were higher in MA than MO (105), and in the others there was no difference (106). In three studies, including one of all premenopausal women, and another of postmenopausal women, there were no significant differences between migraineurs and controls, or between MA and MO subtypes of migraine (22, 27,109).

Table 5.

Association of homocysteine with migraine and headache.

Study author/year	Sample size/study design	Sample characteristics	Observations
Isobe and Terayama, 2010 (105)	60 migraineurs. Observational study.	Migraineurs (30 MA, mean age 40.8 ± 6.7 years; 30 MO, mean age 42.4 ± 5.0 years). Inclusion: Diagnosed by ICHD criteria. Exclusion: Smokers, hypertensive, diabetics, dyslipidemics.	Compared to controls (84.9 ± 24.5): Higher homocysteine in MA (311.0 ± 62.2; p < 0.0001). Higher homocysteine in MO (120.6 ± 30.4; p < 0.001). Compared to MO: Higher homocysteine in MA (p < 0.001).
Bahadir et al., 2013 (106)	150 migraineurs +107 controls. Case-control study.	Migraineurs (52% MA, 48% MO), mean age of 37.0 ± 9.0 years. Inclusion: Diagnosed by ICHD criteria. Exclusion: Diabetes mellitus DM, vasculitis, prior stroke, pregnancy, MI, SLE, NSAIDs, antiplatelet agents.	Compared to controls: Higher homocysteine in MA. Higher homocysteine in MO. Compared to MO: No difference in homocysteine in MA.
Nelson et al., 2010 (26)	2295 headache patients +9475 controls. Population study.	Children (4–11 yo) and adolescents (12–19 yo) with severe headache. Inclusion: 4–19 yo; diagnosis of headache by self-report.	Compared to non-headache controls: Higher homocysteine in headache (OR 1.87, 95% CI 1.60, 2.18), and in girls and boys subsets. More children with headache were in the highest-risk quintile.
Gavgani and Hoseinian, 2012 (107)	65 migraineurs +65 controls. Case-control study.	Migraineurs (61.5% female) average age of 27.69 ± 9.50, range 15–55 years, matched for age and gender. Inclusion: Diagnosed by ICHD criteria.	Compared to controls: Higher mean homocysteine levels in migraine (14 vs 11, p < 0.001). Higher hyperhomocysteinemia in migraine (OR 2.56), especially in women.
Moschiano et al., 2008 (108)	136 migraineurs +117 controls. Case-control study.	Migraineurs (90 females, 14–61 years) matched for age and sex. Inclusion: Diagnosed by ICHD criteria. Exclusion: Conditions inducing increased homocysteine levels.	Compared to controls: Higher mean homocysteine levels and hyperhomocysteinemia in migraine, especially in men.
Kurth et al., 2008 (27)	5087 migraineurs +22,539 controls. Population study (Women’s Health Study).	Migraineurs (70.5% with active migraine, of whom 39.7% have MA). Inclusion: Women > 45 yo; headaches diagnosed by validated questionnaire.	Compared to controls: No difference in homocysteine any migraine, active MA, active MO, and prior migraine.
Tietjen et al., 2009 (22)	125 migraineurs +50 controls. Case-control study.	Migraineurs (61 MA, 64 MO) 100% female, matched for age, race, education, and income. Mean frequency 12 ± 9 days/month. Mean age 37 ± 9 years. Inclusion: 18–50 yo; premenopausal, diagnosis by ICHD criteria. Exclusion: Stroke, MI, DM, vasculitis and anti-platelet drugs.	Compared to controls: No difference in homocysteine in migraine, MA or MO.
Hering-Hanit et al., 2001 (109)	78 migraineurs + 126 controls. Case-control study.	Migraineurs (22 MA, 56 MO; 11 men, 67 women) matched to controls (74 men, 52 women) for age and gender. Exclusion: HTN, DM, CAD, and smokers.	Compared to controls: No difference in homocysteine in migraineurs.

MA: migraine with aura; MO: migraine without aura; BMI: body mass index; ICHD: International Classification of Headache Disorders; MI: myocardial infarction; DM: diabetes mellitus; LDL: low-density lipoprotein; HDL-c: high-density lipoprotein cholesterol; EM: episodic migraine; CM: chronic migraine; SLE: systemic lupus erythematosus; NSAIDs: nonsteroidal anti-inflammatory drugs; CSF: cerebrospinal fluid; yo: years old; OR: odds ratio; CI: confidence interval.

Biomarkers of endothelial activation and dysfunction

EMPs

Circulating EMPs are novel biomarkers of ED. The mechanism of EMP generation in vivo is uncertain, but cell culture experiments demonstrate EMP production in response to endothelial exposure to inflammatory cytokines. EMPs then trigger endothelial release of chemokines and attract leukocytes to the endothelial surface, thus promoting inflammation and thrombosis (12). EMPs also inhibit endothelial nitric oxide synthase, which impairs vasodilation and increases arterial stiffness. In a recent case-control study (110) of premenopausal women, investigators demonstrated that activated EMP levels were higher in those with MA, the subtype associated with elevated stroke risk. The levels also correlated with the degree of arterial stiffness as estimated from the finger tonometry-derived augmentation index. Elevated levels of platelet and monocyte microparticles indicate a pro-inflammatory and hypercoaguable state, and it is uncertain whether this is a cause or a consequence of aura.

ADMA

ADMA is an endogenous inhibitor of nitric oxide synthase, thereby serving to regulate nitric oxide production in the endothelium. Elevated levels of ADMA have been associated with oxidative stress, ED, atherosclerosis, and cardiovascular disease (111). In one study of ADMA in migraineurs during the interictal period, plasma levels of ADMA and NO (nitrates and nitrites) did not differ between migraineurs and controls (112). In a second study, ADMA and NO concentrations were elevated in migraineurs compared to controls, but in the migraine cohort there were no differences between the ictal and interictal levels (113).

ET-1

ET-1 is a peptide produced by the endothelium that regulates vascular tone. Known primarily as a potent vasoconstrictor of smooth muscle cells, ET-1 also binds to ET receptor type B receptors on the endothelium to effect NO-mediated vasodilation.

Interictal studies

Three studies showed an increase in ET-1 levels measured between migraine attacks compared to controls (24,101,114), and one showed ET-1 levels to be lower in the migraine cohort (115).

Ictal studies

Levels of ET-1 rose during the course of the migraine attack in four of five studies investigating this (114,116 –119). A population-based case-control study of young women with ischemic stroke found, however, no evidence that genes regulating endothelin-1 or the endothelin receptor type B mediate the association between migraine and stroke (120).

Pro-brain natruretic peptide (Pro-BNP)

Pro-BNP is a cardiac neurohormone secreted from the ventricles in response to volume expansion and pressure overload. This hormone inhibits the sympathetic nervous system as well as the renin-angiotensin-aldosterone system. As a possible marker of preclinical cardiac involvement, this neurohormone has been evaluated in one study of individuals with migraine, and results demonstrated that patients with migraine had higher pro-BNP concentration than healthy controls (40).

Biomarkers of endothelial regeneration and repair

EPCs

Newer lines of investigation include the study of EPCs, which are circulating cells released from the bone marrow (14). Able to differentiate to become part of the functioning endothelial lining, EPCs have a role in endothelial regeneration and repair of injured vessels. The first study demonstrated that EPC numbers were reduced in migraine, especially MA, compared to controls and to those with TTH (121). Function of EPCs was also reduced in people with migraine, showing reduced migratory capacity and increased cellular senescence. These findings were interpreted as being related to chronic inflammation in migraine, and as a sign of greater cardiovascular risk. The second study compared patients with migraine (during the interictal period) to controls in a case-control design. Similar to findings in the Lee study, the number of EPCs was reduced in those with migraine (122). In addition the EPCs counts decreased with the longer duration of migraine. These findings were taken as an indication of altered endothelial function in migraine. The third study enrolled patients with episodic migraine (EM), chronic migraines (CM) and controls (123). The investigators evaluated the number of EPCs and their capacity to make colony-forming units (CFUs), in addition to stratifying EPCs as “early” or “late” based on surface markers. The main finding was that CFUs and the total number of EPC were no different in the three main clinical groups (CM, EM and controls). The more mature EPCs, markers for vascular damage and the need for endothelial repair, were significantly lower in controls than in any migraine subset (CM, EM, MA or MO).

Discussion

Summary of findings

No biomarkers investigated to date have the sensitivity and specificity to diagnose migraine outside of its clinical context. A number of positive vascular biomarker studies in migraine populations lend insight into possible pathophysiological mechanisms by which migraine may be associated with stroke. The lack of consistency among study results and the often poor quality of evidence, however, makes it difficult to conclude with certainty that migraine is a pro-inflammatory and pro-coagulatory condition. Although cytokine and procoagulant biomarkers only indirectly implicate the vasculature, ictal increases lend credence to the hypothesis that biomarker changes are a consequence of the migraine attack. The finding that adverse cholesterol changes (i.e. elevated total cholesterol, total cholesterol:HDL ratio, and LDL, or decreased HDL) are associated with migraine is relatively consistent among many investigations. Whether these differences, which are often small, have clinical relevance and contribute to the migraine-vascular disease association is uncertain. The exciting identification of EMP, an emerging vascular biomarker that allows both for qualitative and quantitative ascertainment of endothelial activation, places the data on inflammation and coagulation in a unifying context. Further support that the endothelium is involved in migraine comes from the studies of precursor cells, which function as vehicles for regeneration and repair.

Implications

Mechanistic

As detailed in this review, there is growing evidence from biomarker studies of endothelial activation in migraine. However, endothelial activation is only one component of ED; impaired vascular reactivity is the other. Although there is solid evidence of greater arterial stiffness in migraine, in studies of vascular reactivity in migraine, as summarized in a recent comprehensive review (124), a greater number showed no difference in comparison to the control groups, as showed impairment. In the natural history of ED, a precursor of atherosclerosis, endothelial activation precedes vascular reactivity impairment. Curiously, evidence suggests that in migraine this may not be the case (62). Migraine-related stroke is most prevalent in people with favorable vascular risk profiles, whether young (125,126) or old (3). Also, many of the vascular conditions associated with migraine, including Raynaud’s phenomenon, livedo reticularis, vasospastic angina, and pre-eclampsia, are endotheliopathies but are not atherogenic in nature. It is uncertain, based on current evidence in migraine, whether endothelial activation is uncoupled from the atherosclerotic process, or whether strokes, possibly related to activation-induced thrombosis, occur prior to development of either impaired reactivity or atherosclerosis. Prothrombotic physiological changes in the brain endothelium may be precipitated by aura-related oligemia or oxidative stress. Or CSD may occur as a consequence of thrombosis-induced (mild) ischemia in susceptible individuals, i.e. aura as a transient ischemic attack (TIA) variant. These hypotheses are not mutually exclusive. Aura may precipitate release of inflammatory cytokines, which then activate the endothelium, thus leading to increased coagulation, and thrombosis-induced ischemia and finally either aura or stroke. In support of this, epidemiological evidence suggests that prothrombotic risk factors, such as OCP use and presence of patent foramen ovale, predispose to aura and also place MA migraineurs at particular risk of stroke.

Clinical

The implications of our findings from the clinical and public health perspective revolve around the issues of screening and treatment. If MA carries a risk of stroke, should all MA patients be screened with a select battery of the vascular biomarkers? If biomarker testing reveals evidence of endothelial activation, what is the next step? Options include nonpharmacological therapies that promote endothelial health, including exercise, meditation, and weight loss. Pharmacological agents that improve endothelial function and are also proven to prevent migraine include angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers (127,128). Daily aspirin is also a consideration based on a preliminary retrospective analysis, which showed it to be effective prophylaxis for MA (82). The longitudinal WHS, however, did not find evidence that aspirin therapy decreased stroke risk more in women with migraine than those without (129).

Limitations of this review

We relied on previously published research studies and the availability of these studies using the method outlined in the study procedures. Our criteria for selection, restricting studies to those published in English from 1990 to present, may have introduced bias. In the case of platelet studies in migraine, with the majority of studies published before 1990, we made an exception to our publication date criterion. This is a narrative review and not a quantitative meta-analysis. We cannot, therefore, comment on aspects such as effect sizes for all studies, correlation coefficients, and other quantitative measures. Various evaluation methodologies and outcome indices were used in the chosen studies. In our study selection criteria, attempts were not made to filter studies based on methodology or outcome indicators. Therefore, differences between individual studies based on biochemical assay methods and outcome measures could be extensive, thus limiting the results of this review. Also, for the vast majority of studies, only blood assay and blood data reporting studies were included. For the studies we included in this review, there were substantial differences in study participants based on age, gender, income, education, and geographic location. In addition, there was a large variation in sample size of studies included in this review (from n of 7 to more than 20,000 study participants). The majority of studies used a convenience sample of participants from clinics, which could have introduced a sampling bias. With self-reported variables (e.g. migraine history, frequency, medication usage, personal and medical history, and other background characteristics), reliability may be limited. Because of the conflicting nature of results from various studies, our findings on the association of specific vascular biomarkers with migraine remain inconclusive.

Recommendations for future studies

One of the most intriguing and promising avenues for future biomarker research in migraine is in the field of microparticles related to endothelial cells, platelets, and monocytes. Understanding the underlying nature (i.e. apoptosis, injury, activation) and severity of the producing stimulus is valuable information while investigating EMPs in larger, more diverse populations of migraineurs with regards to age, sex, race, migraine subtype, frequency, duration, and severity. The role of EMPs in individuals with patent foramen ovale or with subclinical white matter and infarct-like lesions on cerebral magnetic resonance imaging (MRI) remains uncertain. Determining the relationship of EMPs to EPCs, vascular reactivity, and atherosclerosis in migraineurs may also bring new insights on pathogenesis of migraine and migraine-related stroke. Following EMP levels after various treatments will provide a gauge of effects on endothelial health in relationship to clinical course. Ultimately, it would be beneficial to be able to test biomarkers in longitudinal studies of migraine to determine those that best predict adverse consequences such as stroke and other vascular events.

Clinical implications

There are no definitive vascular biomarkers in people with migraine.

Promising biomarkers needing further investigation in the migraine population include endothelial microparticles and endothelial precursor cells.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

None declared.

References

Lipton

Bigal

Diamond

. Migraine prevalence, disease burden, and the need for preventive therapy. Neurology 2007; 68: 343–349.

Schürks

Rist

Bigal

. Migraine and cardiovascular disease: Systematic review and meta-analysis. BMJ 2009; 339: b3914– b3914.

Kurth

Schürks

Logroscino

. Migraine, vascular risk, and cardiovascular events in women: Prospective cohort study. BMJ 2008; 337: a636– a636.

O’Keeffe

Tsapatsaris

Beetham

WP . Association between Raynaudnts in women: disease burden, and the need for preventive therapy. J Rheumatol 1993; 20: 1187–1188.

Rose

Carson

Sanford

. Migraine and other headaches: Associations with Rose angina and coronary heart disease. Neurology 2004; 63: 2233–2239.

Adeney

Williams

. Migraine headaches and preeclampsia: An epidemiologic review. Headache 2006; 46: 794–803.

Tietjen

Gottwald

Al Qasmi

. Migraine is associated with livedo reticularis: A prospective study. Headache 2002; 42: 263–267.

Tietjen

. Migraine as a systemic vasculopathy. Cephalalgia 2009; 29: 989–996.

Baumgartner-Parzer

Waldhäusl

. The endothelium as a metabolic and endocrine organ: Its relation with insulin resistance. Exp Clin Endocrinol Diabetes 2001; 109(Suppl 2): S166–S179.

10.

Rajendran

Rengarajan

Thangavel

. The vascular endothelium and human diseases. Int J Biol Sci 2013; 9: 1057–1069.

11.

Bonetti

Lerman

. Endothelial dysfunction: A marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol 2003; 23: 168–175.

12.

Chironi

Boulanger

Simon

. Endothelial microparticles in diseases. Cell Tissue Res 2009; 335: 143–151.

13.

Markiewicz

Richard

Marks

. Impact of endothelial microparticles on coagulation, inflammation, and angiogenesis in age-related vascular diseases. J Aging Res 2013; 2013: 734509–734509.

14.

Hill

Zalos

Halcox

. Circulating endothelial progenitor cells, vascular function, and cardiovascular risk. N Engl J Med 2003; 348: 593–600.

15.

Mortensen

. C-reactive protein, inflammation, and innate immunity. Immunologic Research 2001; 24: 163–176.

16.

Du Clos

. Function of C-reactive protein. Ann Med 2000; 32: 274–278.

17.

Vermeire

Van Assche

Rutgeerts

. C-reactive protein as a marker for inflammatory bowel disease. Inflamm Bowel Dis 2004; 10: 661–665.

18.

Ridker

Hennekens

Buring

. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000; 342: 836–843.

19.

Pearson

Mensah

Alexander

. Markers of inflammation and cardiovascular disease: Application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003; 107: 499–511.

20.

Di Napoli

Papa

Bocola

. C-reactive protein in ischemic stroke: An independent prognostic factor. Stroke 2001; 32: 917–924.

21.

Vanmolkot

de Hoon

. Increased C-reactive protein in young adult patients with migraine. Cephalalgia 2007; 27: 843–846.

22.

Tietjen

Herial

White

. Migraine and biomarkers of endothelial activation in young women. Stroke 2009; 40: 2977–2982.

23.

Güzel

Taşdemir

Celik

. Evaluation of serum transforming growth factor in young women. graine. Prevention and thene patients. Neurol Neurochir Pol 2013; 47: 357–362.

24.

Hamed

Ezz Eldin

. Vascular risk factors, endothelial function, and carotid thickness in patients with migraine: Relationship to atherosclerosis. J Stroke Cerebrovasc Dis 2010; 19: 92–103.

25.

Tietjen

Khubchandani

Herial

. Migraine and vascular disease biomarkers. A population-based study. Cephalalgia 2011; 31(1 Suppl): 11–12.

26.

Nelson

Richardson

. Headache and biomarkers of vascular disease in a representative sample of US children. Arch Pediatr Adolesc Med 2010; 164: 358–362.

27.

Kurth

Ridker

Buring

. Migraine and biomarkers of cardiovascular disease in women. Cephalalgia 2008; 28: 49–56.

28.

Guldiken

Demir

. Soluble CD40 ligand and prolactin levels in migraine patients during interictal period. J Headache Pain 2011; 12: 355–360.

29.

Guldiken

Demir

. Insulin resistance and high sensitivity C-reactive protein in migraine. Can J Neurol Sci 2008; 35: 448–451.

30.

Bernecker

Ragginer

Fauler

. Oxidative stress is associated with migraine and migraine-related metabolic risk in females. Eur J Neurol 2011; 18: 1233–1239.

31.

Gruber

Bernecker

Pailer

. Lipid profile in normal weight migraineurs – evidence for cardiovascular risk. Eur J Neurol 2010; 17: 419–425.

32.

Silva

Rueda-Clausen

Silva

. Endothelial function in patients with migraine during the interictal period. Headache 2007; 47: 45–51.

33.

Gudmundsson

Aspelund

Scher

. C-reactive protein in migraine sufferers similar to that of non-migraineurs: The Reykjavik study. Cephalalgia 2009; 29: 1301–1310.

34.

Denes

Pinteaux

Rothwell

. Interleukin-1 and stroke: Biomarker, harbinger of damage, and therapeutic target. Cerebrovasc Dis 2011; 32: 517–527.

35.

Kocer A, Kocer E, Memisogullari R, et al. Interleukin-6 levels in tension headache patients. Clin J Pain 2010; 26: 690–693.

36.

Bruno

Carpino

. An overview on immune system and migraine. Eur Rev Med Pharmacol Sci 2007; 11: 245–248.

37.

Tietjen

Khubchandani

Khan

. Adiponectin and inflammatory cytokines in young women with migraine. Headache 2010; 50: S81–S81.

38.

Perini

D’Andrea

Galloni

. Plasma cytokine levels in migraineurs and controls. Headache 2005; 45: 926–931.

39.

Fidan

Yüksel

Ymir

. The importance of cytokines, chemokines and nitric oxide in pathophysiology of migraine. J Neuroimmunol 2006; 171: 184–188.

40.

Uzar

Evliyaoglu

Yucel

. Serum cytokine and pro-brain natriuretic peptide (BNP) levels in patients with migraine. Eur Rev Med Pharmacol Sci 2011; 15: 1111–1116.

41.

Gergont

Kaciński

Kwinta

. Proinflammatory cytokines in children with idiopathic headache [in Polish]. Przegl Lek 2005; 62: 1269–1275.

42.

Sarchielli

Alberti

Baldi

. Proinflammatory cytokines, adhesion molecules, and lymphocyte integrin expression in the internal jugular blood of migraine patients without aura assessed ictally. Headache 2006; 46: 200–207.

43.

van Hilten

Ferrari

Van der Meer

. Plasma interleukin-1, tumour necrosis factor and hypothalamic-pituitary-adrenal axis responses during migraine attacks. Cephalalgia 1991; 11: 65–67.

44.

Rozen

Swidan

. Elevation of CSF tumor necrosis factor alpha levels in new daily persistent headache and treatment refractory chronic migraine. Headache 2007; 47: 1050–1055.

45.

Bø

Davidsen

Gulbrandsen

. Cerebrospinal fluid cytokine levels in migraine, tension-type headache and cervicogenic headache. Cephalalgia 2009; 29: 365–372.

46.

Vallejo

Tilley

Vogel

. The role of glia and the immune system in the development and maintenance of neuropathic pain. Pain Pract 2010; 10: 167–184.

47.

Munno

Marinaro

Bassi

. Immunological aspects in migraine: Increase of IL-10 plasma levels during attack. Headache 2001; 41: 764–767.

48.

Ishizaki

Takeshima

Fukuhara

. Increased plasma transforming growth factor-beta1 in migraine. Headache 2005; 45: 1224–1228.

49.

Schürks

Kurth

Buring

. A candidate gene association study of 77 polymorphisms in migraine. Pain 2009; 10: 759–766.

50.

Song

Lee

. Adiponectin as a new paradigm for approaching Alzheimer’s disease. Anat Cell Biol 2013; 46: 229–234.

51.

Peterlin

Alexander

Tabby

. Oligomerization state-dependent elevations of adiponectin in chronic daily headache. Neurology 2008; 70: 1905–1911.

52.

Bernecker

Pailer

Kieslinger

. Increased matrix metalloproteinase activity is associated with migraine and migraine-related metabolic dysfunctions. Eur J Neurol 2011; 18: 571–576.

53.

Peterlin

Tietjen

Gower

. Ictal adiponectin levels in episodic migraineurs: A randomized pilot trial. Headache 2013; 53: 474–490.

54.

Nozari

Dilekoz

Sukhotinsky

. Microemboli may link spreading depression, migraine aura, and patent foramen ovale. Ann Neurol 2010; 67: 221–229.

55.

Rist

Tzourio

Kurth

. Associations between lipid levels and migraine: Cross-sectional analysis in the epidemiology of vascular ageing study. Cephalalgia 2011; 31: 1459–1465.

56.

Scher

Terwindt

Picavet

. Cardiovascular risk factors and migraine: The GEM population-based study. Neurology 2005; 64: 614–620.

57.

Bigal

Kurth

Santanello

. Migraine and cardiovascular disease: A population-based study. Neurology 2010; 74: 628–635.

58.

Gilad

Boaz

Dabby

. Migraine and vascular risk factors in the elderly. Geriatr Gerontol Int 2014; 14: 220–225.

59.

Monastero

Pipia

Cefalù

. Association between plasma lipid levels and migraine in subjects aged > or =50 years: Preliminary data from the Zabùt Aging Project. Neurol Sci 2008; 29(Suppl 1): S179–S181.

60.

Saberi

Hatamian

Kazemnejad

. Hyperlipidemia in migraine: Is it more frequent in migraineurs? Iran J Neurol 2011; 10: 46–50.

61.

Winsvold

Hagen

Aamodt

. Headache, migraine and cardiovascular risk factors: The HUNT study. Eur J Neurol 2011; 18: 504–511.

62.

Stam

Weller

Janssens

. Migraine is not associated with enhanced atherosclerosis. Cephalalgia 2013; 33: 228–235.

63.

Benseñor

Goulart

Lotufo

. Cardiovascular risk factors associated with migraine among the elderly with a low income: The Sao Paulo Ageing & Health Study (SPAH). Cephalalgia 2011; 31: 331–337.

64.

Jahromi

Abolhasani

Meysamie

. The effect of body fat mass and fat free mass on migraine headache. Iran J Neurol 2013; 12: 23–27.

65.

Harandi

Togha

Sadatnaseri

. Cardiovascular risk factors and migraine without aura: A case-control study. Iran J Neurol 2013; 12: 98–101.

66.

Fava

Pirritano

Consoli

. Chronic migraine in women is associated with insulin resistance: A cross-sectional study. Eur J Neurol 2014; 21: 267–272.

67.

Rockett

Perla Ada

Perry

. Cardiovascular disease risk in women with migraine. J Headache Pain 2013; 14: 75–75.

68.

Hanington

Jones

Amess

. Migraine: A platelet disorder. Lancet 1981; 2: 720–723.

69.

Borgdorff

Tangelder

. Migraine: Possible role of shear-induced platelet aggregation with serotonin release. Headache 2012; 52: 1298–1318.

70.

Couch

Hassanein

. Platelet aggregability in migraine. Neurology 1977; 27: 843–848.

71.

Manotti

Quintavalla

Manzoni

. Platelet activation in migraine. Thromb Haemost 1983; 50: 758–758.

72.

Hilton

Cumings

. 5-Hydroxytryptamine levels and platelet aggregation responses in subjects with acute migraine headache. J Neurol Neurosurg Psychiatry 1972; 35: 505–509.

73.

Jones

Forsythe

Amess

. Platelet aggregation in migraine patients during the headache-free interval. Adv Neurol 1982; 33: 275–278.

74.

Lechner

Ott

Fazekas

. Evidence of enhanced platelet aggregation and platelet sensitivity in migraine patients. Cephalalgia 1985; 5(Suppl 2): 89–91.

75.

Pawlowska

Kozubski

Walkowiak

. Increased platelet glycoprotein IIb reflects an abnormality of the platelet membrane in migraine. Headache 1988; 28: 60–60.

76.

Kozubski

Walkowiak

Cierniewski

. Platelet fibrinogen receptors in migraine patients. Headache 1987; 27: 431–434.

77.

Kovacs

Herman

Filep

. Platelet aggregation of migraineurs during and between attacks. Cephalalgia 1990; 10: 161–165.

78.

Jha

Varma

Garg

. Platelet aggregation in migraine. J Assoc Physicians India 1992; 40: 153–154.

79.

Deshmukh

Meyer

. Cyclic changes in platelet dynamics and the pathogenesis and prophylaxis of migraine. Headache 1977; 17: 101–108.

80.

Hermán

Magyar

Kovács

. Decreased sensitivity of platelets to platelet-activating factor in migraine patients during the headache-free interval. Thromb Haemost 1989; 62: 818–818.

81.

D’Andrea

Cananzi

Perini

. Decreased collagen induced platelet aggregation and increased platelet arginine levels in migraine: A possible link with the NO pathway. Cephalalgia 1994; 14: 352–356.

82.

Savi

Condello

Bert

. Prophylaxis of migraine with aura: A place for acetylsalicylic acid. J Headache Pain 2013; 1(Suppl 1): P195–P195.

83.

Blann

. Plasma von Willebrand factor, thrombosis, and the endothelium: The first 30 years. Thromb Haemost 2006; 95: 49–55.

84.

McCrary

Nolasco

Hellums

. Direct demonstration of radiolabeled von Willebrand factor binding to platelet glycoprotein Ib and IIb-IIIa in the presence of shear stress. Ann Biomed Eng 1995; 23: 787–793.

85.

Bongers

de Maat

van Goor

. High von Willebrand factor levels increase the risk of first ischemic stroke: Influence of ADAMTS13, inflammation, and genetic variability. Stroke 2006; 37: 2672–2677.

86.

Spencer

Felmeden

Blann

. Haemorheological, platelet and endothelial indices in relation to global measures of cardiovascular risk in hypertensive patients: A substudy of the Anglo-Scandinavian Cardiac Outcomes Trial. J Intern Med 2007; 261: 82–90.

87.

Qizilbash

Duffy

Prentice

. Von Willebrand factor and risk of ischemic stroke. Neurology 1997; 49: 1552–1556.

88.

Folsom

Rosamond

Shahar

. Atherosclerosis Risk in Communities (ARIC) Study Investigators. Prospective study of markers of hemostatic function with risk of ischemic stroke. Circulation 1999; 100: 736–742.

89.

Cesar

García-Avello

Vecino

. Increased levels of plasma von Willebrand factor in migraine crisis. Acta Neurol Scand 1995; 91: 412– 413.

90.

Tietjen

Al Qasmi

Athanas

. Increased von Willebrand factor in migraine. Neurology 2001; 57: 334–336.

91.

Tietjen

Al Qasmi

Athanas

. Altered hemostasis in migraineurs studied with a dynamic flow system. Thromb Res 2007; 119: 217–222.

92.

Kirtava

Drews

Lally

. Medical, reproductive and psychosocial experiences of women diagnosed with von Willebrand’s disease receiving care in haemophilia treatment centres: A case-control study. Haemophilia 2003; 9: 292–297.

93.

Kannel

. Influence of fibrinogen on cardiovascular disease. Drugs 1997; 54(Suppl 3): 32– 40.

94.

Fisher

Meiselman

. Hemorheological factors in cerebral ischemia. Stroke 1991; 22: 1164–1169.

95.

Iwamoto

Kubo

Takasaki

. Platelet activation in the cerebral circulation in different subtypes of ischemic stroke and Binswanger’s disease. Stroke 1995; 26: 52–56.

96.

Smith

Keen

Grant

. Fate of fibrinogen in human arterial intima. Arteriosclerosis 1990; 10: 263–275.

97.

Moons

Bots

Salonen

. Prediction of stroke in the general population in Europe (EUROSTROKE): Is there a role for fibrinogen and electrocardiography? J Epidemiol Community Health 2002; 56(Suppl 1): i30–i36.

98.

Kannel

Wolf

Castelli

D’Agostino

. Fibrinogen and risk of cardiovascular disease: The Framingham Study. JAMA 1987; 258: 1183–1186.

99.

Danesh

Lewington

. Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: An individual participant meta-analysis. JAMA 2005; 294: 1799–1809.

100.

Tsuda

Satoh

Kitadai

. Hemorheologic profiles of plasma fibrinogen and blood viscosity from silent to acute and chronic cerebral infarctions. J Neurol Sci 1997; 147: 49–54.

101.

Bianchi

Pitari

Amenta

. Endothelial, haemostatic and haemorheological modifications in migraineurs. Artery 1996; 22: 93–100.

102.

Walkowiak

Kozubski

Pawlowska

. Expression of fibrinogen receptors in platelets of migraine patients – correlation with platelet GPIIb content and plasma cholesterol. Thromb Haemost 1989; 61: 419–422.

103.

Yucel

Tanriverdi

Arıkanoglu

. Increased fibrinogen, D-dimer and galectin-3 levels in patients with migraine. Neurol Sci 2014; 35: 545–549.

104.

Macko

Kittner

Epstein

. Elevated tissue plasminogen activator antigen and stroke risk: The Stroke Prevention In Young Women study. Stroke 1999; 30: 7–11.

105.

Isobe

Terayama

. A remarkable increase in total homocysteine concentrations in the CSF of migraine patients with aura. Headache 2010; 50: 1561–1569.

106.

Bahadir

Eroz

Dikici

. Investigation of MTHFR C677T gene polymorphism, biochemical and clinical parameters in Turkish migraine patients: Association with allodynia and fatigue. Cell Mol Neurobiol 2013; 33: 1055–1063.

107.

Gavgani

Hoseinian

. Comparative study on homocysteine levels in migraine patients and normal peoples. Ann Bio Res 2012; 3: 1804–1807.

108.

Moschiano

D’Amico

Usai

. Homocysteine plasma levels in patients with migraine with aura. Neurol Sci 2008; 29(Suppl 1): S173–S175.

109.

Hering-Hanit

Gadoth

Yavetz

. Is blood homocysteine elevated in migraine? Headache 2001; 41: 779–781.

110.

Liman T, Bachelier-Walenta K, Neeb L, et al. Circulating endothelial microparticles in female migraineurs with aura. Cephalalgia. 2015; 35: 88–94.

111.

Sibal

Agarwal

Home

. The role of asymmetric dimethylarginine (ADMA) in endothelial dysfunction and cardiovascular disease. Curr Cardiol Rev 2010; 6: 82–90.

112.

Guldiken

Demir

Guldiken

. Asymmetric dimethylarginine and nitric oxide levels in migraine during the interictal period. J Clin Neurosci 2009; 16: 672–674.

113.

Uzar

Evliyaoglu

Toprak

. Increased asymmetric dimethylarginine and nitric oxide levels in patients with migraine. J Headache Pain 2011; 12: 239–243.

114.

Gallai

Sarchielli

Firenze

. Endothelin-1 in migraine and tension-type headache. Acta Neurol Scand 1994; 89: 47–55.

115.

Komatsumoto

Nara

. Lower level of endothelin-1 in migraine with aura [in Japanese]. Rinsho Shinkeigaku 1995; 35: 1250–1252.

116.

Hasselblatt

Köhler

Volles

. Simultaneous monitoring of endothelin-1 and vasopressin plasma levels in migraine. Neuroreport 1999; 10: 423–425.

117.

Kallela

Färkkilä

Saijonmaa

. Endothelin in migraine patients. Cephalalgia 1998; 18: 329–332.

118.

Färkkilä

Palo

Saijonmaa

. Raised plasma endothelin during migraine attack. Cephalalgia 1992; 12: 383–384.

119.

Nattero

Mengozzi

Inconis

. Nitric oxide, endothelin-1, and transcranial Doppler in migraine. Findings in interictal conditions and during migraine attack. Headache 1996; 36: 307–311.

120.

MacClellan

Howard

Cole

. Relation of candidate genes that encode for endothelial function to migraine and stroke: The Stroke Prevention in Young Women study. Stroke 2009; 40: e550–e557.

121.

Lee

Chu

Jung

. Decreased number and function of endothelial progenitor cells in patients with migraine. Neurology 2008; 70: 1510–1517.

122.

Rodríguez-Osorio

Sobrino

Brea

. Endothelial progenitor cells: A new key for endothelial dysfunction in migraine. Neurology 2012; 79: 474–479.

123.

Oterino

Toriello

Palacio

. Analysis of endothelial precursor cells in chronic migraine: A case-control study. Cephalalgia 2013; 33: 236–244.

124.

Sacco

Ripa

Grassi

. Peripheral vascular dysfunction in migraine: A review. J Headache Pain 2013; 14: 80–80.

125.

Pezzini

Grassi

Lodigiani

. Predictors of migraine subtypes in young adults with ischemic stroke: The Italian Project on Stroke in Young Adults. Stroke 2011; 42: 17–21.

126.

MacClellan

Giles

Cole

. Probable migraine with visual aura and risk of ischemic stroke: The Stroke Prevention in Young Women study. Stroke 2007; 38: 2438–2445.

127.

Schrader

Stovner

Helde

. Prophylactic treatment of migraine with angiotensin converting enzyme inhibitor (lisinopril): Randomised, placebo controlled, crossover study. BMJ 2001; 322: 19–22.

128.

Tronvik

Stovner

Helde

. Prophylactic treatment of migraine with an angiotensin II receptor blocker: A randomized controlled trial. JAMA 2003; 289: 65–69.

129.

Kurth

Diener

Buring

. Migraine and cardiovascular disease in women and the role of aspirin: Subgroup analyses in the Women’s Health Study. Cephalalgia 2011; 31: 1106–1115.

Vascular biomarkers in migraine

Abstract

Aim

Methods

Results

Conclusion

Keywords

Introduction

Methods

Results

Biomarkers of inflammation

CRP

Cytokines

TNF-α

Interictal studies

Ictal studies

IL-6

Interictal studies

Ictal studies

Interleukin-1β (IL-1β)

Interictal studies

Ictal studies

IL-10

Interictal

Ictal

Other interleukins

TGF-β1

Interictal studies

Adiponectin

Lipids

Biomarkers of thrombosis and coagulation

Platelet activation

vWF

Fibrinogen

tPA

Homocysteine

Biomarkers of endothelial activation and dysfunction

EMPs

ADMA

ET-1

Interictal studies

Ictal studies

Pro-brain natruretic peptide (Pro-BNP)

Biomarkers of endothelial regeneration and repair

EPCs

Discussion

Summary of findings

Implications

Mechanistic

Clinical

Limitations of this review

Recommendations for future studies

Clinical implications

Footnotes

Funding

Conflict of interest

References