Conventional vascular risk factors: Their role in the association between migraine and cardiovascular diseases

Abstract

Introduction

Migraine, in particular migraine with aura, has been found to be associated with cardiovascular disease. However, the role of conventional vascular risk factors in the association is still debated. The aim of the present review is to address the association between migraine and conventional cardiovascular risk factors as well as to address their possible role in the association between migraine and cardiovascular disease.

Methods

Data for this review were obtained through searches in multiple sources up to May 2014 using the terms “migraine” OR “headache” in combination with all the vascular risk factors of interest.

Results

Data about the possible association between migraine and high blood pressure values are heterogeneous, hindering any final conclusion. Data addressing the possible association between migraine and diabetes mellitus indicate the lack of any association or in some cases a negative association between the two conditions. The body of evidence on the role of dyslipidemia in migraineurs is relatively homogeneous and, with few exceptions, reports an association between migraine and an unfavorable lipid profile; however, the difference in lipid levels between migraineurs and non-migraineurs is small and its clinical implication unclear. Regarding obesity, a trend has been observed of increased risk of migraine with increasing obesity, especially in young patients, albeit in the midst of conflicting data. Evidence about the association between cigarette smoking and migraine mostly indicates that migraineurs are more commonly smokers than non-migraineurs. On the other hand, the majority of the available studies report less alcohol use in migraineurs than in non-migraineurs. Finally, many of the available studies suggest a more frequent family history of cardiovascular disease in migraineurs as compared to non-migraineurs. Since most of the studies that supported the association between migraine and cardiovascular disease adjusted the analyses for the presence of several vascular risk factors, they cannot entirely explain this association.

Conclusions

Based on the available reported data, it seems unlikely that the higher risk of cardiovascular disease in migraineurs is mediated by any single vascular risk factor. For this reason the role of specific interactions among risk factors with the contribution of genetic, environmental, personality and psychological factors should be appropriately investigated.

Keywords

Migraine vascular disease arterial hypertension diabetes mellitus dyslipidemia cigarette smoking alcohol cardiovascular disease

Introduction

Migraine, in particular migraine with aura (MA), has been found to be associated with cardiovascular disease (CVD) (1 –19). Mechanisms underlying the hypothesized causal relationship between migraine and CVD have not yet been established. It is improbable that CVD occurs exclusively through the conventional mechanisms commonly recognized in the general population (e.g. atherosclerosis and small vessel disease); migraine-specific mechanisms are probably involved. Additionally, whether ischemic stroke and ischemic heart disease share common mechanisms in migraineurs is still unproven. Several studies have addressed the burden of conventional vascular risk factors (RFs) in migraineurs. However, in some areas results from the available studies are conflicting and definitive conclusions cannot be drawn. In this review we will examine the association between migraine and conventional vascular RFs such as arterial hypertension, diabetes mellitus (DM), dyslipidemia, obesity, cigarette smoking, alcohol consumption, and family history of CVD; additionally we will briefly address their possible role in the association between migraine and CVD.

Methods

Data for this review were obtained through searches in Embase, Google Scholar, ISI Web of Science, and Medline from their first availability up to May 2014. The search terms included “migraine” OR “headache” AND (“arterial hypertension,” “diabetes mellitus,” “dyslipidemia,” “hypercholesterolemia,” “obesity,” “body mass index,” “cigarette smoking,” “alcohol,” “history of cardiovascular disease”). We also searched reference lists and papers cited in the identified articles. We considered only studies addressing, as their primary or secondary aim, the difference in the distribution of vascular RFs between migraineurs and non-migraineurs. We excluded studies using as the reference group participants with non-migraine headache, not defining migraine or the RF of interest, performed on individuals with metabolic disease only, or not adopting a matching procedure or adjusted analyses to account for possible confounders. When studies provided more than one adjusted model, we used the one including the largest number of factors. Where a cohort study analyzed the association between migraine and the RF of interest considering only baseline data, we have reported the study as cross-sectional in our results. We reviewed only studies published in English.

Results

Blood pressure (BP)

Several studies have attempted to investigate the association between migraine and BP values, with contradictory results (details on sample size, demographic characteristics of included participants, and findings are shown in Table 1) (6,20 –32). An association between headache, including migraine, and high BP values has been found by some studies (6,23 –26,30). Other studies have pointed to the lack of a clear association between migraine and increased BP values (20 –22,28) and some have demonstrated an inverse relationship between migraine and high BP values (20,21,27,29,31,32).

Table 1.

Summary of selected studies addressing the relationship between migraine and blood pressure.

Study (first author, year; acronym)	Design	Sample size (n)	Age range (years)	Women (%)	Matching procedure or adjustment for confounders	Findings
Benseñor, 2011; Saõ Paulo Ageing & Health Study (SPAH) (20)	Cross-sectional	1,450 subjects; 165 migraineurs	≥65	78 migraineurs 56 non-migraineurs	OR adjusted for age, gender, body mass index, smoking status, alcohol consumption, monthly income	Similar SBP (140 vs 146 mmHg) and DBP (84 vs 86 mmHg) values, and similar prevalence of arterial hypertension (75% vs 80%) in migraineurs and non-migraineurs; no increased risk of hypertension in migraineurs, inverse association in women (OR 0.58, 95% CI 0.38–0.90)
Gudmundsson, 2005; Reykjavik Study (21)	Cross-sectional*	21,537 subjects; 2,116 migraineurs	19–87	52	OR adjusted for age and year entering the study	No increased risk of migraine associated with arterial hypertension in men and women. Decreased risk of migraine in women with SBP ≥160 mmHg (OR 0.76, 95% CI 0.62–0.92)
Hagen, 2002; Nord-Trøndelag Health Study (HUNT) study (22)	Cohort, prospective	22,685 subjects; 1,648 migraineurs	≥20	NR	OR adjusted for age, occupational status, and use of antihypertensive medication	No increased risk of migraine in women and men with high (≥160 mmHg) SBP and high (≥90 mmHg) DBP
Ikeda, 2012 (23)	Case-control	221 subjects; 111 migraineurs	NR	69	Age and gender matching	Higher SBP and DBP values in women (109 vs 107 mmHg P<0.01 and 69 vs 66 mmHg; P<0.01) and men (120 vs 116 mmHg P<0.01 and 74 vs 71 mmHg; P<0.01) with migraine than in non-migraineurs
Kuo, 2013; Taiwan National Health Insurance Research Database (6)	Cross-sectional*	125,550 subjects; 20,925 migraineurs	≥18	70	Age and gender matching	Hypertension more common in migraineurs than in non-migraineurs (16% vs 12%; P<0.0001)
Merikangas, 1997; National Health and Nutrition Examination Survey (NHANES) (24)	Cohort, prospective	12,220 subjects; 1,109 migraineurs	25–74	60	OR adjusted for age and gender	Increased risk of migraine in subjects with arterial hypertension (OR 1.7, 95% CI 1.5–2.0)
Scher, 2005; Genetic Epidemiology of Migraine (GEM) study (25)	Cross-sectional	5,755 subjects; 620 migraineurs	20–65	52	OR adjusted for sex, gender, socioeconomic status, smoking status, and alcohol intake	Arterial hypertension more likely in migraineurs overall (OR 1.64, 95% CI 1.3–2.0), in MA (OR 1.73, 95% CI 1.2–2.4) and in MO (OR 1.62, 95% CI 1.3–2.0) than in non-migraineurs; high BP values (SBP ≥140 mmHg or DBP ≥90 mm Hg) at examination similar in migraineurs and non-migraineurs
Schwaiger, 2008; Bruneck Study (26)	Cross-sectional*	574 subjects; 111 migraineurs	55–94	29	Adjusted for age and gender	Hypertension more common in migraineurs than in non-migraineurs (72% vs 68%)
Tronvik, 2008; Nord-Trøndelag Health Study (HUNT-1 and HUNT-2) study (27)	Cross-sectional (HUNT-1) Cohort prospective (HUNT-2)	Prospective analysis: 37,995 subjects; 17,536 migraineurs Cross-sectional analysis: 77,066 subjects; 35,020 migraineurs	≥20	NR	OR adjusted for age and years of education	Prospective analysis: linear trend (p<0.01) of decreasing migraine prevalence with increasing SBP in women who had not used antihypertensives; no trend for males and for females using antihypertensives Cross-sectional analysis: trend of increasing migraine prevalence with increasing DBP values in migraineurs not using antihypertensives
Tronvik, 2011; Nord-Trøndelag Health (HUNT- Youth) Study (28)	Cross-sectional	5,847 subjects; 341 migraineurs	13–18	52	OR adjusted for age and weight	No increased risk of migraine according to SBP values groups
Tzourio, 2003; Epidemiology of Vascular Ageing (EVA) study (29)	Cross-sectional	1,373 subjects; 140 migraineurs	59–71	54 migraineurs 86 non-migraineurs	OR adjusted for age, gender, alcohol consumption, and history of depression	Decreasing risk of migraine with increasing SBD (OR 0.62, 95% CI 0.39-1.00 for 120-139 mmHg group and OR 0.44 95% CI 0.27-0.75 for ≥140 group vs 1<20 group; P for trend<0.001)
Wang, 2014; Taiwan National Health Insurance Research Database (30)	Cross-sectional*	23,082 subjects; 11,541 migraineurs	18–45	71	Age and gender matching	Hypertension more common in migraineurs than in non-migraineurs (3% vs 2%; P<0.0001)
Wiehe, 2002 (31)	Cross-sectional	1,174 subjects; 201 migraineurs	≥17	56	Risk ratio of migraine adjusted for age, gender, schooling, income, race, alcohol consumption, physical activity at leisure time, and depression	Reduced migraine risk in subjects with high normal BP to severe hypertension than in subjects with optimal or normal BP (adjusted risk ratio 0.56, 95% CI 0.41–0.77)
Winswold, 2010; Nord-Trøndelag Health (HUNT-2) study (32)	Cross-sectional	48,713 subjects; 3,412 migraineurs	≥20	62 non-migraineurs 76 MA 74 MO	Adjusted for age, gender, and education level	Lower SBP in subjects with MA (135 vs 137; P<0.01) and MO (136 vs 137; P<0.01) than controls; lower DBP (81 vs 80 mmHg; P<0.001) in subjects with MO than in controls

BP=blood pressure, CI=confidence interval, DBP= diastolic blood pressure, MA=migraine with aura, MO=migraine without aura, NR=not reported, OR=odds ratio, SBP=systolic blood pressure.

Cohort studies reporting the association in a cross sectional analysis.

The Genetic Epidemiology of Migraine (GEM), a large, well-conducted cross-sectional study, reported that individuals with migraine had a 1.64 increase in the risk of presenting a history of physician-diagnosed hypertension (25). This finding was consistent both in patients with MA and with migraine without aura (MO). In line with this observation, the National Health and Nutrition Examination Survey (NHANES), a prospective cohort study involving 12,220 participants, reported an increased risk of migraine in individuals with arterial hypertension; no details were provided regarding MA or MO (24). The prospective cohort Bruneck study also indicated, in an analysis of cross-sectional data, that hypertension was more common in migraineurs than in controls (26). Additionally, two cross-sectional studies in Taiwan using administrative coding data supported this same observation (6,30). In line with those findings, a case-control study on Japanese patients, showed higher values for both systolic BP (SBP) and diastolic BP (DBP) in migraineurs as compared to non-migraineurs, regardless of migraine subtype (23).

On the other hand, some studies did not find a clear association between the presence of hypertension and migraine, or indeed indicated the presence of an inverse relationship (Table 1). Several reports from the Nord-Trøndelag Health (HUNT) Study, a prospective cohort study involving a large number of participants from adolescence to adulthood, indicated in cross-sectional analyses that individuals with hypertension or high BP values were not at increased risk of migraine (22,27,28). Similarly, a cross-sectional population-based study, the Saõ Paulo Ageing & Health (SPAH) Study, did not find an increased risk of hypertension in migraineurs but suggested the presence of an inverse association in women (20). This last finding is in line with those reported by the large cross-sectional Reykjavik Study (21), by cross-sectional data from the prospective Epidemiology of Vascular Aging (EVA) Study (29), and by some reports from a cross-sectional Brazilian study (31) and from the HUNT-2 study (32) suggesting a decreased risk of migraine associated with high BP values (31) or lower mean BP values in migraineurs (32).

Diabetes Mellitus (DM)

Studies addressing the association between migraine and DM (details on sample size, demographic characteristics of included participants, and findings are shown in Table 2) have reported either no association (6,20,23,24,26,30,33) or an inverse (34 –36) association between the two conditions.

Table 2.

Summary of selected studies addressing the relationship between migraine and diabetes mellitus.

Study (first author, year; acronym)	Design	Sample size (n)	Age range (years)	Women (%)	Matching procedure or adjustment for confounders	Findings
Aamodt, 2007; Nord-Trøndelag Health (HUNT) study (34)	Cross-sectional	51,383 subjects; 6,209 migraineurs	≥20	54 no DM 50 all DM types	OR adjusted for age, gender, and educational level	Decreased risk of migraine in subjects with DM1 (OR 0.4, 95% CI 0.2–0.9) and DM2 (OR 0.7, 95% CI 0.5–0.9) than in non-diabetics
Benseñor, 2011; (20) Saõ Paulo Ageing & Health Study (SPAH)	Cross-sectional	1,450 subjects; 165 migraineurs	≥65	78 migraineurs 56 non-migraineurs	OR adjusted for age, sex, body mass index, smoking status, alcohol consumption, monthly income	Similar prevalence of DM in migraineurs and non-migraineurs (27 vs 23%); no increased risk of DM in migraineurs
Berge, 2013 (35)	Cross-sectional	4,640,219 subjects; 81,225 migraineurs	≥20	49	OR adjusted for age and gender	Lower risk of receiving migraine drugs in patients with DM than in non-DM subjects (OR 0.7, 95% CI 0.68–0.75)
Burch, 2012; Women’s Health Study (WHS) (33)	Cohort, prospective	38,620 subjects; 5,062 migraineurs	≥45	100	HR adjusted for age, BMI, exercise, ever used postmenopausal hormones of any type, postmenopausal status, randomized treatment assignment to aspirin and vitamin E, alcohol consumption, history of hypertension ≥140/90 mmHg, baseline treatment with medication for high BP, history of high cholesterol ≥240 mg/dl, baseline treatment with cholesterol-lowering medication, smoking status, SBP, family history of DM, and ever used oral contraceptives for ≥2 months.	No association between any migraine and incident DM; no association between MA or MO and incident DM
Burn, 1984 (36)	Case-control	850 subjects; 188 migraineurs	≥10	51	Adjusted for age and gender	Lower migraine prevalence in patients with DM than in non-diabetics (17 vs 29%; P<0.05)
Ikeda, 2012 (23)	Case-control	1,195 subjects; 596 migraineurs	NR	69	Age and gender matching	No differences in fasting blood sugar (93.5 vs 92.6 mg/dL in women and 99.9 vs 99.6 mg/dL in men) and hemoglobin A_1C (5.3 vs 5.3% in women and 5.3 vs 5.4% in men) between migraineurs and non-migraineurs
Kuo, 2013; Taiwan National Health Insurance Database (6)	Cross-sectional*	125,550 subjects; 20,925 migraineurs	≥18	70	Age and gender matching	Similar prevalence of DM in migraineurs and non-migraineurs (6.0 vs 5.9%)
Merikangas, 1997; National Health and Nutrition Examination Survey (NHANES) (24)	Cohort, prospective	12,220 subjects; 1,109 migraineurs	25–74	60	OR adjusted for age and sex	No increased risk of migraine in subjects with DM than in non-diabetics
Schwaiger, 2008; Bruneck Study (26)	Cross-sectional*	574 subjects; 111 migraineurs	55–94	29 migraineurs	Adjusted for age and gender	No difference in DM prevalence among subjects with MA (11.1%), MO (9.3%) or non-migraineurs
Wang, 2014; Taiwan National Health Insurance Research Database (30)	Cross-sectional*	23,082 subjects; 11,541 migraineurs	18–45	71	Age and gender matching	Similar prevalence of DM in migraineurs than in non-migraineurs (1.00 vs 0.86%; P=0.2802)

BMI= body mass index, BP= blood pressure, CI=confidence interval, DM=diabetes mellitus, DM1=diabetes mellitus type 1, DM2=diabetes mellitus type 2, HR=hazard ratio, MA=migraine with aura, MO=migraine without aura, NR=not reported, OR=odds ratio, SBP= systolic blood pressure.

Cohort studies reporting the association in a cross sectional analysis.

Data from the NHANES (24) and the SPAH study (20) indicated that migraineurs were not at increased risk of DM nor were diabetics at increased risk of migraine. Similarly, the analysis of data from the Women’s Health Study (WHS) cohort of women without DM at baseline showed that patients with a past history of migraine, active MA or active MO were not at increased risk of developing type 2 DM as compared to non-migrainous women (33). In addition, a similar prevalence of DM in migraineurs and non-migraineurs has been reported by the cross-sectional Bruneck study (26) and by the two retrospective cross-sectional studies in Taiwan using administrative coding data (6,30). The case-control study on Japanese subjects showed no differences in indicators of DM (fasting blood sugar and hemoglobin A_1C) between migraineurs and non-migraineurs (23).

At variance with this, some data indicate a negative association between migraine and DM. A cross-sectional analysis of data from the prospective cohort HUNT study found that both patients with type 1 and type 2 DM were less likely to report a previous history of migraine than patients without DM (34). A lower frequency of migraine in patients with DM as compared to patients without DM was also reported by an early case-control study (36). Similar conclusions were drawn by a Norwegian cross-sectional study based on the analysis of data related to drug prescription (35). The study found that patients with DM had an overall reduced prevalence of medically treated migraine as compared to the non-diabetic population, and that the trend was progressively more pronounced with advancing age.

Dyslipidemia

Several studies have investigated the association between migraine and unfavorable lipid profiles (details on sample size, demographic characteristics of included participants, and findings are shown in Table 3) (6,23,25,26,30,32,37 –41). The majority of these studies found an association between the two conditions (6,23,25,30,37 –39,41) in the presence of some conflicting findings (26,32,40).

Table 3.

Summary of selected studies addressing the relationship between migraine and dyslipidemia.

Study (first author, year; acronym)	Design	Sample size (n)	Age range (years)	Women (%)	Matching procedure or adjustment for confounders	Findings
Chen, 2012; Taiwan National Health Insurance Research Database (37)	Cross-sectional	4,738 subjects; 948 CM	None	75.3	OR adjusted for age, gender, urbanization level of the residence and income	Increased risk of hyperlipemia in subjects with CM compared with non-migraineurs (OR 1.84, 95% CI 1.45–2.32)
Ikeda, 2012 (23)	Case-control	221 subjects; 111 migraineurs	NR	69	Age and gender matching	Higher mean TC levels (mg/dL) in female (210.3 vs 201.5; P<0.01) and male (209.3 vs 202.0; P<0.01) migraineurs than non-migraineurs. Higher mean TC levels (mg/dL) in MA (210.5) and MO (209.3) than non-migraineurs (201.5; P<0.001). Lower mean HDL-C levels (mg/dL) in female (67.4 vs 71.8; P<0.001) and male (53.5 vs 56.5; P<0.01) migraineurs than non-migraineurs. Similar mean HDL-C levels (mg/dL) in MA (62.9) and MO (63.2) than non-migraineurs (64.4).Higher mean LDL-C levels (mg/dL) in female (118.9 vs 109.6; P<0.01) and male (129.8 vs 121.5; P<0.01) migraineurs than non-migraineurs. Higher mean LDL-C levels (mg/dL) in MA (125.9) and MO (123.8) than non-migraineurs (115.4; P<0.001).
Kuo, 2013; Taiwan National Health Insurance Database (6)	Cross-sectional*	125,550 subjects; 20,925 migraineurs	≥18	70	Age and gender matching	Higher prevalence of dyslipidemia (ICD-9-CM code 272) in migraineurs than in non-migraineurs (8.0 vs 5.4%; P<0.0001)
Kurth, 2008; Women’s Health Study (WHS) (38)	Cross-sectional*	27,626 subjects; 5,087 migraineurs	≥45	100	Adjusted for age, body mass index, smoking, systolic blood pressure, antihypertensive treatment, physical activity, alcohol consumption, history of diabetes, menopausal status, postmenopausal hormone use, history of oral contraceptive use and family history of myocardial infarction prior to age 60	Higher risk of elevated TC (OR 1.09, 95% CI 1.01-1.18) and elevated non-HDL-C (OR 1.14, 95% CI 1.05-1.23) in women with migraine than in non-migraineurs
Rist, 2011; Epidemiology of Vascular Ageing (EVA) Study (39)	Cross-sectional	1,155 subjects; 166 migraineurs	61–69	82 migraineurs; 54 non-migraineurs	OR adjusted for age, gender, smoking status, ever cholesterol medication use, alcohol consumption, systolic blood pressure, diastolic blood pressure and BMI	High risk of MA vs non headache in subjects in the 3rd tertile of total cholesterol (RR 5.97, 95% CI 1.29–27.61). High risk of MA vs non headache in subjects in the 3nd tertile of triglycerides (RR 4.42, 95% CI 1.32–14.77)
Rockett, 2013 (40)	Case-control	59; 30 migraineurs	18–55	100	Age and BMI matching	Lower HDL-C levels in obese migraineurs (42.5±11.5) than normal weight controls (54.9±11.9), obese controls (46.7±14.0) and normal weight migraineurs (54.8±13.3)
Saberi, 2011 (41)	Case-control	205 subjects; 102 migraineurs	None	80	Age, gender, hypertension, and diabetes mellitus matching	Increased risk of hypertriglyceridemia (OR 3.09, 95% CI: 1.6- 5.8), hypercholesterolemia (OR 5.29, 95% CI 2.4-11.4), low HDL-C (OR 0.28, 95% CI 0.13-0.6),high LDL-C (OR 2.43, 95% CI 1.1-5.3) in migraineurs than in non-migraineurs
Scher, 2005; Genetic Epidemiology of Migraine (GEM) study (25)	Cross-sectional	5,755 subjects; 620 migraineurs	20–65	52	OR adjusted for sex, age, socioeconomic status, smoking status, and alcohol intake	Higher mean values of TC in migraineurs (204 mg/DL; P<0.05) and MA (206; P<0.05) but not in MO (202) than in non-migraineurs (201). Lower mean values of HDL-C in MA (50 mg/DL; P<0.05) but not in all migraineurs (51) or MO (52) than in non-migraineurs (52). In the adjusted analysis only an increased risk of TC >240 mg/dL in MA women (adjusted OR 1.8, 95% CI 1.1–2.7)
Schwaiger, 2008; Bruneck Study(26)	Cross-sectional*	574 subjects; 111 migraineurs	55–94	29 migraineurs	Adjusted for age and gender	No differences in mean TC (6.01 vs 6.02 vs 5.86 mmol/L), HDL-C (1.69 vs 1.66 vs 1.66 mmol/L) and LDL-C (3.63 vs 3.67 vs 3.59 mmol/L) among MA, MO and non-migraineurs
Wang, 2014; Taiwan National Health Insurance Database (30)	Cross-sectional*	23,082 subjects; 11,541 migraineurs	18–45	71	Age and gender matching	Higher prevalence of dyslipidemia (ICD-9-CM code 272) in migraineurs than in non-migraineurs (1.78 vs 0.95%; P<0.0001)
Winswold, 2011; Nord-Trøndelag Health (HUNT) Study (32)	Cross-sectional	48,713 subjects; 3,412 migraineurs	≥20	62 non-migraineurs 76 MA 74 MO	Adjusted for age, gender, and education level	High values of TC (≥7.5 mmol/L) associated with decreased risk of MO (OR 0.73, 95% CI 0.60-0.89) but not of MA than lower values (<5.2 mmol/L); low values of TC (<1.1 mmol/L) associated with increased risk of MO (OR 1.32, 95% CI 1.15-1.51) and of MA (OR 1.77, 95% CI 1.35-2.32) than higher values (≥1.6 mmol/L)

BMI= body mass index, CI=confidence interval, CM= chronic migraine; HDL-C=high density lipoprotein cholesterol, ICD-9-CM=International Classification of Diseases 9^th revision Clinical Modification, TC=total cholesterol, LDL-C=low density lipoprotein cholesterol, MA=migraine with aura, MO=migraine without aura, NR=not reported, OR=odds ratio, RR=relative risk.

Cohort studies reporting the association in a cross sectional analysis.

The cross-sectional GEM study demonstrated that patients with MA were more likely to have high total cholesterol and high total/high-density lipoprotein (HDL) cholesterol ratio as compared to other groups while there was no association between high total cholesterol and MO (25). However, in this study, the difference in mean blood values of total cholesterol between patients with MA and the other groups was modest and the clinical relevance of this small difference needs to be clarified. Cross-sectional data from the WHS cohort supported these findings indicating an increased risk of having higher level of total cholesterol and non-HDL cholesterol in participants with migraine than in non-migraineurs; in this study the analysis performed according to migraine type was inconclusive (38). Data from the cross-sectional EVA study also indicated an association between migraine and dyslipidemia; individuals with high total cholesterol and high triglyceride levels had an increased risk of MA but not of MO (39). Further evidence for an association between migraine and dyslipidemia came from a case-control study on 205 patients; migraineurs as compared to non-migraineurs had an increased risk of hypertriglyceridemia, hypercholesterolemia, low HDL cholesterol, and high low-density lipoprotein (LDL) cholesterol (41). Three studies on Taiwanese individuals, all with a retrospective design and using administrative coding data, recognized a higher prevalence of dyslipidemia in migraineurs as compared to non-migraineurs (6,30) or an increased risk of hyperlipidemia in patients with chronic migraine than non-migraineurs (37), while the previously mentioned case-control study of Japanese participants found that migraineurs had increased values of total cholesterol and LDL cholesterol and decreased values of HDL cholesterol, without any difference between migraine subtypes (23).

Some data have pointed to a negative association between migraine and dyslipidemia or to the lack of any association. In the HUNT study, lower levels of both total and HDL cholesterol values were reported, especially in MA (32). The multivariate analysis showed an inverse association between high total cholesterol values and MO but not MA, along with a positive association between high HDL cholesterol levels and MA or MO. A negative association between migraine and dyslipidemia was reported also by a case-control study with a small sample size (40). This study reported results stratified for obesity status, making comparison with other studies difficult. The Bruneck study reported no differences in mean values of blood lipids between migraineurs and non-migraineurs (26).

Obesity

Numerous studies addressing the possible association between migraine and obesity have provided conflicting results (details on sample size, demographic characteristics of included participants, and findings are shown in Table 4) (23 –26,32,42 –62) but many of the available studies suggest an association between migraine and obesity with a greater risk of migraine chronification with increasing obesity status (32,42 –47,54,57,62). Moreover, some studies have highlighted the presence of an association between migraine and obesity more in younger than in older individuals (45,51,52,54). A cross-sectional analysis of data of the Australian Longitudinal Study on Women’s Health found a greater risk of migraine in obese than in non-obese women (45). Data from the cross-sectional American Migraine Prevalence and Prevention (AMPP) study supported the increased risk of migraine in obese and morbidly obese subjects when compared to non-obese (44). Cross-sectional data of the Romany National Health Survey in Spain also indicated an increased risk of migraine associated with obesity (47). Further evidence for the above relationship came from a large cross-sectional study on Asian individuals showing a more than twofold increase in the odds of migraine in participants with severe obesity but no significant increase in migraine risk in obese and overweight individuals (58). Finally, a general population, cross-sectional analysis reported an increased risk of migraine in obese individuals as compared to those of normal weight with a trend of increasing odds of episodic migraine with increasing obesity status from normal weight to overweight to obese especially in participants younger than 50 years of age (52). Those findings were supported by a cross-sectional analysis of data from the HUNT study that also reported that the risk both of MA and MO was greater in obese than non-obese patients (32). Data from the cross-sectional University of Washington Twin Registry showed an increased risk of headache including migraine in obese compared to normal-weight participants but no significant increase in the risk in overweight compared to normal-weight individuals (62). Two other cross-sectional analyses conducted with NHANES participants demonstrated an increased risk of migraine or severe headache in obese patients as compared to those of normal weight (46,51), with an association in younger but not in older individuals (51). The presence of an age-dependent association has been further confirmed by a study showing an increased risk of migraine in overweight adolescents (54) and by the cross-sectional Omega study recognizing an increased risk of migraine with increasing body mass index (BMI) in reproductive age women (57). Other data from a large cross-sectional study indicated either no increased risk of migraine associated with obesity but a greater risk of chronification in patients with higher BMI as compared to participants of normal weight (42) or an increased risk of chronic daily headache in the obese and morbidly obese compared to normal-weight individuals (43). In a cross-sectional analysis of the WHS cohort, there was a higher prevalence of migraine in obese individuals, but this association disappeared after adjustment for cardiovascular factors and postmenopausal status (59). However, in that study, increasing BMI among migraineurs was associated with increased frequency of headaches (59). An analysis of prospective data in the same cohort found an increased risk of becoming overweight but not obese over time in migraineurs than in non-migraineurs (61).

Table 4.

Summary of selected studies addressing the relationship between migraine and obesity.

Study (first author, year; acronym)	Design	Sample size (n)	Age range (years)	Women (%)	Matching procedure or adjustment for confounders	Findings
Bigal, 2006 (42)	Cross-sectional	30,215; 3,791 migraineurs	≥18	65	Risk of migraine stratified by gender and adjusted for marital status, income, use of daily medications, and depression scores; unadjusted risk of chronification	No increased risk of migraine in overweight (BMI 25-29.9), obese (BMI 30-34.9), and morbidly obese subjects (BMI ≥35) than in normal weight subjects (BMI 18.5-24.9) Greater risk of chronification in patients with higher BMI (OR 1.5, 95% CI 1.2-1.8 for obese; OR 2.0, 95% CI 1.4-2.4 for morbidly obese)
Bigal and Lipton, 2006 (43)	Cross-sectional	30,215; 3,791 migraineurs	18–89	62	Adjusted for age, race, use of acute or preventive medication, socioeconomic status, and marital status	Increased risk of CDH in the obese (OR 1.5, 95% CI 1.2-1.8) and morbidly obese (OR 2.0, 95% CI 1.4-2.4) subjects than in normal weighted (same categories of BMI of Bigal, 2006); BMI associated with a dignosis of TM
Bigal, 2007; American Migraine Prevalence and Prevention (AMPP) Study (44)	Cross-sectional	162,576; 18,968 migraineurs	≥12	NR	OR adjusted for age, gender, race, income, duration of illness, comorbidities, use of preventive medication, and use of opioids	Risk of migraine increases as increasing BMI (for obese subjects OR 1.31, 95% CI 1.10-1.56 and for morbidly obese subjects OR 1.74, 95% CI 1.41-1.93) (same categories of BMI of Bigal, 2006)
Brown, 2000; Australian Longitudinal Study on Women's Health (45)	Cross-sectional	14,779; 12,812 migraineurs	18–23	100	Adjusted for age, area of residence, education, smoking and exercise	Increased risk of headache including migraine (OR 1.47, 95% CI 1.25-1.73) in obese (BMI ≥30) than in non-obese (BMI 25–29.9) women
Ford, 2008; National Health and Nutrition Examination Survey (NHANES) (46)	Cross-sectional	7,601; 1,660 migraineurs	≥20	28	OR adjusted for age, gender, ethnicity, educational status, marital status, smoking status, alcoholuse, systolic blood pressure, concentrations of C-reactive protein and total cholesterol, physical activity and self-reported diabetes	Risk for headache including migraine higher for BMI <18.5 (OR 2.01, 95% CI 1.34-3.02) or ≥30 (OR 1.37, 95% CI 1.09-1.72) than participants with a BMI of 18.5–<25
Ikeda, 2012 (23)	Case-control	221; 111 migraineurs	NR	69	Age and gender matching	Similar BMI values in female (21.5 vs 21.2) or male (24.0 vs 23.7) migraineurs and non-migraineurs
Jimenez-Sanchez, 2013; Romany National Health Survey (47)	Cross-sectional	993; 292 migraineurs	16–80	47	OR adjusted for age, gender, occupational status, self-related health, sleep habits, osteoarthritis, allergy, depression	Increased risk of migraine (OR 1.75, 95% CI 1.15-2.65) in obese (BMI >30) than in non-obese (BMI ≤30) subjects
Keith, 2008 (48)	Cross-sectional	220,370	≥18	100	Adjusted for age, race, and smoking	Increased BMI associated with significantly increased risk of headache, but not migraine
Mattson, 2007 (49)	Cross-sectional	684; 225 migraineurs	40–74	100	OR adjusted for age and education level	No increased risk of active migraine or inactive migraine in obese (BMI ≥30) subjects
Merikangas, 1997; National Health and Nutrition Examination Survey (NHANES) (24)	Cross-sectional*	12,220; 1,109 migraineurs	25–74	60	OR adjusted for age and gender	No increase in the risk of migraine associated with obesity
Molarius, 2008 (50)	Cross-sectional	43,770; 1,782 migraineurs	18–79	53	OR adjusted for lifestyle and socioeconomic factors	No increased risk of headache including migraine in overweight (BMI 25–29.9) and obese (BMI ≥30) than in normal weight subjects (BMI 18.5–24.9)
Peterlin, 2010; National Health and Nutrition Examination Survey (NHANES) (51)	Cross-sectional	21,783; 4,664 migraineurs	≥20	28 migraineurs	OR adjusted for age, income, education, race, smoking, alcohol use, and diabetes	Increased risk of migraine in obese men (OR 1.38, 95% CI 1.20-1.59) and women (OR 1.39, 95% CI 1.24-1.55) aged ≤55 years; no association in men and women aged >55 years
Peterlin, 2013; National Comorbidity Survey Replication (52)	Cross-sectional	3,862; 188 migraineurs	≥18	50 49 non-migraineurs 80 migraineurs	OR adjusted for age, gender, race, smoking, poverty index, depression and diabetes	Increased risk of episodic migraine in obese (BMI ≥30) subjects as compared to those of normal weight (OR 1.81, 95% CI 1.27-2.57) with a trend of increasing odds of episodic migraine with increasing obesity status from normal weight to overweight to obese (P=0.001) especially in participants younger than 50 years of age (OR 1.86, 95% CI 1.20-2.89)
Queiroz, 2009 (53)	Cross-sectional	3,848; 627 migraineurs	18–79	60	Adjusted for age, gender, education level, marital status, household income, job status, BMI and physical activity	No increased risk of migraine in obese (BMI ≥30) than in non-obese (BMI <25) subjects
Robberstad, 2010; Nord-Trøndelag Health (HUNT) Study (54)	Cross-sectional	5,847; 392 migraineurs	13–18	50	Adjusted for age, gender, smoking, and physical activity	Increased risk of migraine in overweight subjects (OR 1.6, 95% CI 1.4-2.2)
Scher, 2005; Genetic Epidemiology of Migraine (GEM) study (25)	Cross-sectional	5,755; 620 migraineurs	20–65	52	Adjusted for age, sex, and socioeconomic status	Similar mean BMI values in migraineurs (25.6), MA (25.8), MO (25.6), and non-migraineurs (25.4)
Schramm, 2013; the population-based German Headache Consortium Study (GHC) (55)	Cross-sectional	9,944; 1,709 migraineurs	18–65	71	OR adjusted for age, gender, smoking, drinking, obesity and education, and frequent intake of acute pain drugs	No increased risk of CM in obesity than non-obese subjects
Schwaiger, 2008; Bruneck Study (26)	Cross-sectional*	574; 111 migraineurs	55–94	29 migraineurs	Adjusted for age and gender	Similar mean BMI values in MA (24.7), MO (26.5), and non-migraineurs (25.9)
Téllez-Zenteno, 2010 (56)	Case-control	1,983; 1,371 migraineurs		83; 81 migraineurs	Age matching	Obesity percentage is lower in migraineurs (10.3%) than in controls (13.6%); morbidly obese percentage is lower in migraineurs (3.4%) than in controls (4.2%)
Vo, 2011; Omega Study (57)	Cross-sectional*	3,733; 670 migraineurs	≥18 (reproductive age)	100	OR adjusted for age, ethnicity, educational attainment, marital status, history of chronic hypertension or diabetes mellitus, smoking and exercise status.	Increased risk of migraine in obese women (OR 1.48, 95% CI 1.12-1.96); highest risk in severely obese (OR 2.07; 95% CI 1.27-3.39) and morbidly obese (OR 2.75; 95% CI 1.60-4.70) than normal weight women (same categories of BMI of Bigal, 2006]
Yu, 2012 (58]	Cross-sectional*	5,029; 467 migraineurs	18–65	49	OR adjusted for age, gender, habitation, marital status, occupation, educational level	No increased risk of migraine in overweight (BMI 23.0-24.9) and obese (BMI 25.0-29.9) than normal weight (BMI 18-5-23.0) subjects; increased risk of migraine in severely obese (BMI ≥30) subjects (OR 2.10, 95% CI 1.39- 3.12)
Winswold, 2011; Nord-Trøndelag Health (HUNT) Study (32)	Cross-sectional	48,713; 3,412 migraineurs	≥20	62 non- migraineurs 76 MA 74 MO	OR adjusted for age, gender and education level.	Increased risk of MA (OR 1.32, 95% CI 1.04-1.32) and MO (OR 1.17, 95% CI 1.04-1.32) in obese (BMI ≥30) than in non-obese (BMI <25) subjects
Winter, 2009; Women’s Health Study (WHS) (59)	Cross-sectional*	63,467; 12,613 migraineurs	≥45	100	OR adjusted for age, smoking, exercise, alcohol consumption, history of hypertension, postmenopausal status, postmenopausal hormone use and history of elevated cholesterol	No increased risk of migraine in women with BMI 30.0- 34.9 and BMI ≥35 than in women with BMI <23
Winter, 2011; German DMKG Headache Study (60)	Cross-sectional (3 study populations)	7,417; 427 migraineurs	35–75	52	OR adjusted for age and gender	No increased risk of migraine in subjects with BMI 25.0-29.9 and BMI ≥30 than in subjects with BMI <25
Winter, 2012 Women’s Health Study (WHS) (61)	Cohort, prospective	19,162; 3,483 migraineurs	≥45	100	HR adjusted for age, race, randomized treatment assignments, baseline BMI, total calorie intake, alcohol consumption, exercise, smoking status, postmenopausal hormone use, postmenopausal status, history of cholesterol >240 mg/dl, history of hypertension, history of depression	Increased risk of becoming overweight (HR 1.11, 95%CI 1.05–1.17) but not obese (HR 1.00, 95%CI 0.83–1.19) in migraineurs than in non-migraineurs
Wright, 2010; University of Washington Twin Registry (62)	Cross-sectional	3,471; 632 with headache including migraine	NR	62	OR adjusted for age and gender	No increased risk of headache including migraine in overweight (BMI 25.0-29.9) than in normal weight subjects; increased risk of headache including migraine in obese (BMI ≥30) than in normal weight subjects (OR 1.70, 95% CI 1.32- 2.19)

BMI=body mass index (expressed in Kg/m²), CDH= chronic daily headache, CI=confidence interval, CM=chronic migraine, HR=hazard ratio, MA=migraine with aura, MO=migraine without aura, NR=not reported, OR=odds ratio, TM=transformed migraine.

Cohort studies reporting the association in a cross sectional analysis.

At variance several other studies did not support an association between migraine and obesity. Data from a cross-sectional analysis of the NHANES cohort showed the absence of an increased risk of migraine in obese participants (24). Similarly, a large cross-sectional study pooling data from different data sets showed an increased risk of headache in general among obese women, but not specifically of migraine (48). In line with these findings, two large cross-sectional population-based studies conducted in the United States (US) did not find an increased risk of migraine associated with obesity (42,53); this same finding was supported by the German DMKG Headache Study (60). In agreement with those results, a cross-sectional study did not find an increased risk of active migraine or inactive migraine in obese women (49). A further cross-sectional study indicated that obesity was not associated with an increased risk of headache including migraine (50). In the cross-sectional GEM study, there was no difference between mean BMI values in migraineurs and non-migraineurs (25). Additionally, the cross-sectional Bruneck study (26), the case-control Japanese study (23), and a further case-control study on Mexican participants (56) reported similar BMI values or obesity prevalence in migraineurs when compared to non-migraineurs. Data from the cross-sectional, population-based German Headache Consortium Study showed that obese individuals did not have an increased risk of chronic migraine as compared to non-obese participants (55).

Cigarette smoking

Most of the available data point to an association between cigarette smoking and migraine (24,25,32,50,54,63 –68) although there are some discordant findings (20,26,69).

In the cross-sectional Collaborative Perinatal Project (involving 508 migrainous pregnant women and 3192 non-migrainous pregnant women), migraineurs were reported to smoke more heavily and to have a longer smoking history as compared to non-migraineurs (64). Similarly, a higher prevalence of smokers among migraineurs was reported by the cross-sectional population-based Erasmus Rucphen Family (ERF) study, in particular for MA (49% vs 31% in controls; p < 0.001) (63) and by the HUNT Study (32). Data from the prospective cohort NHANES reported a positive association between lifetime smoking history and migraine (odds ratio (OR) for migraine 1.2, 95% confidence interval (CI) 1.0–1.3; p < 0.001) while no association was reported between current smoking and migraine (OR for migraine 1.1, 95% CI 0.9–1.4) (24). Interesting data also came from the cross-sectional GEM study, showing a greater likelihood of migraine in smokers (25). In detail, there was an association between any migraine and being a former smoker (OR 1.25, 95% CI 1.0–1.6; p < 0.05), smoking fewer than 20 cigarettes per day (OR 1.35; 95% CI 1.0–1.8; p < 0.05), smoking more than 20 cigarettes per day (OR 1.59, 95% CI 1.2–2.2; p < 0.05) and between MA and smoking more than 20 cigarettes per day (OR 2.14, 95% CI 1.3–3.5; p < 0.05) while MO was not associated with any smoking status. A positive association between migraine and smoking was also reported by the prospective cohort Coronary Artery Risk Development in Young Adults (CARDIA) study, showing an increased risk of developing migraine along a follow-up period in current smokers at baseline (adjusted relative hazard 1.35, 95% CI 1.08–1.68) (65). The prospective cohort HUNT study identified an increased risk of migraine in younger (<40 years) smokers as compared to “never smokers” (OR 1.2, 95% CI 1.1–1.4 in those smoking <10 cigarettes/day and OR 1.3, 95% CI 1.2–1.5 in those smoking ≥10 cigarettes day) (66). Interestingly, a positive association was confined to younger smokers whereas a negative association was found in elderly smokers (OR 0.8, 95% CI 0.7–0.9 in those smoking <10 cigarettes/day and OR 0.8, 95% CI 0.7–0.9 in those smoking ≥10 cigarettes day). Further support for a positive association was provided by a cross-sectional Swedish population-based study (adjusted OR for headache in current smokers 1.3, 95% CI 1.2–1.4) (50), and by a cross-sectional study on German students, reporting an association between daily smoking and migraine (adjusted OR 2.7, 95% CI 1.3–5.6) (67). Lastly, similar conclusions have also been drawn from the youth part of the HUNT study, which found an association between smoking and migraine (adjusted OR for migraine in smokers 1.8, 95% CI 1.4–2.2) (54) and from a hospital-based case-control study conducted in Dhaka, Bangladesh, that found a six-fold increased risk of migraine in tobacco smokers as compared to non-smokers (adjusted OR 6.6, 95% CI 2.2–19.6) (68).

The prospective cohort Bruneck study found no difference in the prevalence of cigarette smoking, in an analysis of cross-sectional data, according to migraine status and subtype (26). Additionally, data from the SPAH (20) and from the large cross-sectional Spanish National Health Survey (SNHS) (69) confirmed the lack of any association and rejected the hypothesis of an association between migraine and smoking.

Alcohol consumption

Most of the available data indicate the occurrence of lower alcohol consumption in migraineurs as compared to non-migraineurs or the lack of any association between the two conditions. The cross-sectional GEM study found that migraineurs overall as well as migraineurs with MA and MO were less likely to consume alcohol than non-migraineurs (25). A cross-sectional study on medical students reported that individuals with migraine had lower alcohol consumption than participants without any headache (70). The absence of any association has been suggested by the Bruneck study (26), the SPAH study (20), and the ERF study (63), this latter showing similar alcohol consumption between migraineurs, irrespective of migraine subtype, and controls. In line with these results, the NHANES showed no increased risk of migraine in alcohol users (>12 drinks per year) than non-users (OR 1.1, 95% CI 0.9–1.2) (24). Moreover, an inverse relationship between alcohol consumption and headache including migraine, with a decreased risk of migraine in alcohol users, has been highlighted by a Swedish population-based study (adjusted OR 0.7, 95% CI 0.7–0.8 for headache in heavy alcohol users at least once a month) (50) and by the large cross-sectional SNHS (crude OR 0.54, 95% CI 0.49–0.59) (69). Further evidence came from the HUNT Study (32), which highlighted the presence of a higher proportion of alcohol abstainers among migraineurs, both with MA and MO, as compared to non-migraineurs. Finally, contrasting results have been reported by a cross-sectional study on German students that found an association between wine consumption and migraine (adjusted OR for migraine in consumers of one or more glasses/week 2.2, 95% CI 1.1–4.3) (67).

Family history of CVD

Many of the available studies indicate that a family history of CVD is frequently associated with migraine. A parental history of early myocardial infarction (<60 years of age) was more frequently found in migraineurs in the cross-sectional GEM study (25). Further evidence came from cross-sectional analysis of data from the large cohort HUNT Study, showing a higher likelihood of having a first-degree relative with early myocardial infarction in patients with MA and a higher likelihood of having a first-degree relative with stroke in patients with both MA and MO (28). With respect to migraine subtypes, no differences between patients with MA and MO were found by the Japanese case-control study. This study found an association between family history of CVD and migraine, irrespective of migraine subtype (23).

Global role of RFs in the association between migraine and CVD

Vascular RFs might contribute to increasing the risk of vascular events in migraineurs through an atherosclerosis-mediated pathway. However, there is no agreement in considering migraineurs as more prone to developing atherosclerosis as compared to non-migraineurs (71 –74). Although little information is available on the frequency of ischemic stroke subtypes in migraineurs, the available data suggest that most strokes in migraineurs are not of atherothrombotic etiology (72,75). In the same way, the majority of cardiac events in migraineurs seem not to be attributable to atherothrombosis (71,73). In addition, migraine has not been reliably associated with an increase in intima-media thickness (29,63,76,77), which is a marker of forthcoming atherosclerosis and an independent predictor of future risk for ischemic cardiac and cerebral events (78). In addition, conventional vascular RFs do not play a confounding role in the association between migraine and CVD, as suggested by studies that support the association between the two conditions even after the adjustment of results for conventional vascular RFs (online Supplementary Table) (1 –15).

Some studies have suggested that stroke risk is increased most in migraineurs with a favorable cardiovascular RF profile (5,75). Non-hypertensives, nonsmokers and nondiabetics are the individuals with the greatest increase in the risk of ischemic stroke associated with migraine (5,75) while there is an increased risk of ischemic stroke in migraineurs in the presence of cigarette smoking (79,80). Moreover, data from the WHS (49) indicated that the association between MA and vascular events differed according to vascular risk status: in fact, the association between MA and ischemic stroke was proven only among women in the lowest Framingham risk score group, while the association with myocardial infarction was proven only among women in the highest score group. With regard to the individual components of the Framingham risk score, this pattern of association was driven by a particularly increased risk of ischemic stroke among women with MA who were young (aged 45–49 years) and who had low total cholesterol levels. In contrast, women with MA who had high total cholesterol levels had an increased risk of myocardial infarction.

Discussion

Summary of findings

We found various studies that have addressed the burden of conventional vascular RFs in migraineurs. However, in some areas results from the available studies are conflicting and definitive conclusions cannot be drawn. Several factors may account for the extreme variability of the above results, including the use of different sampling criteria (selected versus unselected populations), the study design, the selection of criteria and procedures to diagnose migraine and to detect RF of interest (questionnaires, personal visits, medical records) with changing definitions over time, and inadequate control for confounders other than age and gender in some of the studies. The available data about the possible association between migraine and high BP values are particularly heterogeneous. Data addressing the possible association between migraine and DM indicate the lack of any association or in some cases a negative association between the two conditions. The body of evidence referring to the presence of dyslipidemia in migraineurs is relatively homogeneous and, with few exceptions, reports an association between migraine and an unfavorable lipid profile. However, the clinical relevance of the association remains questionable, as the difference in lipid levels between migraineurs and non-migraineurs was small despite statistical significance. Regarding obesity, a trend of increased risk of migraine with increasing obesity was observed, especially in young participants. However, no conclusive statement can be made because there are numerous conflicting data. Obesity may also represent an RF for migraine chronification. Evidence for an association between cigarette smoking and migraine is less controversial, and most of the available data support the existence of a positive relationship. On the other hand, most of the available studies did not find an association between migraine and increased alcohol consumption, and some have indicated reduced alcohol use in migraineurs compared with non-migraineurs. If moderate alcohol consumption is cardioprotective, migraineurs may be disadvantaged in this regard. Finally, many of the available studies suggested a more frequent family history of CVD in migraineurs as compared to non-migraineurs.

If vascular RFs played an active role, a greater frequency of atherothrombosis would be expected in patients with migraine, whereas most of the studies did not lead to this conclusion. In line with this argument, the majority of the studies have demonstrated, through adjusted data analysis, the independence of the association between migraine and CVD from the presence of conventional vascular RFs. Indeed, the studies carried out to date do not provide enough data to establish any possible adjunctive role of vascular RFs in comorbidity with migraine.

Implications

It seems unlikely that the higher risk of CVD in migraineurs is mediated by conventional vascular RFs. Different and alternative mechanisms should be appropriately investigated. Currently, it is not possible to stratify the vascular risk of migraineurs on the basis of the concomitant presence of vascular RFs. The recognition of a comorbid vascular risk profile that exposes migrainous patients to a further increase in the risk of CVD is important for clinicians as it might help to identify individuals who may require a more strict control of the comorbid vascular RFs.

Strengths and limitations

A variety of studies have investigated the possible association between migraine and conventional vascular RFs. However, while for some factors (e.g. arterial hypertension, dyslipidemia, obesity) the studies are numerous, for others the data are more limited. Striking differences among the results of the available studies hinder definite conclusions and suggest that additional confounding elements may play a significant role. Differences in the research design of the studies may partly account for the heterogeneity of the results. Some of the data reported in this review were drawn by comparisons between group analyses adjusted for gender and age only and consequently the possible role of some major confounders cannot be excluded. In addition, there is a paucity of data addressing the question of any differences according to migraine subtype.

Potential clinical/public health implications

While the evidence that links migraine to CVD is robust, the overall increase in absolute risk of CVD in migraineurs is rather small. Unfortunately, at the moment there are no reliable features that may indicate which individuals, across the overall migraine population, are at the highest risk of vascular events. MA and conventional vascular RFs are independent predictors of CVD events. The multiplicative role that coexistent vascular RF may play in the overall risk for vascular events, in a context of mutual enhancement, is well recognized (81). However, in migraineurs the exact impact that comorbid vascular RFs may exert on overall CVD is still a matter of debate. It is imperative, however, in daily clinical practice, to pay attention to the presence of RFs, even in young migrainous patients, and to be sure that comorbidities such as arterial hypertension, DM and dyslipidemia are adequately treated by a vascular neurologist or by a primary care physician. All patients who have migraine should be advised not to smoke, while obese patients should be encouraged to lose weight using appropriate strategies. At the moment it is not clear if the vascular risk associated with migraine can be modified and there is currently no direct evidence to support the hypothesis that migraine treatment may reduce future stroke risk. No drugs are currently recommended for vascular event prevention in migraineurs. Patients with migraine should not be prescribed aspirin or other anti-thrombotics for cardiovascular prevention unless other (i.e. non-migraine) indications are clearly present. The same applies to migraineurs for whom magnetic resonance imaging of the brain shows white matter hyperintensities or infarct-like lesions. Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers that in preliminary studies have shown some efficacy in migraine prevention (82) have also been associated with reduction of cardiovascular risk independently of their BP-lowering effect (83,84). However, further evidence is needed to support their role as a migraine-preventive treatment in the absence of high BP values and to demonstrate any possible benefit for the CVD risk of migraineurs. In addition, ergot alkaloids and triptans, two effective acute migraine treatments, may raise concerns regarding CVD safety because of their vasoconstrictive effect. However, the available data suggest that triptans when used at the suggested doses and in individuals without a history of CVD do not lead to increased CVD risk (4).

Recommendations for future studies

Future studies aimed at evaluating the burden of conventional vascular RFs in migraineurs should apply rigorous methodology with respect to the selection of cases, the diagnosis of migraine and the assessment of RFs of interest. The study sample should be adequately estimated in order to avoid the emergence of any biased association, and any possible confounding factors should be taken into account. Future studies should also clarify the role that comorbid RFs may play in the association between migraine and CVD to establish if comorbidities magnify the vascular risk of migraineurs.

Future research should also clarify the mechanisms that may explain the increased CVD risk in individuals with migraine. Studies have shown an association between migraine and stress (85) while stress is also a risk factor for CVD (86). Stress, personality traits and psychopathological status may be confounding factors in the association between migraine and CVD that, to date, are not accurately evaluated. Recent findings indicate that migraineurs have a vascular vulnerability in the brain but also at systemic levels, making them more susceptible to ischemic events in the presence of even mild triggers of ischemia (87). Biomarkers such as homocysteine, C-reactive protein, methylenetetrahydrofolate reductase, methionine synthase reductase, and von Willebrand factor should also be studied to understand their contribution to triggering ischemia or their role as markers of increased vascular risk in migraineurs (74,82,87). Studies should address the role of genetic factors. It is in fact possible that inherited factors may lead both to migraine and to vascular events. From this perspective, genome-wide-association studies may lead to some important conclusions. It would be useful to develop biomarkers or instrumental tests to enable the identification, among migraineurs, of any subgroup at particular increased risk of specific cardiovascular events. Studies should also establish if migraine is a modifiable or non-modifiable risk factor for CVD. Drugs acting on the renin-angiotensin system should be better studied, since there are data indicating that they may have a role in the prevention of migraine attacks and they may positively affect vascular risk.

Clinical implications

There is no solid evidence to demonstrate an increased burden of conventional vascular risk factors in migraineurs, with the only exceptions being dyslipidemia and cigarette smoking.

The presence of conventional vascular risk factors cannot entirely explain the association between migraine and cardiovascular disease.

Attempts to explain the possible mechanisms of an association between migraine and cardiovascular disease should not rely exclusively on the presence of comorbidities with vascular risk factors.

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

Bigal

Kurth

Santanello

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

Carolei

Marini

De Matteis

. History of migraine and risk of cerebral ischemia in young adults. The Italian National Research Council study group on stroke in the young. Lancet 1996; 347: 1503–1506.

Carter

Anderson

Jamrozik

. Migraine and risk of subarachnoid haemorrhage: A population-based case-control study. J Clin Neurosci 2005; 12: 534–537.

Hall

Brown

. Triptans in migraine: The risks of stroke, cardiovascular disease, and death in practice. Neurology 2004; 62: 563–568.

Henrich

Horwitz

. A controlled study of ischemic stroke risk in migraine patients. J Clin Epidemiol 1989; 42: 773–780.

Kuo

Yen

Chen

. Increased risk of hemorrhagic stroke in patients with migraine: A population-based cohort study. PLoS One 2013; 8: e55253–e55253.

Kurth

Slomke

Kase

. Migraine, headache, and the risk of stroke in women: A prospective study. Neurology 2005; 64: 1020–1026.

Kurth

Gaziano

Cook

. Migraine and risk of cardiovascular disease in women. JAMA 2006; 296: 283–291.

Kurth

Gaziano

Cook

. Migraine and risk of cardiovascular disease in men. Arch Intern Med 2007; 167: 795–801.

10.

Kurth

Kase

Schürks

. Migraine and risk of haemorrhagic stroke in women: Prospective cohort study. BMJ 2010; 341: c3659–c3659.

11.

Lidegaard

. Oral contraceptives, pregnancy and the risk of cerebral thromboembolism: The influence of diabetes, hypertension, migraine and previous thrombotic disease. Br J Obstet Gynaecol 1995; 102: 153–159.

12.

Lidegaard

Kreimer

. Contraceptives and cerebral thrombosis: A five-year national case-control study. Contraception 2002; 65: 197–205.

13.

Marini

Carolei

Roberts

. Focal cerebral ischemia in young adults: A collaborative case-control study. The National Research Council study group. Neuroepidemiology 1993; 12: 70–81.

14.

Nightingale

Farmer

RDT

. Ischemic stroke in young women: A nested case-control study using the UK General Practice Research Database. Stroke 2004; 35: 1574–1578.

15.

Sternfeld

Stang

Sidney

. Relationship of migraine headaches to experience of chest pain and subsequent risk for myocardial infarction. Neurology 1995; 45: 2135–2142.

16.

Tzourio

Iglesias

Hubert

. Migraine and risk of ischaemic stroke: A case-control study. BMJ 1993; 307: 289–292.

17.

Sacco

Ricci

Carolei

. Migraine and vascular diseases: A review of the evidence and potential implications for management. Cephalalgia 2012; 32: 785–795.

18.

Sacco

Ornello

Ripa

. Migraine and hemorrhagic stroke. A meta-analysis. Stroke 2013; 44: 3032–3038.

19.

Schürks

Rist

Bigal

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

20.

Benseñor

Goulart

Lotufo

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

21.

Gudmundsson

Thorgeirsson

Sigfusson

. Migraine patients have lower systolic but higher diastolic blood pressure compared with controls in a population-based study of 21,537 subjects. The Reykjavik Study. Cephalalgia 2005; 26: 436–444.

22.

Hagen

Stovner

Vatten

. Blood pressure and risk of headache: A prospective study of 22,685 adults in Norway. J Neurol Neurosurg Psychiatry 2002; 72: 463–466.

23.

Ikeda

Kawase

Takazawa

. Cardiovascular disease risk profile in Japanese midlife migraineurs. J Neurol Res 2012; 2: 10–15.

24.

Merikangas

Fenton

Cheng

. Association between migraine and stroke in a large-scale epidemiological study of the United States. Arch Neurol 1997; 54: 362–368.

25.

Scher

Terwindt

Picavet

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

26.

Schwaiger

Kiechl

Stockner

. Burden of atherosclerosis and risk of venous thromboembolism in patients with migraine. Neurology 2008; 71: 937–943.

27.

Tronvik

Stovner

Hagen

. High pulse pressure protects against headache: Prospective and cross-sectional data (HUNT study). Neurology 2008; 70: 1329–1336.

28.

Tronvik

Zwart

Hagen

. Association between blood pressure measures and recurrent headache in adolescents: Cross-sectional data from the HUNT-Youth study. J Headache Pain 2011; 12: 347–353.

29.

Tzourio

Gagnière

El Amrani

. Relationship between migraine, blood pressure and carotid thickness. A population-based study in the elderly. Cephalalgia 2003; 23: 914–920.

30.

Wang

Lin

. Increased risk of ischemic heart disease in young patients with migraine: A population-based, propensity score-matched, longitudinal follow-up study. Int J Cardiol 2014; 172: 213–216.

31.

Wiehe

Fuchs

Moreira

. Migraine is more frequent in individuals with optimal and normal blood pressure: A population-based study. J Hypertens 2002; 20: 1303–1306.

32.

Winsvold

Hagen

Aamodt

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

33.

Burch

Rist

Winter

. Migraine and risk of incident diabetes in women: A prospective study. Cephalalgia 2012; 32: 991–997.

34.

Aamodt

Stovner

Midthjell

. Headache prevalence related to diabetes mellitus. The Head-HUNT study. Eur J Neurol 2007; 14: 738–744.

35.

Berge

Riise

Fasmer

. Does diabetes have a protective effect on migraine? Epidemiology 2013; 24: 129–134.

36.

Burn

Machin

Waters

. Prevalence of migraine in patients with diabetes. BMJ 1984; 289: 1579–1580.

37.

Chen

Tang

. Comorbidity profiles of chronic migraine sufferers in a national database in Taiwan. J Headache Pain 2012; 13: 311–319.

38.

Kurth

Ridker

Buring

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

39.

Rist

Tzourio

Kurth

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

40.

Rockett

Perla Ada

Perry

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

41.

Saberi

Hatamian

Kazemnejad

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

42.

Bigal

Liberman

Lipton

. Obesity and migraine: A population study. Neurology 2006; 66: 545–550.

43.

Bigal

Lipton

. Obesity is a risk factor for transformed migraine but not chronic tension-type headache. Neurology 2006; 67: 252–257.

44.

Bigal

Tsang

Loder

. Body mass index and episodic headaches: A population-based study. Arch Intern Med 2007; 167: 1964–1970.

45.

Brown

Mishra

Kenardy

. Relationships between body mass index and well-being in young Australian women. Int J Obes Relat Metab Disord 2000; 24: 1360–1368.

46.

Ford

Pearson

. Body mass index and headaches: Findings from a national sample of US adults. Cephalalgia 2008; 28: 1270–1276.

47.

Jiménez-Sánchez

Fernández-de-las-Peñas

Jiménez-García

. Prevalence of migraine headaches in the Romany population in Spain: Sociodemographic factors, lifestyle and co-morbidity. J Transcult Nurs 2013; 24: 6–13.

48.

Keith

Wang

Fontaine

. BMI and headache among women: Results from 11 epidemiologic datasets. Obesity 2008; 16: 377–383.

49.

Mattsson

. Migraine headache and obesity in women aged 40–74 years: A population-based study. Cephalalgia 2007; 27: 877–880.

50.

Molarius

Tegelberg

Ohrvik

. Socio-economic factors, lifestyle, and headache disorders—a population-based study in Sweden. Headache 2008; 48: 1426–1437.

51.

Peterlin

Rosso

Rapoport

. Obesity and migraine: The effect of age, gender and adipose tissue distribution. Headache 2010; 50: 52–62.

52.

Peterlin

Rosso

Williams

. Episodic migraine and obesity and the influence of age, race, and sex. Neurology 2013; 81: 1314–1321.

53.

Queiroz

Peres

Piovesan

. A nationwide population-based study of migraine in Brazil. Cephalalgia 2009; 29: 642–649.

54.

Robberstad

Dyb

Hagen

. An unfavorable lifestyle and recurrent headaches among adolescents: The HUNT study. Neurology 2010; 75: 712–717.

55.

Schramm

Obermann

Katsarava

. Epidemiological profiles of patients with chronic migraine and chronic tension-type headache. J Headache Pain 2013; 14: 40–40.

56.

Téllez-Zenteno

Rishi Pawha

Hernandez-Ronquillo

. Association between body mass index and migraine. Eur Neurol 2010; 64: 134–139.

57.

Ainalem

Qiu

. Body mass index and adult weight gain among reproductive age women with migraine. Headache 2011; 51: 559–569.

58.

Liu

Yang

. Body mass index and migraine: A survey of the Chinese adult population. J Headache Pain 2012; 13: 531–536.

59.

Winter

Berger

Buring

. Body mass index, migraine, migraine frequency and migraine features in women. Cephalalgia 2009; 29: 269–278.

60.

Winter

Hoffmann

Meisinger

. Association between lifestyle factors and headache. J Headache Pain 2011; 12: 147–155.

61.

Winter

Wang

Buring

. Migraine, weight gain and the risk of becoming overweight and obese: A prospective cohort study. Cephalalgia 2012; 32: 963–971.

62.

Wright

Schur

Noonan

. Chronic pain, overweight, and obesity: Findings from a community-based twin registry. J Pain 2010; 11: 628–635.

63.

Stam

Weller

Janssens

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

64.

Chen

Leviton

Edelstein

. Migraine and other diseases in women of reproductive age. The influence of smoking on observed associations. Arch Neurol 1987; 44: 1024–1028.

65.

Hozawa

Houston

Steffes

. The association of cigarette smoking with self-reported disease before middle age: The Coronary Artery Risk Development in Young Adults (CARDIA) study. Prev Med 2006; 42: 193–199.

66.

Aamodt

Stovner

Hagen

. Headache prevalence related to smoking and alcohol use. The Head-HUNT Study. Eur J Neurol 2006; 13: 1233–1238.

67.

Milde-Busch

Blaschek

Borggräfe

. Associations of diet and lifestyle with headache in high-school students: Results from a cross-sectional study. Headache 2010; 50: 1104–1114.

68.

Sarker

Rahman

Harun-Or-Rashid

. Association of smoked and smokeless tobacco use with migraine: A hospital-based case-control study in Dhaka, Bangladesh. Tob Induc Dis 2013; 11: 15–15.

69.

Fernández-de-Las-Peñas

Hernández-Barrera

Carrasco-Garrido

. Population-based study of migraine in Spanish adults: Relation to socio-demographic factors, lifestyle and co-morbidity with other conditions. J Headache Pain 2010; 11: 97–104.

70.

Domingues

. Headache is associated with lower alcohol consumption among medical students. Arq Neuropsiquitriac 2011; 69: 620–623.

71.

Ahmed

Bairey Merz

McClure

. Migraines, angiographic coronary artery disease and cardiovascular outcomes in women. Am J Med 2006; 119: 670–675.

72.

Rist

Buring

Kase

. Migraine and functional outcome from ischemic cerebral events in women. Circulation 2010; 122: 2551–2557.

73.

Rose

Carson

Sanford

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

74.

Sacco

Ripa

Grassi

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

75.

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.

76.

De Hoon

Willigers

Troost

. Cranial and peripheral interictal vascular changes in migraine patients. Cephalalgia 2003; 23: 96–104.

77.

Vanmolkot

Van Bortel

de Hoon

. Altered arterial function in migraine of recent onset. Neurology 2007; 68: 1563–1570.

78.

Lorenz

Markus

Bots

. Prediction of clinical cardiovascular events with carotid intima-media thickness: A systematic review and meta-analysis. Circulation 2007; 115: 459–467.

79.

Chang

Donaghy

Poulter

. Migraine and stroke in young women: Case-control study. The World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Br Med J 1999; 318: 13–18.

80.

Tzourio

Tehindrazanarivelo

Iglesias

. Case-control study of migraine and risk of ischaemic stroke in young women. BMJ 1995; 310: 830–833.

81.

Grundy

Pasternak

Greenland

. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation 1999; 100: 1481–1492.

82.

Ripa

Ornello

Pistoia

. The renin-angiotensin system: A possible contributor to migraine pathogenesis and prophylaxis. Exp Rev Neurother 2014; 14: 1043–1055.

83.

Dahlöf

Devereux

Kjeldsen

. LIFE study group. Cardiovascular morbidity and mortality in the losartan intervention for endpoint reduction in hypertension study (LIFE): A randomized trial against atenolol. Lancet 2002; 359: 955–1003.

84.

Yusuf

Sleight

Pogue

. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The heart outcomes prevention evaluation study investigators. N Engl J Med 2000; 342: 145–153.

85.

Sauro

Becker

. The stress and migraine interaction. Headache 2009; 49: 1378–1386.

86.

Steptoe

Kivimäki

. Stress and cardiovascular disease. Nat Rev Cardiol 2012; 9: 360–370.

87.

Eikermann-Haerter

. Spreading depolarization may link migraine and stroke. Headache 2014; 54: 1146–1157.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB