Abstract
Background
Metabolic syndrome (MetS) and migraine are known to be associated. This study assessed the risk of MetS and its clinical characteristics in migraine with aura (MA) and without aura (MO) based on a large-scale cross-sectional survey.
Methods
The study material consisted of 751 participants in the Nutrition, Environment and CardioVascular Health (NESCaV) survey. Diagnosis of migraine was based on the ef-ID migraine questionnaire and MetS was defined according to the Revised-Adult Treatment Panel III criteria. Sociodemographic and risk factors were also recorded. Weighted logistic regression was used to assess the risk of MetS.
Results
After adjusting for stratification (gender, age, district) and other factors (smoking status, sedentary lifestyle, family history of stroke, myocardial infarction and hypertension), MA subjects were at higher risk of MetS (OR 3.45; 95% CI: 1.63–7.29) while MO individuals were not, when compared to non-migraineurs. When considering MetS components, MA was positively associated with low HDL-cholesterol (OR 2.26; 95% CI: 1.08–4.74), hyperglycemia (OR 2.77; 95% CI: 1.30–5.88) and abdominal obesity (OR 2.03; 95% CI: 1.07–3.86).
Conclusion
Migraineurs with aura are at higher risk of MetS, suggesting that these subjects, already more exposed to stroke, may benefit from a systematic screening for the metabolic syndrome.
Introduction
Migraine is one of the most common neurological disorders, with an important individual and societal burden (1). Up to one third of migraineurs present visual disturbances prior to or during an attack, a subtype named “migraine with aura” (MA) (2). Metabolic syndrome (MetS), a cluster of interacting cardiovascular risk factors including hypertension, insulin resistance, dyslipidemia and central adiposity, is known as a major public health problem due to its association with an increased risk of atherosclerotic cardiovascular disease and type 2 diabetes. In Europe, the prevalence rates of MetS vary from 10 – 26%. Although the pathogenesis of MetS is still unclear, several factors are involved in its development such as sedentary lifestyle, dietary habits, obesity, insulin resistance and genetics (3). A recent prospective study showed that migraine, particularly migraine with aura (MA), is a recognized risk factor for the development of MetS (4). Although the link between MetS and migraine is still little understood, close monitoring of vascular risk factors is particularly important in migraine, especially in MA. Indeed, given the increased risk of cardiovascular disease (5) and the use of migraine-preventive medications that may favor metabolic disorders such as weight gain (6,7), MA patients constitute a specific risk group. Therefore, the present study aimed to assess the risk of MetS in subjects among the general population suffering from migraine, whether MA or MO, after accounting for potential confounders such as sociodemographic features, family history, dietary and behavioural determinants. We also investigated the clinical components of MetS in the two migraine subtypes.
Methods
Study design
Within the NESCaV (Nutrition, Environment and CardioVascular Health) inter-regional European project, a cross-sectional study was conducted between 2010 and 2012 on a random sample of 1017 subjects from the province of Liège, Belgium. Sampling was stratified into 10-year age classes, gender and four districts (Liège, Verviers, Huy, Waremme). The design and conduct of NESCaV have been described in detail in previous publications (8,9). Of the 1017 subjects, 751 (73.8%) who filled out the ef-ID Migraine questionnaire for screening migraine were included in the study (10). Exclusion criteria included pregnancy and subjects living in institutions. All participants were duly informed and signed a consent form before the data collection. Ethical approval was obtained from the Ethics Committee of the Faculty of Medicine of the University of Liège (B70720097541) (9).
Subject characteristics
Sociodemographic and lifestyle characteristics of participants were recorded as described previously (9). Briefly, education was classified into four categories, “primary and lower secondary”, “secondary”, “bachelor” and “university”. Economic level was treated as a binary variable, below or above poverty risk level. In relation to MetS and/or migraine, study subjects were asked about their family history of diabetes mellitus, hypertension, myocardial infarction and stroke by means of the self-administered questionnaire. Smoking status was categorized as current, past and non-smoker. Inactivity was defined as the absence of sport practice at least once a week (9). Nutritional habits were acquired by means of a validated food frequency questionnaire (FFQ) (11). Food data collected this way were used to define a diet quality index (DQI) developed on the basis of WHO recommendations for the prevention of chronic diseases (12,13). This has 13 components: total fat, saturated fat (SFA), poly-unsaturated fat (PUFA), mono-unsaturated fat (MUFA) expressed as a percentage of total daily energy intake, the ratio of n-6 to n-3 fatty acids, cholesterol intake in mg/day, total carbohydrates, simple sugars, total protein expressed as a percentage of total daily energy intake, sodium, fruits/vegetables, and lastly total and soluble fiber intakes in g/day. One point was attributed for each component intake that met the WHO recommendations, otherwise a zero was given (13). Alcohol intake was evaluated in grams of alcohol per day by the FFQ as well. A standard drink contains 10 g of alcohol. Three categories were defined: non-consumer (no drink per day), moderate consumer (≤ 2 drinks per day for women and ≤ 3 drinks per day for men) and heavy consumer (> 2 drinks per day for women and > 3 drinks per day for men). These categories were established according to the National Program on Nutrition and Health (PNNS) in France in 2001 (14).
Migraine
The diagnosis of migraine was based on a validated screening instrument: “ef-ID Migraine”. According to this questionnaire, a subject reporting any disabling headaches during the previous 12 months accompanied by nausea/vomiting and/or photophobia was classified as “migraine without aura (MO)”. However, when the subject also responded positively to visual disturbances before headache, he/she was classified as “migraine with aura (MA)”. Otherwise, subjects were considered as “non-migraineurs” (10,16).
Metabolic Syndrome
Anthropometric measurements including waist circumference (WC) were performed by a trained health professional with subjects wearing light clothes without shoes. Blood pressure (BP) was measured at least three times with a minimum of one-minute intervals. Analyses were based on the mean values of the second and third measurements of blood pressure. Fasting morning blood samples were analyzed by the central Laboratory of the University Hospital of Liege (CHU), using standardized methods (9). The Revised-Adult Treatment Panel III (RATP III) definition was used to define MetS. MetS was present when three or more of the following criteria were met: (1) WC ≥ 102 cm for men and ≥ 88 cm for women; (2) raised level of triglycerides (TG) ≥ 150 mg/dl or specific treatment for this lipid anomaly; (3) reduced level of high density lipoprotein cholesterol (HDL-C) < 40 mg/dl for men and < 50 mg/dl for women or specific treatment for this lipid anomaly; (4) BP ≥ 130/85 mmHg or treatment of previously diagnosed hypertension; (5) fasting plasma glucose (FPG) level ≥ 100 mg/dl or use of medication for hyperglycemia (15).
Statistical analyses
Statistical analyses were performed by using the SAS 9.4 survey procedure for complex sampling design (SAS© Institute Inc., Cary, NC, USA). To account for the stratified random sampling method, weighted statistical methods were applied (9,16). Results for categorical variables were expressed as frequencies and percentages, and results for quantitative variables were expressed as median and interquartile range (IQR). The association between MetS and migraine adjusted for stratification factors (age, gender and district) (Model I) and for other related factors significant at p < 0.10 (Model II) was assessed by logistic regression analysis and quantified by the odds ratio (OR) with 95% confidence interval (95% CI). Missing data were not replaced. Results were considered significant at the 5% critical level (p < 0.05).
Results
Migraine status
Sociodemographic characteristics and health indicators of the study participants (N=751) from the NESCaV study according to migraine status.
Adjusted for age, gender, district.
Metabolic syndrome
Sociodemographic characteristics and health indicators of the study participants from the NESCaV study according to the presence or absence of MetS (N = 751).
Adjusted for age, gender, district.
Association between migraine status and MetS
Association between migraine and metabolic syndrome adjusted for stratification factors (model I) and for factors related to either migraine or metabolic syndrome at p < 0.10 (model II) from the NESCaV study according to the presence or absence of MetS (N = 751).
Adjusted for age, gender, district.
Adjusted for age, gender, district, sedentary lifestyle, smoking status and family history of myocardial infarction, stroke and hypertension.
MetS components and migraine status
Association between metabolic syndrome components and type of migraine.
Logistic regression adjusted for age, gender, district.
Logistic regression adjusted for age, gender, district, sedentary lifestyle, smoking status and family history of myocardial infarction, stroke and hypertension.
Discussion
The findings of this cross-sectional study corroborate those of the Nord-Trøndelag Health prospective study (HUNT), which showed a relationship between MA and MetS. In this 11-year longitudinal study, Winsvold et al. reported a 14.6% incidence of MetS in MA. The risk of developing MetS in subjects with MA was 40% higher compared to people without headache (incident risk ratio [IRR] = 1.39, 95% CI: 1.03–1.86). The risk rose to 2.10 when subjects with MA were daily smokers (IRR = 2.10, 95% CI: 1.53–2.89). In the HUNT study, among the MetS components, hyperglycemia, abdominal obesity and low HDL-C level were positively associated with MA (4). Despite the cross-sectional design of our study, we found the same higher risk of hyperglycemia, abdominal obesity and low HDL-C level in MA.
Several other studies have explored the association between MetS and migraine (17). In a case-control study, Salmasi et al. found no difference in MetS frequency between migraineurs and controls (17% and 15%, respectively). However, migraine patients were significantly more obese and had a higher level of waist circumference than the control group (18). In the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), the results of the multivariate-adjusted regression analyses showed an inverse association between MetS and confirmed migraine in men (OR 0.62; 95% CI: 0.43–0.99) but not in women (19). In another study design, Guldiken et al. reported that 19.5% of patients with MetS had migraine (11.9% in men and 22.5% in women) (20). Our findings parallel those results in that impaired glucose metabolism and increased waist circumference are more frequent MetS components in subjects with migraine than in non-migraineurs (p < 0.05). In another study, MetS was found in 31.9% of migraine patients: a low HDL-C level was reported in 80%, increased waist circumference in 57.8%, hypertension in 27.4%, hypertriglyceridemia in 20.7%, insulin resistance in 11.1% and increased fasting blood sugar in 9.6% (21). It should be mentioned, however, that insulin resistance was not confirmed in the study by Sacco et al. of MO and MA patients (22). Gozke et al. reported, in their observational study, a higher frequency of hypertriglyceridemia in MO compared to tension type headache and headache-free groups (23). In our study, none of the MetS components was significantly more prevalent in MO. Amongst 142 female migraine patients, He et al. found a MetS prevalence of 12%. The most represented components of MetS were hypertension (35.3%), followed by hyperglycemia (30.7%), low HDL-C level (25.8%), high triglycerides (21.2%) and abdominal obesity (16.0%). Chronic migraine in females was associated with MetS, especially when it was comorbid with medication overuse headache (MOH) (7). Differing results between our study and other studies are probably due to study design, study population, definition of MetS, and diagnosis of migraine, in particular of MA.
The present study findings suggest that MetS, a constellation of traditional risk factors, may provide an explanation of the association between migraine with aura and cardiovascular disease (17, 24). The mechanisms underlying the relationship between migraine and MetS are still unclear. Several hypotheses have been proposed and discussed. Among them, it has been argued that migraine may lead to MetS by adopting an unhealthy lifestyle (4,17,25) or by using migraine or affective disorder treatments (6,7,17). Another hypothesis is that migraine and MetS share common underlying pathways such as specific genetic variants (4,26,27) or oxidative stress (28). Recent cross-phenotype analysis obtained through genome wide association study (GWAS) suggested that genetic factors play a role in migraine and cardiovascular events (26,27). Further research work in this particular field is awaited to elucidate how both diseases interact. Concerning oxidative stress, recent studies reported an association between this specific condition and migraine (28–30) and also with MetS (31). Oxidative stress results from an imbalance between oxidant exposure and antioxidant protections, leading to an increased production of oxidative free radicals and reactive oxygen species (ROS) (32,33), which can cause damage to biomolecules like DNA, lipids, and proteins (30,32,33). Beckener et al. showed in their study that an increased oxidative stress in migraineurs contributes to an atherogenic lipid profile, hyperinsulinemia and hence leads to metabolic risk (28).
Understanding the mechanisms involved in the activation of oxidative stress could be helpful to identify novel biomarkers and molecular targets, allowing the development of novel therapeutics to prevent and/or treat both migraine and MetS (28,33).
The association between MA and MetS must also be taken into account in the pharmacological treatment of migraine. In particular, drugs favoring weight gain (6), hypertension (7) or dyslipidemia (17), which are components of MetS and lead to oxidative stress (33), should be avoided.
The strengths of the present study rely on the stratified random sample of the general adult population, the use of a validated migraine screening tool and objective measurements of anthropometric, clinical and biochemical features, and the large amount of covariate information, which allowed controlling for potential confounding factors. The study has some limitations. First, despite the use of a validated migraine screening tool, it is considered that MO and, particularly MA, can only be accurately diagnosed through a face-to-face interview by a clinician. Moreover, this tool is likely to miss the screening of other types of migraine with aura such as MA without visual aura (10%) or migraine aura without headache (2). Second, the cross-sectional design precludes any conclusion about a causal relation between MetS and MA. Third, results of the robust model (model II) included only 533 participants (71%) of the original 751 study participants. This can lead to a possible selection bias. Fourth, the number of participants in each migraine group is limited, especially in the MA group (n = 79). However, despite the small number of MA subjects, significant associations were found with MetS, low HDL-C level, hyperglycemia and abdominal obesity in this subgroup. Due to the small number of subjects in each group, some potential confounders could not be analyzed in the multivariate logistic analysis, such as anti-migraine treatment (n = 5). Despite adjustment for a large number of confounders, some potential confounders may be missing including the effect of anxiety, depression and sleep disorders that were not explored in this study. Therefore the outcomes of our study need to be confirmed in large-sized studies with a prospective design.
Conclusion
Compared to non-migraineurs, there is a higher risk of MetS in MA but not in MO subjects. Therefore, a systematic routine screening of MetS may be recommended in migraineurs with aura only. This proposal is reinforced by the fact that MA subjects are already more exposed to stroke.
Article highlights
The risk of MetS is significantly higher in MA subjects compared to non-migraineurs, while it is not in MO subjects. When considering MetS components separately, MA was positively associated with low HDL-cholesterol, hyperglycemia and abdominal obesity. A systematic screening of MetS in migraineurs with aura should be advised.
Footnotes
Acknowledgments
Ethical approval was obtained from the Ethics Committee of the Faculty of Medicine of the University of Liège (B70720097541).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: the NESCaV survey project was supported by the INTERREG IV A program, “Greater Region”, 2007–2013. It was co-funded by the European Regional Development Fund (ERDF) [NESCAV N°39/GR/3/3/056], the Walloon Region and the University of Liège in Belgium. This study was an ancillary study of the NESCaV survey project.