Original Paper

Eur Neurol 2010;64:134139 DOI: 10.1159/000316656

Received: April 21, 2010 Accepted: June 7, 2010 Published online: July 22, 2010

Association between Body Mass Index and Migraine

Jos F.Tllez-Zenteno a DaveRishiPahwa a LizbethHernandez-Ronquillo a GuillermoGarca-Ramos b AntonioVelzquez c

Division of Neurology, Department of Medicine, Royal University Hospital, University of Saskatchewan, Saskatoon, Sask., Canada; b Department of Neurology and Psychiatry, and c Blood Bank, National Institute of Medical Sciences and Nutrition Salvador Zubirn, Mexico City, Mexico

Key Words Overweight Migraine Body mass index Aura

ly significant. No association was found between the disability and severity of migraine and BMI. Conclusions: This study did not find associations between severity or disability of migraine and BMI. Copyright 2010 S. Karger AG, Basel

Abstract Objective: To explore the prevalence of overweight and obesity in patients with migraine. Background: Previous studies support the concept that obesity is an exacerbating factor for migraine. Also, some studies have found an increased frequency of obesity and overweight in migraine patients compared to the normal population. Methods: We studied 1,371 patients with migraine and 612 controls. The migraine population was matched by gender with a healthy control group. Results: Mean age of patients with migraine was 38.0 8 13.3 years and in the controls it was 34.8 8 12.1 years. The percentage of females in both groups was similar (migraine 81.6% vs. control 83.3%, p = 0.40). The distribution of body mass index (BMI) in migraine patients and controls was as follows: underweight patients (BMI !18.5) 3.1% migraine versus controls 1.5%; normal (BMI 18.524.9) 44.8% migraine versus controls 47.1%; overweight (BMI 2529.9) 38.3% migraine versus controls 33.7%; obese (BMI 3034.5) 10.3% migraine versus controls 13.6%; morbidly obese (BMI 35) 3.4% migraine versus controls 4.2%. Overweight and obesity in migraine patients versus controls were statistical-

Background

Migraine and obesity are highly prevalent disorders in the general population. The prevalence of migraine ranges from 5 to 35% in females and from 3 to 20% in males [1]. The highest rates have been found in North-America, Latin-America and Europe compared with Africa and Asia [1]. Migraine is a public health concern with a significant impact on the individual and society. Some studies have shown that obesity is comorbid with a number of chronic pain syndromes, including fibromyalgia, back and neck pain. Little is known about the influence of baseline weight status on the prevalence, severity, and disability of episodic migraine. Migraine and obesity may be linked from a biochemical perspective [2, 3]. It has been demonstrated that obesity is a pro-inflammatory state; adipocytes can secrete a variety of cytokines, including IL-6 and tumor necrosis factor, which are cytoJos F. Tllez-Zenteno, MD, PhD Division of Neurology, Department of Medicine, Royal University Hospital Saskatoon, SK S7N 0W8 (Canada) Tel. +1 306 966 8011, Fax +1 306 966 8008, E-Mail jft084@mail.usask.ca

2010 S. Karger AG, Basel 00143022/10/06430134$26.00/0 Fax +41 61 306 12 34 E-Mail karger@karger.ch www.karger.com Accessible online at: www.karger.com/ene

kines that promote inflammation [2, 3]. On the other hand, migraine is associated with neurovascular inflammation involving the same cytokines [4]. The relationship between overweight, and migraine and headache in general has been the focus of clinical research in recent years. Some studies have shown that general obesity or total body obesity, which is estimated with the body mass index (BMI), has been shown to be related to headache disorders in several epidemiological studies. However, there is a considerable uncertainty about the nature of the obesity/headache relationship and whether it is specific to migraine, or chronic daily headache, or headache in general [2]. The uncertainty is caused by inconsistent observations between studies. Some studies have demonstrated that obesity in patients with migraine (PWM) is associated with aura, high frequency of migraines, greater severity and with an increased frequency of photophobia and phonophobia [5, 6]. On the other hand, other studies have not found any association between obesity and migraine and its characteristics [7]. The most consistent information reports a potential association between frequent headache in patients with overweight and chronic daily headache [8]. Finally, some studies have not demonstrated that obesity is more frequent in PWM than in the general population [5, 7]. We report a large population-based study investigating (1) the influence of BMI on the prevalence of episodic migraine, and (2) the influence of BMI on migraine clinical features, including headache frequency, pain severity, disability, and associated symptoms.

into the seven subtypes accepted by the IHS. This test assesses the type and frequency of headache, medications and it includes a brief neurological exam to rule out secondary causes. It has been previously validated in the Mexican population [911]. None of the patients were evaluated during a migraine episode. Control Groups The control group was a group of healthy people attending the blood donor bank of the National Institute of Medical Sciences and Nutrition on the days of the migraine case recruitment. These controls participated voluntarily, had a general examination and answered a standardized questionnaire to rule out other relevant medical conditions. None of the controls had migraine and were attending the clinic for clinical reasons, and all were determined to be healthy. The process of headache ascertainment was performed with the same questionnaires as in PWM. Headache-Related Disability and Frequency of Headaches We used the migraine disability assessment (MIDAS) questionnaire to evaluate headache-related disability [12]. The MIDAS score was classified into four grades of severity: (I) scoring 05 (minimal or infrequent disability) these migraine sufferers tend to have little or no treatment needs; (II) scoring 610 (mild or infrequent disability) these migraine sufferers tend to have moderate treatment needs; (III) scoring 1120 (moderate disability); (IV) scoring 21 and more (severe disability) grade III and IV migraine sufferers have high disability and tend to have urgent treatment needs [12]. Frequency of headaches was ascribed to three groups: mild = patients had migraines at least 6 times a year or every other month, moderate = monthly headaches, and severe = weekly headaches. Body Mass Index BMI was calculated according to the following formula: BMI = weight (kg)/height (m)2. We defined five categories based on BMI: underweight (!18.5), normal weight (18.524.9), overweight (2529.9), obese (3034.9), and morbidly obese (135). Analysis We used descriptive statistics to assess frequencies and distributions. As appropriate, proportions were compared with either 2 test or Fishers exact test. All analyses were performed using SPSS version 17 (SPSS Inc., Chicago, Ill., USA).

Methods
Methods and Sample Design This is a cross-sectional study performed in two large teaching hospitals in Mexico City (National Institute of Medical Sciences and Nutrition, and National Institute of Neurology and Neurosurgery) and affiliated hospitals. The Institutional Review Boards of both institutions approved the research protocol. The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. Migraine Patients Adult PWM were identified from the outpatient neurology clinic of the previously mentioned hospitals. After their clinic appointment, PWM were approached and asked to voluntarily participate in the study. Informed consent of participants was obtained before interviews and measures. A trained physician conducted a standardized diagnostic interview adhering to the criteria of the International Headache Society (IHS), evaluating the presence of headache and its subtypes in the previous year. The headache test contains 51 questions and classifies headache

Results

General Characteristics of the Cohort Our sample consists of 1,985 subjects. 1,371 (70%) were PWM and 612 (30%) were healthy blood donors (HBD) matched by gender. In the migraine group 1,122 (81.6%) patients were females and in the HBD group 509 patients (83.3%) (p = 0.36). The height was 159.6 8 0.09 cm in HBD versus 160.3 8 0.82 cm in PWM (p = 0.15), the weight was 66.4 8 13.0 kg in HBD versus 64.7 8 13.3 kg in PWM. Finally, BMI was 25.3 8 4.4 in PWM versus 25.8 8 4.4 in HBD (table1).
Eur Neurol 2010;64:134139

Body Mass Index in Migraine

135

Patients with Migraine 1,371 patients with headache fulfilled the criteria for migraine according to the standardized questionnaire. Only patients with episodic migraine were included in this study. Patients with chronic migraines were excluded. The age at onset of migraine was 19.5 8 13.3 years, 725 patients (53%) had migraine with aura (MA) and 646 (47%) had migraine without aura. The disability score (MIDAS) showed little or no disability in 192 patients (9.7%), mild disability in 138 (10.1%), moderate disability in 460 (33.6%), and severe disability in 581 (42.4%). Regarding self-reported comorbidity ascertained in this study, 74 patients had hypertension (9%), 2 Reynaud phenomenon (0.1%), 6 mitral valve prolapse (0.4%), 6 angina/ myocardial infarction (0.4%), 3 stroke (0.2%), 17 epilepsy (1.2%), 32 vertigo (2.3%), 206 functional bowel disorder (15%), 20 asthma (1.5%), 126 allergies (9.2%), 250 depression (18.3%), 25 mania (1.8%), 6 panic attacks (0.4%) and 188 anxiety (13.7%). Body Mass Index in Patients with Migraine and Controls PWM had a higher proportion of overweight (38.3%) compared with HBD (33.7%), p = 0.03. On the other hand, HBD had a higher proportion of obesity (13.6%) and morbid obesity (4.2%) than PWM (10.3 and 3.4%, respectively); only obesity was statistically significant. The presence of underweight was more frequent in HBD than PWM (p = 0.03). The category of normal weight was similar in both groups (table2; fig.1). Disability of Migraine and Body Mass Index There was no correlation between the disability scores and BMI. PWM and overweight were homogeneously distributed in all the weight categories and no statistical association was identified. The same trend was seen in obese and morbidly obese PWM (table3). Severity of Migraine and Body Mass Index There was no correlation between the severity of migraine and BMI. PWM were well distributed according to severity, and the different categories of weight had no significant differences (table4). Migraine with Aura and Body Mass Index We did not find any correlation between the presence of MA and BMI in PWM. The distribution of weight was similar between patients with and without MA (table5).

% 50 45 40 35 30 25 20 15 10 5 0 Underweight Normal Overweight Weight category Obesity Morbid obesity Migraine Blood donors

Fig. 1. BMI distribution in PWM and HBD. No difference was

seen between groups in any of the categories.

Table 1. General characteristics of the sample

Variable Age Gender, female Height Weight BMI

PWM (n = 1,371) 38.0813.3 1,125 (81.6%) 160.380.08 64.7813.3 25.384.46

HBD (n = 612) 34.87812.17 509 (83.3%) 159.680.093 66.3813.05 25.884.46

p value 0.00 0.40 0.07 0.01 0.02

PWM = Patients with migraine; HBD = healthy blood donors.

Table 2. Comparison of BMI between PWM and HBD

Weight category Underweight (n = 52) Normal (n = 903) Overweight (n = 732) Obesity (n = 224) Morbid obesity (n = 73) Total

PWM 43 (3.1) 615 (44.8) 526 (38.3) 141 (10.3) 47 (3.4) 1,371

HBD 9 (1.5) 288 (47.1) 206 (33.7) 83 (13.6) 26 (4.2) 612

p value 0.03 0.3 0.03 0.01 0.3

Underweight (<18.5), normal weight (18.524.9), overweight (2529.9), obese (3034.9), and morbid obesity (>35). Numbers in parentheses denote percentages. PWM = Patients with migraine; HBD = healthy blood donors.

136

Eur Neurol 2010;64:134139

Tllez-Zenteno et al.

Table 3. Correlation between MIDAS grade and BMI

Grade I Underweight (n = 43) Normal (n = 614) Overweight (n = 525) Obesity (n = 141) Morbid obesity (n = 47) 6 (3.1) 86 (45) 70 (36.6) 22 (11.5) 7 (3.7)

Grade II 4 (2.9) 58 (42) 49 (35.5) 21 (15.2) 6 (4.3)

Grade III 10 (2.2) 214 (46.5) 174 (37.8) 43 (9.3) 19 (4.1)

Grade IV 23 (4.0) 257 (44.2) 231 (39.8) 55 (9.5) 15 (2.6)

p 0.28 0.47 0.55 0.24 1.64

Underweight (<18.5), normal weight (18.524.9), overweight (2529.9), obese (3034.9), and morbid obesity (>35). MIDAS questionnaire; I = little or no disability 05 points; II = mild disability 610 points; III = moderate disability 1120 points; IV severe disability 21+ points. Figures in parentheses are percentages.

Table 4. Correlation between severity of migraine and BMI

Table 5. BMI distribution between patients with and without aura

Mild % Underweight (n = 43) 2.9 (1) Normal (n = 614) 51.4 (18) Overweight (n = 525) 40 (14) Obesity (n = 141) 1.4 (2) Morbid obesity (n = 47) 0

Moderate Severe % %

p Underweight Normal Overweight Obesity Morbid obesity

MA (n = 725) 19 (2.6) 329 (45.4) 275 (37.9) 76 (10.5) 26 (3.6)

Migraine without aura (n = 646) 24 (3.7) 285 (44.2) 250 (38.8) 65 (10.1) 21 (3.3)

p value 0.24 0.65 0.75 0.69 0.73

2.2 (12) 3.8 (30) 0.28 46.2 (249) 43.6 (347) 0.46 36.5 (197) 39.4 (314) 0.55 45.3 (64) 53.1 (75) 0.23 3.2 (17) 3.8 (30) 0.44

Underweight (<18.5), normal weight (18.524.9), overweight (2529.9), obese (3034.9), and morbid obesity (>35).

Underweight (<18.5), normal weight (18.54.9), overweight (2529.9), obese (3034.9), and morbid obesity (>35). Numbers in parentheses denote percentages. MA = Migraine with aura.

Body Mass Index and Characteristics of Migraine In our analysis, photophobia has a trend to be more frequent in patients with overweight (92.4%), obesity (93.6%) and morbid obesity (89.4%), compared with patients with normal weight (87.6%) and underweight (83.7%). Only the group of patients with photophobia and overweight had a statistical association (table 6). Also, nausea had a trend to be more frequent in overweight, obesity and morbid obesity.

Discussion

We performed a large population-based cross-sectional study to explore the influence of BMI on the prevalence and the clinical characteristics of patients with episodic migraine. The main findings of our study are the following: (1) we did not find an increased prevalence of obesity in PWM compared with a healthy control group of blood
Body Mass Index in Migraine

donors; (2) our study shows no relation between disability and severity of migraine and BMI; (3) no correlation was found between the presence of aura and BMI; (4) photophobia was more frequent in PWM and obesity. This controlled study shows that obesity is not comorbid with episodic migraine and supports previous evidence. Bigal et al. [5] performed a large population-based study including 30,125 subjects from the general population, 3,091 of whom were patients with episodic migraine. In this study, 27.3% of PWM had overweight compared with 31.1% in the general population, obesity was present in 9.7 versus 10.4% and morbid obesity in 5.1 versus 4.5%; none of these rates were statistically significant. Our study shows similar findings as the study by Bigal et al. [5]. When compared, BMI and weight were slightly higher in HBD than PMW, which was statistically significant, but when the BMI was broken down into different categories, the distribution did not show consistent trends. Overweight was slightly more frequent in PWM than
Eur Neurol 2010;64:134139

137

Table 6. Correlation between migraine characteristics and BMI

Nausea Underweight (n = 43) Normal (n = 614) Overweight (n = 525) Obesity (n = 141) Morbid obesity (n = 47) 35 (81.4) 542 (88.1) 478 (91) 123 (87.2) 42 (89.4)

p value 0.10 0.36 0.05 0.48 0.93

Vomiting 29 (67.4) 376 (61.1) 338 (64.4) 85 (60.3) 36 (76.65)

p value 0.54 0.19 0.41 0.47 0.50

Photophobia p value 36 (83.7) 539 (87.6) 485 (92.4) 132 (93.6) 42 (89.4) 0.16 0.08 0.02 0.13 0.88

Phonophobia 35 (81.4) 497 (80.8) 431 (82.1) 115 (81.6) 38 (80.9)

p value 0.99 0.61 0.60 0.95 0.92

Underweight (<18.5), normal weight (18.524.9), overweight (2529.9), obese (3034.9), and morbid obesity (>35). Numbers in parentheses denote percentages.

HBD, but obesity and morbid obesity was more frequent in HBD than PWM. Other observations, such as in the study by Mattson et al. [7] in a Swedish population, support the findings of our study, where obesity was not more frequent in PWM compared with the general population. A review of 11 epidemiological studies by Keith et al. [13] concluded that obesity was associated with an increased likelihood of headache or severe headache. Although this analysis showed a possible association, the included studies had different methods of ascertainment of headache and the used controls were different, which makes them difficult to compare. An important methodological aspect in our study that validates and strengthens our observation is the type of controls. While Bigal et al. [5] used subjects from the general population without any formal screen, we used subjects recruited in a blood bank with a formal assessment to corroborate that they were healthy. In addition, these subjects were paired with PWM by gender, which is usually the main confounder in a disease like migraine, which is more prevalent in females. Another consideration in our study is the inclusion of patients with nonchronic migraine, considering that some studies in the past have linked obesity mainly with chronic daily headache [8, 14]. Winter et al. [6] explored the association of BMI with migraine characteristics in 12,613 patients. This study reported that PWM with a BMI 135 had an increased risk of high migraine frequency. Also Bigal et al. [5] reported an association between BMI and the frequency of migraines. In their study, 5.8% of PWM with overweight, 13.6% of PWM with obesity, and 20.7% of PWM with morbid obesity had between 10 and 15 attacks per month. Our study did not show any association of disability and severity of episodic migraine with BMI. We believe that the main explanation for our results is the exclusion of
138
Eur Neurol 2010;64:134139

patients with chronic headache, which is the group of patients where a potential association between obesity and high frequency of headaches has been reported [5, 14, 15]. Some studies have described an association between BMI and some clinical characteristics of PWM. Horev et al. [16] reported that the frequency of aura was more frequent in patients with overweight. The main observation of this study is controversial because of the reduced sample size, lack of controls and the selection of the sample. In keeping with previous studies with stronger methodology [6] than the study of Horev et al. [16], our study did not find any association between the presence of aura and BMI. Winter et al. [6] reported that those PWM with a BMI 135 had an increased risk of photophobia and phonophobia. Our study in part confirms this association; there was a trend in some symptoms like nausea and photophobia to be more frequent in patients with overweight, and this was statistically significant (table6). MA is a primary headache disorder that affects about 30% of migraine sufferers [17]. One of the most interesting findings in our study was the high frequency of MA. In the majority of studies performed in the general population, the frequency of MA varies from 15 to 30%. We assume that this finding could be explained by a reference bias. Both institutions are national reference centers for complex neurological conditions and this could be the reason for the high prevalence of MA. On the other hand, the high frequency of accompanying symptoms in patients with aura is worth noting; this observation could be explained in the same way as the high frequency of patients with aura. It is possible that the most complicated cases are referred to specialized centers in Mexico. This preponderance of patients with MA with more complications and more symptoms has been seen in studies of epilepsy clinics and tertiary centers in other populations [18].
Tllez-Zenteno et al.

Our study had some strengths and limitations. The main strength in our study is the ascertainment method of migraine using the international criteria of the IHS. Another strength of our study is the ascertainment of HBD, who were all screened to corroborate that they were healthy. Previous studies used the general population as controls, mainly in surveys, including populations that could have other comorbidities. Finally, the measure of weight and height were standardized and done at the same time as the inclusion of patients. Previous studies used self-reported weight and height, which can cause potential differences [19]. A limitation of our study is the lack of inclusion of patients with chronic daily headache, which is the group where all the potential associations between BMI and migraine have been reported. Our study mainly includes patients with episodic migraine,

but not chronic migraine, and our results should be applied fo this specific population. Another limitation is the potential reference bias in our study. The high prevalence of MA could indicate a potential reference bias, because the study was done mainly in two tertiary centers in Mexico City. Unfortunately, there are not many epidemiological studies in Mexico [20] exploring different settings to explain variability in the rate of MA, considering that the majority of studies have shown that the prevalence of MA ranges from 10 to 15%.
Acknowledgments
Dr. Tllez-Zenteno receives grants from the Royal University Hospital Foundation in Saskatoon and the University of Saskatchewan.

References
1 Lipton RB, Bigal ME: The epidemiology of migraine. Am J Med 2005; 118(suppl 1):3S 10S. 2 Peterlin BL, Rapoport AM, Kurth T: Migraine and obesity: epidemiology, mechanisms, and implications. Headache; epub ahead of print. 3 Peterlin BL, Rosso AL, Rapoport AM, Scher AI: Obesity and migraine: the effect of age, gender and adipose tissue distribution. Headache; epub ahead of print. 4 Arulmozhi DK, Veeranjaneyulu A, Bodhankar SL: Migraine: current concepts and emerging therapies. Vascul Pharmacol 2005; 43:176187. 5 Bigal ME, Liberman JN, Lipton RB: Obesity and migraine: a population study. Neurology 2006;66:545550. 6 Winter AC, Berger K, Buring JE, Kurth T: Body mass index, migraine, migraine frequency and migraine features in women. Cephalalgia 2009;29:269278. 7 Mattsson P: Migraine headache and obesity in women aged 4074 years: a populationbased study. Cephalalgia 2007;27:877880. 8 Bigal ME, Lipton RB: Obesity and chronic daily headache. Curr Pain Headache Rep 2008;12:5661. 9 Morillo LE, Alarcon F, Aranaga N, Aulet S, Chapman E, Conterno L, Estevez E, GarciaPedroza F, Garrido J, Macias-Islas M, Monzillo P, Nunez L, Plascencia N, Rodriguez C, Takeuchi Y: Prevalence of migraine in Latin America. Headache 2005; 45: 106 117. 10 Tellez-Zenteno JF, Garcia-Ramos G, Zermeno-Pohls F, Velazquez A: Demographic, clinical and comorbidity data in a large sample of 1,147 patients with migraine in Mexico City. J Headache Pain 2005;6:128134. 11 Weder-Cisneros ND, Tellez-Zenteno JF, Cardiel MH, Guibert-Toledano M, Cabiedes J, Velasquez-Paz AL, Garcia-Ramos G, Cantu C: Prevalence and factors associated with headache in patients with systemic lupus erythematosus. Cephalalgia 2004; 24: 1031 1044. 12 Lipton RB, Stewart WF, Sawyer J, Edmeads JG: Clinical utility of an instrument assessing migraine disability: the Migraine Disability Assessment (MIDAS) questionnaire. Headache 2001;41:854861. 13 Keith SW, Wang C, Fontaine KR, Cowan CD, Allison DB: BMI and headache among women: results from 11 epidemiologic datasets. Obesity (Silver Spring) 2008;16:377383. 14 Scher AI, Stewart WF, Ricci JA, Lipton RB: Factors associated with the onset and remission of chronic daily headache in a population-based study. Pain 2003;106:8189. 15 Bigal ME, Lipton RB: Obesity is a risk factor for transformed migraine but not chronic tension-type headache. Neurology 2006; 67: 252257. 16 Horev A, Wirguin I, Lantsberg L, Ifergane G: A high incidence of migraine with aura among morbidly obese women. Headache 2005;45:936938. 17 DAndrea G, Bonavita V, Rigamonti A, Bussone G: Treatment of migraine with aura: comments and perspectives. Neurol Sci 2003;23:271278. 18 Gracia NM: Burden and management differences between migraine with and without aura in neurology clinics. Neurologia 1999; 14:383388. 19 Katsnelson MJ, Peterlin BL, Rosso AL, Alexander GM, Erwin KL: Self-reported vs measured body mass indices in migraineurs. Headache 2009;49:663668. 20 Morillo LE, Alarcon F, Aranaga N, Aulet S, Chapman E, Conterno L, Estevez E, GarciaPedroza F, Garrido J, Macias-Islas M, Monzillo P, Nunez L, Plascencia N, Rodriguez C, Takeuchi Y: Clinical characteristics and patterns of medication use of migraneurs in Latin America from 12 cities in 6 countries. Headache 2005;45:118126.

Body Mass Index in Migraine

Eur Neurol 2010;64:134139

139

Copyright: S. Karger AG, Basel 2010. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.