Causes of migraine headache

Published:

INTRODUCTION

Migraine is a neurological disorder which affects almost 10% of the world's population (Woeber et al., 2007). In 2003, the World Health Organization (WHO) estimated the number of migraineurs worldwide at 303 million people. A similar study in 2004 found that approximately 20 million migraine attacks occur every day (Forshaw, 2003). Individuals who suffer from migraine headaches carry the burden of pain and suffering that can lead to an impaired quality of life. At a community level, migraine headaches can also be problematic because of absences from work or decreased productivity from migraine sufferers (NINDS Migraine Information Page). As a result of the problems associated with migraine headaches, there is significant interest in discovering the triggers for migraines.

A migraine is a specific type of headache characterized by altered bodily perceptions, pulsing pain in the cranial region, and nausea (Forshaw, 2003). Most migraines are unilateral, meaning they affect only one side of the head, and the pain is usually localized to a very specific area (Forshaw, 2003). A typical migraine can last anywhere from 4 to 72 hours. The most frequent symptoms of a migraine include nausea, vomiting, and increased sensitivity to sensory input (Gallagher et al., 2002). Most commonly, individuals affected by migraine headaches have increased sensitivity to light (photophobia) and sound (phonophobia) (Gallagher et al., 2002). A smaller percentage of migraine sufferers report an aura which accompanies their migraine. An aura usually consists of atypical visual, olfactory, or other sensory experiences that give the individual some forewarning that a migraine will soon occur (Gallagher et al., 2002).

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

The diagnosis of migraine can be difficult, especially because this specific type of headache mimics several other types. Migraines are also habitually written off by those who experience them infrequently as a result of the cold or flu (Gallagher et al., 2002). Because of these difficulties, migraines are often underdiagnosed or misdiagnosed (Lyons, 2007). The International Headache Society (IHS) has laid the foundation for the diagnosis and classification of migraine headaches. According to the IHS, there are seven different classes of migraines (Headache Classification Subcommittee, 2004). These include migraine without aura, or common migraine, migraine with aura, hemiplegic migraine, childhood periodic syndromes which are precursors of migraine, retinal migraine, complications of migraine, and probably migraine.

Although there are many classifications of migraine headaches, there are really only two methods of diagnosis, which correlate with the two main types of migraine: migraines that occur with aura those which do not. Migraine without aura can be diagnosed using the "5, 4, 3, 2, 1 criteria": 5 or more attacks, 4 hours to 3 days in duration, 2 or more of - pulsating quality, unilateral location, moderate to severe pain, aggravation by or avoidance of routine physical activity, and 1 or more accompanying symptoms which could include nausea and/or vomiting, photophobia, or phonophobia (Headache Classification Subcommittee, 2004). In the method of diagnosing migraine with aura, only two attacks are necessary to make the diagnosis.

Although migraine headaches are experienced by individuals from many different lifestyles and backgrounds, there are some demographics in which migraine is more commonly seen. Although migraine headaches are equally prevalent in male and female prepubescent patients, 75% of adult patients are women (Lay et al., 2009). Migraines become more common with age, though 98% of patients experience their first migraine before the age of 50 (Forshaw, 2003). Migraines are also thought to be genetically linked, as 70% of migraine patients have some other, first-degree relative (e.g. brother) who has experienced migraine headaches (Forshaw, 2003).

Perhaps the most intriguing question pertaining to migraines is that no one knows what causes them. To date, research has not been able to definitively discern which of the suspected triggers of migraine may actually cause the headaches, nor has anyone determined which of the suspected triggers might play the largest role in producing migraines (Woeber et al., 2007). The list of suspected triggers for migraines is extensive, including, but not limited to: weather, missing a meal, stress, alcohol, various types of food and changes in sleeping patterns (Forshaw, 2003). Women have also reported menses as a trigger, and studies of both men and women have also shown that environmental factors and even certain activities, such as using a computer for too long, can bring on a migraine (Woeber et al., 2007). Because it would be impossible to investigate each and every one of these suspected triggers, the comparison of this studied has been narrowed to include sleep-related, hormonal and food triggers.

Sleep Disorders (Amelia Van Handel)

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

The role of sleep in migraine has not been fully explained (Kelman, 2005). Studies have determined that disturbed sleep patterns may trigger a migraine attack, and it is also widely accepted that sleep can alleviate and even terminate a migraine (Blau, 1982). Although the exact relationship between changes in sleep patterns and migraine is unclear, many researchers believe there is a correlation (Woeber et al., 2007; Kelman, 2005).

Much of the argument for the relationship between migraine and sleep disorders stems from the preponderance of migraine in the morning hours (Fox, 1998). Patients with sleep disorders are far more prone to have morning headaches, and chronic migraine sufferers often experience morning migraines after interruptions in their sleep cycle. This has led researchers to speculate that the circadian clock plays a role in migraine pathophysiology, though no one has yet determined why changes in sleep pattern are a trigger for migraine headache (Cohen, 2005).

Several sleep disorders are speculated to be triggers for migraine. Insomnia, which involves difficulty getting to sleep or staying asleep, is one of the most researched causes for sleep-related migraine (Pallesen, 2001). Excessive daytime sleepiness (EDS) is also commonly associated with migraine. EDS is defined as difficulty maintaining a desired level of wakefulness, and patients diagnosed with the disorder often experience migraine headaches after dosing off during the day (Young, 2004). EDS is relatively prevalent in the general population, experienced by about 10-20% of individuals, and these percentages increase in the very young and very old. (Hasler, 2005). EDS is usually caused by poor sleep quality at night, which can sometimes be associated with insomnia (Carskadon, 1993). To a lesser extent, narcolepsy (symptoms similar to EDS) and sleep apnea (pauses in breathing during sleep which cause an individual to wake up sporadically) have been studied to determine whether or not they might be associated with migraine (Bixler, 2005).

For more than 100 years, medical personnel and researchers alike have noticed an association between sleep problems and headaches (Sahota, 1990). One of the main causes of confusion, however, is whether the headaches are the cause or the result of disrupted sleep. While interruptions in sleep patterns can cause migraine headaches to become more prevalent, migraines can have the same affect on the sleeping disorders themselves (Paiva, 1997). The determination of which comes first, the sleep disruption or the migraine, is the subject of much current research (Woeber, 2007; Lee, 2009).

Hormones (Kelly Pritzl)

Previous research indicates that headaches are three times more common in adult females than adult males. (Evans et al, 2000) The reason for this staggering statistic could be due in part to differences in male and female hormones and levels of hormones. (Evans et al, 2000) The major male and female hormones are estrogens and androgens. Men produce significantly more testosterone, a type of androgen, per day than women (7 mg vs. .5 mg), while women produce more estrogen per day than men. A woman experiences more fluctuations in hormone levels during her life than a male does. During these times of fluctuation, many women will have an increased incidence of migraine, suggesting that fluctuations in hormone levels play a role in the onset of migraine. (Lee, 2009) Some of the hormones that may be involved in the onset of migraine are estrogen, progestin, androgens, testosterone, and serotonin. (Glass, 2009) The mechanisms by which these hormones are involved are not clear, but there is strong evidence for the role of hormones in precipitating migraine attack. (Glass, 2009)

There is particular evidence for the role of hormones in causing migraine attacks in women. Before puberty, males and females tend to experience migraines at the same rate, there is a sharp increase in the number of girls over boys who experience migraine at the mean age that girls begin menstruating. (Dzoljic et al, 2002) Pregnancy also seems to have an effect on the occurrence of migraines. During pregnancy, there is an increased level of estrogen in the body. Many women either experience an absense of headache when they otherwise suffered from migraine on a regular basis, or they experienced an increase in frequency of headache when they typically did not have migraines. (Robbins, 2002) Another instance of the role of hormone involvement in migraine attack in women is the increase in incidence of migraine as women near menopause, a time of decreased estrogen production. (Robbins, 2002)

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

While there are many different hormones that may have an effect on incidence of migraine, the main focus of this research will primarily be on the mechanisms by which estrogen may induce migraine, with respect to different times in a woman's life estrogen levels fluctuate, such as during menstruation, pregnancy, and menopause. Estrogen is a type of steroid hormone and is considered the primary female sex hormone responsible for regulating the normal sexual and reproductive development in women. (Robbins, 2002) Organ systems such as the musculoskeletal system, the cardiovascular system, and the brain are affected by estrogen. (Robbins, 2002) There are two approaches to the current understanding of the role of estrogen in migraines. One type is estrogen withdrawal headache. This happens after a severe drop in estrogen levels in the body, such as during menstruation, during menopause, or post-partum. The second type is exogenous hormone induced headache. This occurs during or after the use of oral contraceptives or hormone replacement therapy. (Kibler et al, 2005)

A comparison of studies that examines the correlation between levels of estrogen during certain periods of a woman's life and the incidence of migraine will allow better understanding of the function of this hormone as a cause of migraine. Very little is known about the way in which estrogen actually precipitates migraine, but with an enhanced understanding of the current research that has been done, future research will be promoted on a topic that affects such a considerable proportion of migraineurs.

Food (Brandon Pellerin)

Various foods have been suspected of triggering migraines for decades (Grant, 1979, Peatfield et al., 1984). In susceptible people, certain foods and particular compounds contained in these foods are believed to induce trigeminovascular (warning system to protect the brain from tissue injury and toxins) neurons to release neurotransmitters such as calcitonin, gene-related peptide and substance P. The release of these neurotransmitters leads to vasodilation (widening of blood vessels), mast cell degranulation (release of molecules from secretory vesicles called granules), increased vascular permeability (capacity of a blood vessel wall to allow the flow of small molecules), and meningeal edema (accumulation of fluid within the meninges) resulting in neurogenic inflammation (release of inflammatory mediators from neurons) (Sun-Edelstein and Mauskop, 2009). Many common foods such as wheat, eggs, beef, and corn are documented migraine triggers (Grant, 1979). However, the most prevalent food precipitants of migraine are alcohol, chocolate, coffee, fatty foods and artificial sweeteners (Peatfield et al., 1984).

Various compounds present in common foods are suspected to play important roles in the triggering of migraines. Certain amines such as tyramine and phenylethylamine are thought to be precipitators of migraines and are present in alcohol and chocolate (Sun-Edelstein and Mauskop, 2009, Marcus et al., 1997). Caffeine, also present in chocolate, is believed to be the culprit of coffee's capacity of being a trigger (Sun-Edelstein and Mauskop, 2009). Artificial sweeteners themselves such as aspartame and more recently sucralose, have been subjects of research as to their ability to precipitate migraines (Sun-Edelstein and Mauskop, 2009, Bigal and Krymchantowski, 2006).

However, not all migraineurs exhibit sensitivity to food and those that do are not equally affected by each trigger. The food that affects one person may not be the same food that triggers migraine in another, while at the same time a third person may be affected by both. The inconsistency of results keeps food as a continued subject of debate and study in migraine precipitation.

The purpose of this research was to determine the role, if any, that sleep disorders, hormones, and food play in the triggering of migraine headaches. The goal of this study was to determine if there is any validity to the conjectures that these are triggers for migraine and if so, which trigger plays the largest role in determining whether or not migraine will occur.

To answer these questions, data was gathered from primary sources by searching PubMed and Biological Abstracts. From these studies, each researcher conducted his or her own analysis of the data found within a particular subtopic to determine what correlation that specific cause might have with migraine headache. The information collected in this portion of the research was then combined to determine the relative relationship between the triggers and migraine, using correlation data and p-values to determine which was the overriding cause of migraine headaches.

METHODS

Sleep Disorders (Amelia Van Handel)

In order to find articles relating sleep disorders to migraine, the database Biological Abstracts was used. This database was chosen because Biological Abstracts includes articles from all science-based subject areas and includes many reviews and other literature forms, which proved helpful for background or supplemental information. This database was also chosen because it was a good resource for primary research sources relating to the specific subject matter.

Having chosen this database, the search was initiated using keywords relating to the topic. Initially, the subtopic for this section of the research was sensory stimuli, so the search began with the keywords "migraine headache*" and "light". The word "headache" was truncated so as to provide a larger base of results. This returned 31 articles, but after looking through them, it was determined that most of them were reviews. The lack of primary research led to a search other types of stimuli, common food triggers, and even specific symptoms of migraine associated with the senses. When none of these provided the intended results, it was determined that this subtopic should be changed. Leaving the idea of sensory stimuli behind completely, a search was performed using the words "migraine" and "sleep", which yielded 38 articles. The number of articles and the quality of the source material fit the needs of the research, and thus the search was completed.

With a manageable number of articles, those which were most pertinent to the subtopic of sleep disorders were chosen for further analysis. The initial 38 were narrowed by removing those articles which were not primary research. Although the reviews and other literature forms would be helpful for background information, they would not be useful in making comparisons and finding correlation. The article selection was further narrowed by looking for those articles which contained the metrics the research would focus on. With these parameters in place, only 11 articles remained, a number which was determined to be appropriate for drawing conclusions about the correlation between certain triggers and migraine.

The metrics of focus chosen for this research pertained to the quality of patients' sleep and the correlation this had to the number of migraines they experienced. Epworth Sleepiness Scale scores and Pittsburgh Sleep Quality Index made it possible to measure the quality of patients' sleep. The association between migraine and sleep disorders was reported in the chosen articles, which aided in the determination of the relationship between sleep quality and the number of migraines experienced.

Hormones (Kelly Pritzl)

All of the primary research articles relating to hormones as a cause of migraines were found online through the search engines Biological Abstracts and PubMed. The same process for finding citations was used with both engines. The search strategy consisted of first examining the results when "migraines" was entered into the search box. This yielded far too many results; the goal was to restrict the number of articles relevant to hormones as precursors of migraines to 40 or less. In order to refine the search, the entities "migraines AND hormones" were entered into the search box. To further refine the search, "migraines AND estrogen" was entered in and results were limited to "only clinical trials" and "articles in English only".

After gathering 40 relevant research articles, five articles within these were found containing specific criteria in order to properly conduct the meta-analysis within the topic of hormones and across the three topics of sleep-induced migraines, nutrition and migraines, and hormones and migraines. The criteria for selecting the five best articles included: relevant and useful primary data, p-values, similar subjects and number of subjects, similar methods of data collection, and recentness of publication.

All of the research articles relating to increased or decreased levels of estrogen and incidence of migraine were found online using the search databases Biological Abstracts and PubMed. The same process for finding articles was used with both search engines. The search strategy consisted of entering "estrogen AND migraine" and searching with one other term specifically related to each of the different reproductive time periods being studied, including, "prepuberty", "menstruation", "combined oral contraceptives", "pregnancy", and "menopause". This strategy resulted in a number of relevant primary research articles. The search was further narrowed down by reviewing the articles, keeping those that contained specific criteria, and eliminating those that did not. The criteria for selecting the five articles were similar to the selection criteria for the first five articles: relevant and useful primary data, similar subjects and number of subjects, similar methods of data collection, and recentness of publication. The studies were then cross-compared and analyzed, examining inconsistencies and similarities between the data.

Food (Brandon Pellerin)

To find relevant articles on the subject of food triggered migraines, the electronic databases Biological Abstracts and PubMed were used. Biological Abstracts was used using a title search for the word "migraine*" with a secondary title search of "food* or diet*". This search turned up 30 records. The asterisk is used to search for any result which contains the root word. More specific searches were done by a title search of "migraine*" with topic searches of "chocolate*", "alcohol*", "caffeine*", "aspartame*", or "sucralose*". Similar methods were used using the PubMed database with the exception of the use of the asterisk and the differentiating of topic searches and title searches. General searches were done using "migraine and diet" and "migraine and food". More specific searches were done using the same keywords used in biological abstracts, joined by the "and" limiter.

The articles searched for were published in relevant scientific journals and pertained to the topic of food and its potential to precipitate migraines. Articles that were chosen contained two types of data. One set of data included the results of general surveys that were done to ascertain details of migraineur's attacks, such as various triggers. The second type of data obtained were results from studies of specific foods documented as triggers for migraine. When an article seemed to contain useful information and was able to be accessed online, it was saved as a PDF file for future reference.

The data collected from the general surveys consisted of questionnaires asking for details of subject's migraines. A vast amount of information was collected in these surveys such as the type of migraine (with or without aura), associated symptoms of migraine (photophobia, nausea, etc.), frequency, duration, and so on. The information important to this study was that concerning precipitants of migraines. Each survey documented the reported triggers of each subject if a trigger existed. This information is used to ascertain the prevalence of foods as triggers within the population that suffer from migraines.

The second data set used results from studies that sought to test whether suspected foods did indeed trigger migraines. The studies relied on correlating migraine occurrences with the consumption of particular foods. The studies analyzed diet and migraine diaries kept by the subjects. The diet records often required subjects to record all food consumption and the time at which it was consumed. Likewise, the migraine diaries required subjects to document the occurrence of migraines and details regarding them such as severity, duration, type, associated symptoms, etc. The studies analyzed the results by comparing the onset of migraine with the consumption of a particular food. If there was a significant increase in the amount of migraines after consumption of a particular food, it was reasonable to conclude it played a role in the triggering of the migraine.

RESULTS

Sleep Disorders (Amelia Van Handel)

In conducting this research, the results of eleven studies were analyzed. These studies were interested in finding the correlation, if one existed, between sleep disorders and migraine. Nine of the eleven articles documented research conducted by asking patients to record the quality of their sleep and the number of migraine headaches they experienced, either in diary format or by answering questions in a comprehensive questionnaire (Woeber et al., 2007; Alstadhaug et al., 2007; Barbanti et al., 2007; Peres et al., 2005; Calhoun et al., 2006; Heng et al., 2006; Pakalnis et al., 2009; Seidel et al., 2009; Kelman, 2007). The two remaining articles focused on removing the stimulus - i.e. sleep problems - by providing targeted behavioral sleep invention and massage therapy, respectively. (Calhoun et al., 2007; Lawler et al., 2006). The researchers then analyzed whether or not there was an improvement in headache frequency to determine if sleeping disorders were correlated with migraine.

In order to conduct research on the relationship between sleeping disorders and migraine, only patients who suffered from both conditions could be included in the studies. To determine the level of sleep disruption, three studies measured excessive daytime sleepiness (EDS) as a function of a score on the Epworth Sleepiness Scale (Barbanti et al., 2007; Peres et al., 2005; Seidel et al., 2009). A score of 10 or higher on the Epworth Sleepiness Scale indicated EDS. The first study (Barbanti et al., 2007) found that EDS was more common in migraineurs than in controls (14% vs. 5%), and the second study (Peres et al., 2005) found EDS occurred in 85% of chronic migraine sufferers. In the same study, dozing off was a headache trigger in 30% of all patients and 70% of patients with EDS. In both of these studies, patients who presented with EDS had more frequent migraines (Barbanti et al., 2007; Peres et al., 2005). The third study (Seidel et al., 2009) found no correlation between daytime sleepiness and migraine headache, nor between fatigue and migraine headache, though it did notice a decrease in the quality of sleep in migraine sufferers.

In the same three studies, the quality of sleep was measured using the Pittsburgh Sleep Quality Index (Barbanti et al., 2007; Peres et al., 2005; Seidel et al., 2009). An overall score of greater than 5 separated poor sleepers from good sleepers on this scale. In the first two studies (Barbanti et al., 2007; Peres et al., 2005), about 90 percent of patients diagnosed with EDS were also categorized as poor sleepers using the Pittsburgh Sleep Quality Index. This provided further evidence that poor or inadequate sleep and migraine headaches often occurred in the same patients. The third study (Seidel et al., 2009) did not see significant difference in the daytime sleepiness of migraine sufferers versus healthy controls, but did find that PSQI total score was highest in patients with frequent migraine and lowest in controls (P = 0.04).

Three of the eleven studies chosen focused on migraine in a particular population. One study interpreted the correlation between migraine and sleep disturbance in children (Heng et al., 2006). The study was conducted by asking parents to answer questions relating to their child's sleeping habits and quality of sleep. This study found that individuals with migraine had higher total sleep (P = .02), sleep delay (P = .03), and daytime sleepiness scores (P = .001) than their healthy control counterparts. The study also concluded that patients with more severe headaches had greater sleep disturbance than patients with mild migraine (Heng et al., 2006). Another study chose to look at sleep disorders and migraine in adolescents, with specific emphasis on the role of serotonin (Pakalnis et al., 2009). The study concluded that patients with sleep problems were significantly more likely to be migraine sufferers. Also, as sleep problems became more prevalent, headache frequency increased (from episodic migraine to chronic migraine). The study did not find an association between serotonin levels and migraine, nor between serotonin levels and sleep disorders (Pakalnis et al., 2009). The last article with a specific population focus studied the prevalence and spectrum of sleep problems in women with transformed migraine (Calhoun et al., 2006). In this study, sleep quality was measured by asking patients to record whether they felt "refreshed" or "tired" upon waking. None of the 147 patients recorded feeling "refreshed", and 83.7% recorded feeling "tired" (Calhoun et al., 2006). Thus, researchers concluded that there was a correlation between migraine and nonrestorative sleep, though they were inconclusive as to whether the migraines were the cause or the effect of the sleep problems (Calhoun et al., 2006).

The final two studies, which attempted to remove the sleep disruptions by using either behavioral sleep intervention or massage therapy, found statistically significant data, though one study was more successful than the other. The researchers who used behavioral sleep intervention found a significant decrease in headache frequency and intensity after successful sleep modification (Calhoun et al., 2007). They were also able to revert chronic migraineurs to episodic migraineurs after improving the quality of sleep in their patients. By the final visit, 48.5% of those who had received behavioral sleep modification instructions had reverted to episodic migraine (Calhoun et al., 2007). The study which used massage therapy as a means of removing sleep disruptions (Lawler et al., 2006) found that, in comparison to controls, participants that were exposed to massage therapy experienced fewer migraines. Patients' sleep quality also improved during the experimental period and in the three weeks following. The massage therapy did not, however, increase sleep quantity, which had been an initial goal of the researchers (Lawler et al., 2006).

On the subject of correlation, all but one of the articles chosen for analysis found a correlation between sleep disorders and migraine, though many were hesitant to state definitively that the sleeping disorders were the direct cause of the migraine. Two studies found that patients with excessive daytime sleepiness and/or insomnia experienced more migraines after a night of restless or inadequate sleep (Alstadhaug et al., 2007; Barbanti et al., 2007). These migraines were more likely to occur during the morning hours (Alstadhaug et al., 2007). Conversely, another study which focused on EDS noticed a correlation between fatigue and migraine, but they did not believe the results to be conclusive enough to state whether migraines lead to EDS or if EDS is the primary condition leading to migraine (Peres et al., 2005). The three studies which had a specific focus group (children, adolescents, or women) found statistically significant evidence that migraine and sleep disorders were correlated within a specific population, but again, they were hesitant to state definitively whether the migraine was the cause or the effect of the sleep problems (Calhoun et al., 2006; Heng et al., 2006; Pakalnis et al., 2009). Two studies were tracking several different sleep disorders, but found their results to be inconclusive in determining which sleep disorder was most correlated with migraine (Woeber et al., 2007; Kelman, 2007). The researchers did, however, come to the conclusion that tiredness increased the risk of headache and migraine (headache ratio increased from 0.689 to 1.184 in cases where patients were tired) (Woeber et al., 2007). Research also showed that sleep disturbance was one of the most common triggers for migraine (49.8% of participants reported sleep related triggers) (Kelman, 2007). The studies which were focused on removing the sleep disruptions found that improved sleep quality was correlated with fewer migraine headaches, though to what degree these were correlated they could not say (Calhoun et al., 2007; Lawler et al., 2006)

Only one study (Seidel et al., 2009) did not find a correlation between sleep disorders and migraine. While the study did note that quality of sleep seemed to be lower in migraine sufferers than in healthy controls, the study found no correlation for fatigue or daytime sleepiness (Seidel et al., 2009).

Hormones (Kelly Pritzl)

The purpose of the individual meta-analysis was to determine when hormones were most involved in the precipitation of migraine. Two of the studies used only females as subjects. (Dezoljic, 2002 and Kibler, 2005) The subjects in two other studies consisted of males and females with medically diagnosed cases of migraine. (Kelman, 2007 and Rasmussen, 1993) The subjects in the remaining study were self-reported male and female migraineurs (Russel, 1996) All of the studies were researching adults. The mean age of subjects for all the studies was the mid-thirties. (Dezoljic, 2002; Kelman, 2007; Kibler, 2005; Rasmussen, 1993; Russell, 1996)

The methods used in all of the articles were very similar. Four of the studies conducted a clinical examination to confirm a diagnosis of migraine and were followed by a questionnaire or an interview to collect data on lifestyle of the subjects and possible causes of their migraines. One of the studies involved self report of migraine status and if the subject indicated positively, they were issued a questionnaire. (Russell, 1996)

In all of the studies, incidence of migraine caused by fluctuations in hormones was overwhelmingly more prevalent in females than by males. This indicates that female sex hormones, such as estrogen, play a significant role in the onset of migraine.

In the second phase of the individual estrogen and migraine meta-analysis, the purpose was to examine differing levels of estrogen as a cause for migraine as it is classified by the International Headache Society criteria across different times of the life cycle of a female migraineur. Various stages in the life cycle in which levels of hormones differ, particularly estrogens, include before puberty, during menstruation, while consuming oral contraceptives, during pregnancy, and at menopause. The times when there is an increased level of estrogen include, 1-2 days before menstruation, during periods of actively taking combined oral contraceptives, and during pregnancy. This corresponds to the research findings of greater incidence in migraine during these stages of life. (Sances et al; MacGreggor et al; Machado et al)

The three studies supporting the relationship between increased levels of estrogen and increased incidence of migraine were conducted in a similar way. Two of the studies examined a medium sized sample of migraineurs. (Sances et al, 2003 and MacGreggor et al, 2004). The remaining study examined 400 women, a much larger sample in comparison. (Machado et al, 2010). In all of the studies, the female subjects were drawn from a population of individuals undergoing routine clinical examinations and attending either an outpatient clinic, a hospital, or a family planning clinic. (Sances et al, 2003; MacGreggor et al, 2004; Machado et al, 2010).

The two studies examining the relationship between decreased levels of estrogen and increased incidence of migraine were conducted in a similar way. (Wang et al, 2002 and Arefeh and Russell, 1994). Both studies administered questionnaires and relied on self-report of migraine status and were followed up by a face to face interview and clinical examination by neurologists. Some inconsistencies between the articles include the age of the participants, the way in which the questionnaires were administered, and demographics. The difference between the mean ages of the participants in each of the studies is expected, given the time period females go through puberty (mean age= 12) (Arefeh and Russell, 1994) versus the time period females experience menopause (mean age= 44). (Wang et al, 2002), but still may cause outliers to play a role in the data collected. Another matter to consider is the difference in demographics between the two studies. The childhood/pre-puberty research was conducted with a random sample of children in one school district and community. (Arefeh and Russell, 1994) The menopause research was conducted in two different townships in Taiwan. It is unknown what affects this might have caused in the collection of data. (Wang et al, 2002)

In the second phase of meta-analysis, there were conflicting findings between preliminary data and the research articles. Preliminary research on the topic of incidence of migraine before puberty said that males and females experience headache at the same rate. (Dzoljic et al, 2002) Further research indicates males actually experience migraine more frequently than females before the age of twelve, and subsequently, the roles reverse. (Arefeh and Russell, 1994).

Overall, current research agrees that there seems to be an increase in the incidence of migraine headache during periods of the female cycle when there is either a sharp increase or sudden decrease in the levels of hormone secreted by the body, with the incidence of migraine being most common for women between the ages of 35-54. This indicates a direct relationship for the role of estrogen in precipitating migraine attack. Recommended future research should examine whether there is a specific balance of hormones needed in the body and look for ways to control rapid fluctuations in levels of hormones in order to alleviate the symptoms female migraineurs suffer during critical reproductive periods in their life cycle.

Food (Brandon Pellerin)

Four articles were used that studied triggers of migraine in various populations (Kelman, 2007, Spierings et al., 2001, Takeshima et al., 2004 and Chabriat et al., 1999). The studies used surveys to gather information about a population. The individuals chosen for the surveys were either random people or known migraine sufferers. In the case of the random surveying, individuals that reported having migraines were instructed to complete a detailed migraine questionnaire. Three of these studies (Kelman, 2007, Spierings et al., 2001 and Chabriat et al., 1999) show evidence of food as a precipitant for migraine. Of these three studies, at least 26% of individuals documented food as being a trigger. The fourth study (Takeshima et al., 2004) shows little evidence of this as it reports less then 1% of surveyors listing food as a trigger. Of these four studies, two (Kelman, 2007, Spierings et al., 2001) listed alcohol as a separate category and reported about 40% of individuals claiming alcohol as a trigger for migraine.

Alcohol, chocolate, and artificial sweeteners are among the most often cited food triggers for migraine (Sun-Edelstein and Mauskop, 2009 and Peatfield et al., 1984). The high frequency of these claims makes these subjects of particular interest. Articles were obtained that studied these particular triggers in order to ascertain whether they did indeed trigger migraine.

Two case studies were found that documented the artificial sweetener sucralose as a probable precipitant of migraine (Bigal and Krymchantowski, 2006 and Patel et al., 2006). In the two studies, migraine attacks were documented at least 90% of the time after the individuals consumed a beverage containing the sweetener. In one study this was ascertained through correlating a food diary with the occurrences of migraine (Bigal and Krymchantowski, 2006). The other study (Patel et al., 2006) relied on administering different sodas that contained and did not contain sucralose. It was found that only the sodas containing the sweetener triggered migraines.

Studies were done to determine whether chocolate can act as a migraine trigger (Gibb et al., 1991, and Marcus et al., 1997). Although both studies used similar methods, the results varied. In one double-blind parallel group study, subjects who self identified chocolate as a trigger were given either chocolate or carob as a placebo, both being heavily flavored with mint to disguise any difference in flavor. Of the 12 participants that received chocolate, 5 developed migraines within an average of 22 hours. In contrast, none of those that received carob developed migraines (Gibb et al., 1991). A second study was performed using similar methods and design. However, this study did not purposefully screen for only those that self indentified chocolate as a trigger, rather the group was intended to be representative of a general sampling of migraine sufferers. Also, before administration of the samples the subjects were to follow a diet the excluded vasoactive amine-rich foods. The presence of amines in food is one suspected cause of a food's capacity to induce migraine (Sun-Edelstein and Mauskop, 2009). Of the 64 subjects 11 (17.2%) reported migraine after consuming chocolate, while 26 (40.6%) reported migraine after consuming the carob sample. Within the group of subjects, 11 claimed chocolate as a trigger of migraine. Of these 11, 2 reported migraine after consuming chocolate, while 4 reported migraine after consuming carob (Marcus et al., 1997).

Patients attending the Princess Margaret Migraine Clinic were questioned regarding details of their migraine attacks. Of the 429 patients that had migraine, 18.4% reported sensitivity to all alcoholic drinks, another 11.8% were only affected by red wine, and 28% of all migraineurs reported beer as a precipitant of migraine. Of the 429 patients, 174 (40.5%) did not consume alcohol at all (Peatfield , 1995). The percent of migraineurs listing alcoholic beverages as triggers is significant. The fact that 40% of patients did not consume alcohol means that some sensitivities to the substance may remain undiscovered. Similar results were obtained in a second study. This study was performed at the Montefiore Headache Unit, where details of patients' migraines were taken. Of the 113 patients that suffered from migraine, 65 (57.5%) listed alcohol as a migraine trigger (Lipton, 1989).

A different type of study was performed that tested a group of migraine sufferers that claimed red wine as a trigger for their migraines and not other types of alcohol. Red wine has been cited as a precipitant more often then other types of alcohol due to its higher concentration of suspected migraine precipitating compounds (Sun-Edelstein and Mauskop, 2009). Wine and a diluted mixture of vodka were given to two different groups of patients. It was served cold and in blacked out containers to conceal identity. Of the 11 given wine, 9 suffered from a migraine attack shortly after. In contrast, none of the 8 that were given the vodka mix suffered from an attack (Littlewood et al., 1988). This study gives credence to the claims that red wine contains types of compounds not found in other types of alcohol, or not found in significant amounts to precipitate migraine.

DISCUSSION

Based on the background information, we know that there are several suspected triggers of migraines and some of them seem to be more prevalent than others. After analysis of each individual trigger, we averaged the correlation p-values and found that some of the suspected causes of migraine were statistically significant. Tables of the individual p-values for each suspected trigger can be found in Tables 1-3 in the Appendix. The average was taken of the p-values of each study in the three sections to serve as a comparison to the other suspected causes. The average of all studies was calculated, as well as an average which excluded outlying data. Table 4 is a quick reference for comparing the p-values calculated for this study.

When looking at the average p-values calculated from all data points, the correlation p-values for sleep disruptions, hormones, and food are 0.10327, 0.02792, and 0.09636, respectively. This indicates that hormones have the strongest correlation with migraine, food the second strongest, and sleep disruptions have the weakest correlation to migraine headache.

In contrast, when the outlying data points are removed, the correlation p-values for sleep disruptions, hormones, and food are 0.0274, 0.01365, 0.038, respectively. This leads to the conclusion that hormones have the strongest correlation to migraine, sleep disruptions the next strongest, and food is least correlated with frequency of migraine. However, these p-values are relatively close in magnitude, suggesting that their relationship to migraine is not all that different.

Although the data seems to conclude that hormones are the most likely cause of migraine headache, there are several factors which might have skewed the study. The factors that might have had the strongest impact on the outcome of the study include the method of data collection by primary sources, the subjects included in the original studies, and the interconnectivity of migraine causes.

The method of data collection in the original studies may have influenced the outcome because many of the studies from which data was drawn used self-reported data in the form of diaries, questionnaires, or interviews. Also, several of the studies, many of them under the topic of sleep disorders, used a rating scale that asked participants to rate the frequency or severity of migraines on a numbered scale. This type of data collection leaves a lot of room for error based on subjectivity, the unreliability of recall, and lack of quantitative measurements independent of human interpretation.

Another flaw in the methods of the primary research studied for this report was that it is difficult to tell the difference between migraine and other types of headache. Many headache variations have similar symptoms and there is currently no way to definitively classify a headache as being a migraine. For this reason, some of the reported headaches may not have been migraine-related, but were reported as such.

Additionally, inconsistencies in the data may have arisen from the participants of the original studies from which we drew our data. Most specifically, the entirety of the hormone study excluded male participants. This could very probably have skewed the data because it is already known that 75% of adult migraine sufferers are women, which may have contributed to the higher correlation for the hormonal study (Lay et al., 2009). Moreover, three of the articles analyzed for the sleep disorders section of study focused on particular target groups, including women, adolescents, and children. Another way the subjects of the studies might have affected the outcome of our research arose in the food-related triggers study. Three of the studies analyzed for this section of the report selected migraineurs who had reported a particular trigger and tested for that specific cause. This might have created problems because of expectations of participants or because the researchers already knew these individuals to be affected by that specific trigger instead of testing migraineurs from a wider sampling.

Another possible source of error pertaining to study participants is that some people are more naturally prone to migraine. In this case, these individuals will be more sensitive to so called triggers, independent to any power the trigger itself has to cause migraine. If these types of participants were included in the studies analyzed for this research, their data may have skewed the results as a whole.

A further area of potential fluctuation in the data lies in the interconnectivity of migraine causes. Our study was created under the assumption that there is one main cause of migraine headache, but in reality there may be several triggers working in conjunction with one another. It is also possible that what is perceived as the cause of a migraine might actually be a secondary response to a completely different trigger. For example, hormone levels can be affected by diet or sleep patterns. Also, the specific cause of migraine may be more narrow than our study can conclude. An example of this would be certain foods, because while one type of food may be known to cause migraines, but the specific compound that results in the headache may be unknown.

Knowing these sources of error exist, it cannot be conclusively stated which is the most likely cause of migraine headache based on this data alone. Based on the results of this study, it can be concluded that sleep disruptions, hormones, and food all play a role in precipitating migraine headache, but the extent of that role remains to be known.

Because the results of this study were inconclusive, further research is necessary to determine the most prominent cause of migraine, if one exists. If there is a specific cause that is speculated to be strongly correlated with migraine headache, a study of the mechanism of that interaction would provide insight and credibility to the claim. Also, studies which involve the collection of quantitative data rather than qualitative would have more scientific credibility and would be less likely to be influenced by the same factors that were mentioned previously. Furthermore, a study of the interconnectivity of suspected migraine causes may be more appropriate if there are indeed several triggers working together to cause a migraine. This could be accomplished by executing a study which combines several independent variables and analyzing the results when a subject is exposed to these variables in conjunction with one another.