Association between migraine and irritable bowel syndrome: a population-based retrospective cohort study.

Association between migraine and irritable bowel

syndrome: a

population-based retrospective cohort study

C.-I. Lau

, C.-C. Lin

, W.-H. Chen

, H.-C. Wang

and C.-H. Kao

Introduction

Migraine and irritable bowel syndrome (IBS) are traditionally considered as functional disorders that are characterized by recurrent pain in the absence of detectable organic causes. These seemingly distinctive clinical disorders share many similarities regarding their epidemiology, female predominance, episodic manifestations, trigger factors, benignity, comorbidities and proposed pathophysiology.

Migraine typically presents relapsing episodes of symptoms including headache, nausea, vomiting, sensory hypersensitivity and mood changes that last for hours to days (International Classification of Headache Disorders, ICHD-3) [1]. Similar to migraine, IBS is

characterized by recurrent abdominal pain or discomfort with altered gastrointestinal motility and visceral

hypersensitivity that returns to normal between attacks [2]. Both pain disorders lack detectable organic causes and depend on a set of diagnostic criteria developed by expert consensus, namely the ICHD-3 for migraine and the Rome III criteria for IBS [1,2]. Migraine is a common disorder that affects 10%–12% of the adult population, with approximately two to three times more

women than men being affected [3]. Likewise, IBS is a common disorder found in 10%–15% of the European and North American populations, although its prevalence varies in other parts of the world [4]. A substantial

gender bias toward female predominance has also been confirmed in IBS [5]. In addition, migraine is a disease

with a substantial heritability and multigenic nature. The migraine risk of first-degree relatives of patients who experience migraine with aura is 4-fold higher than that of the general population [6]. Similarly, IBS has been shown to aggregate in families, being two to three times more common in relatives of individuals with IBS than in other people [7].

Several lines of evidence support the notion of a dysregulated cortical excitation_inhibition balance in

migraine [8]. The episodic activation of the trigeminovascular pain pathway leads to sustained sensitization

of the central nervous system (CNS), which may in turn result in altered pain processing. By contrast, an exaggerated pain response to intestinal distension may also underlie the pathophysiology of IBS [9]. This altered central processing of pain is primarily governed by the enteric nervous system (ENS), where visceral hypersensitivity has been reported [10].

Epidemiological studies have shown that migraine and IBS often coexist with overlapped comorbidities such as fibromyalgia, chronic fatigue syndrome, depression, sleep disturbance and anxiety [11,12].

Therefore, it can be speculated that there is an association between migraine and IBS. Indeed, several studies

have reported an increased risk of migraine in IBS patients [5,13–16]. For example, an early study showed that half of the IBS cases were significantly associated with frequent migrainous headaches, but in the general population only 18% of IBS cases showed this association [13]. Similarly, a Polish study demonstrated a high prevalence of frequent headaches, ranging

from 23% to 53%, in IBS patients [17]. A recent

meta-analysis summarized six existing epidemiological studies, and revealed that IBS patients had an overall odds ratio of 2.66 for experiencing comorbid migraine [18]. An emerging model of the brain_gut axis was proposed to explain migraine and IBS as a result of a genetically sensitive nervous system that develops

hyperexcitability over time as a response to multiple environmental and immunological factors [19]. To examine the incidence of IBS in migraine, a nationwide population-based cohort study was conducted

and data from a large nationwide medical database, namely the National Health Insurance Research Database (NHIRD) of Taiwan, were analyzed.

Methods

Data source

On 1 March 1995, the Bureau of National Health

Insurance, Department of Health of Taiwan, consolidated 13 insurance programs into the National Health

Insurance (NHI) program. The NHIRD of Taiwan is

a comprehensive healthcare database covering approximately 99% of the 23 million people of Taiwan. The

NHIRD was established and is maintained by the National Health Research Institutes of Taiwan. The Longitudinal Health Insurance Database (LHID) is a data subset that is provided to scientists in Taiwan for research purposes. The LHID used in this study contained 1 million people randomly selected from all the

population in the insurance claims database for the period 1996–2010. This data set included the annual

claims data for medical service and registry for beneficiaries. The institutional review board of China Medical

University performed a full ethical review of this study (CMU-REC-101-012) to protect the privacy of the patients.

The disease history was recorded using the International Classification of Disease, Ninth Revision, Clinical

Modification (ICD-9-CM), and the data were collected from inpatient and outpatient files.

Study population

Our research design consisted of a population-based retrospective cohort study. Newly diagnosed migraine cases (ICD-9-CM code 346) for the period 2001–2009 were identified in the LHID. The index date of the migraine cohort corresponded to the date of the initial

migraine diagnosis. Migraine patients with IBS history before the index date were excluded (n = 1137). Using a frequency matched by sex and age (per 5 years), individuals who were free from migraine and IBS history

were randomly selected as the comparison cohort. Date and months and the same index year as the matched case were randomly assigned. The major outcome of this study was IBS (ICD-9-CM code 564.1) occurrence. Every patient was observed until a diagnosis of IBS (ICD-9-CM code 564.1) was made or until the patient withdrew from the study, insurance was terminated or the follow-up period ended (31 December 2010). Some comorbidities that were identified before the index date were also included, such as diabetes (ICD-9-CM code 250), hypertension (ICD-(ICD-9-CM codes 401– 405), hyperlipidemia (ICD-9-CM code 272), fibromyalgia (ICD-9-CM code 7291), depression (ICD-9-CM

codes 296.2–296.3, 300.4 and 311), anxiety (ICD-9-CM code 3000), temporomandibular joint disorder CM code 5246) and dysmenorrhea (ICD-9-CM code 6253).

Statistical analysis

The chi-squared test was used to analyze categorical

variables, such as sex and comorbidities, and Student’s t test was used to analyze continuous variables, such

as age, for comparing the migraine cohort with the comparison cohort at the baseline demographic status. The cumulative incidence of IBS for each cohort was calculated by dividing the total number of IBS events by the total sum of the follow-up year (per 10 000 person-years). The Cox proportional hazards regression

model was used to analyze the overall risk, age-specific risk, sex-specific risk and each comorbidity-specific risk of developing IBS associated with migraine. After adjusting for sex, age and comorbidities, the adjusted hazard ratio (HR) of IBS presented a 95% confidence interval (CI) for the migraine cohort.

based on the average number of migrainerelated hospital visits (outpatient visits or hospital

admissions) per year: seldom (<2 times per year), often (2–4 times per year) and usually (>4 per year). The Cox proportional hazards regression model was used to estimate the IBS risk with increasing number of migraine visits.

SAS 9.3 statistical package (SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses, and R software (R Foundation for Statistical Computing, Vienna, Austria) was used to calculate the

Kaplan–Meier estimate for measuring the cumulative IBS incidence of both groups. The differences between the cumulative incidence curves of the two groups were determined using the log-rank test. P < 0.05 in two-tailed tests was considered statistically significant. Results

The demographics and comorbidities of the migraine cohort (N = 14 117) and comparison cohort

(N = 56 468) are shown in Table 1. Both groups exhibited similarities in average age (42.5 years) and sex distribution (predominantly female, 72.6%). The migraine

cohort exhibited a higher prevalence of comorbidities than did the comparison cohort at the baseline (P < 0.05).

Table 2 shows that the overall cumulative incidence of IBS was 2.45-fold higher in the migraine cohort than in the comparison cohort (73.87 vs. 30.14 per 10 000 person-years). Figure 1 shows the cumulative incidence of IBS between the two cohorts. Migraine patients exhibited a greater incidence of IBS than did migraine-free individuals in the follow-up years (logrank test: P < 0.0001). After adjusting for age, sex

and comorbidities, migraine patients exhibited a 1.95-fold increased risk of IBS than did individuals without migraine (HR 1.95, 95% CI 1.75–2.18).

Table 2 also shows the demographic-specific and comorbidity-specific incidence of IBS, as well as the

risk of IBS for each age group, sex and comorbidities stratified. In the youngest group (<30 years old), migraine sufferers exhibited a dramatically increased risk of IBS compared with individuals without

migraine (HR 3.36, 95% CI 2.44–4.63). Male migraineurs exhibited a 2.11-fold increased risk of IBS compared with males without migraine (HR 2.11, 95%

CI 1.72–2.58). Females with and without migraine exhibited a similar IBS risk (HR 1.89, 95% CI 1.66– 2.16). Migraine patients without any comorbidities exhibited a significantly increased risk of IBS compared with migraine-free individuals without any comorbidities (HR 2.65, 95% CI 2.22–3.17). Similar

results were observed for migraine patients without

hypertension (HR 2.26), diabetes (HR 1.96), hyperlipidemia (HR 2.15), fibromyalgia (HR 2.07), depression

(HR 2.02), anxiety (HR 2.01), temporomandibular

joint disorder (HR 1.97) or dysmenorrhea (HR 1.94). Table 3 shows the incidence of IBS and the hazard

ratio adjusted using the frequencies of migraine diagnosis. The incidence of IBS and the adjusted HR

increased as the number of migraine diagnoses increased. In patients experiencing frequent migraine attacks, the incidence of IBS was 348.68 per 10 000 person-years, being 7.61-fold (95% CI 6.33–9.13) higher in the migraine cohort than in the comparison cohort.

Discussion

In accordance with the limited previous studies, our nationwide population-based cohort study revealed

that migraine was associated with a 1.95-fold increased risk of IBS after adjusting for age, gender

and medical comorbidities. Based on our research, this is the largest and only Asian population-based study to demonstrate an increased risk of IBS in migraine patients.

The major strength of the present study is that a large, national data set containing a representative

cohort of 1 000 000 citizens covered by the National Health Insurance of Taiwan was analyzed. The large sample size and the long (10 years) observation period offered significant power for statistical analyses. Moreover, this is the first study to examine the association

between the two diseases by searching for subsequent incidence of IBS in a migraine cohort. In addition, a

trend showing higher incidence rates of IBS in migraineurs with more frequent migraine-related healthcareseeking behaviors was found. After adjusting for age,

hypertension, diabetes, hyperlipidemia, fibromyalgia, depression, anxiety, temporomandibular joint disorder and dysmenorrheal disorders, the hazard ratios for IBS in migraineurs remained strong. Nevertheless, the increased risk of IBS in the migraine cohort seemed to be highest in the youngest group (i.e. under 30 years old), which had an adjusted HR of 3.36.

Altered cortical excitability and sensory processing in migraine patients

There is a growing acceptance of the notion that a genetically predisposed hyperexcitable nervous system may underlie the pathophysiology of migraine and IBS [8,10,19–21]. Recent studies have begun not only to explore the increased cerebral activation during migraine attacks but also the altered interictal cortical processing of various sensory inputs. For example, an increasing number of functional magnetic resonance imaging (fMRI) studies have revealed increased activations in the hypothalamus [22], thalamus [23],

brainstem [24,25] and occipital lobe [26] during migraine attacks, indicating the presence of increased ictal cortical excitability. During the interictal period, it has been consistently demonstrated in electrophysiological studies that the normal habituation of cortical

responses on repetitive stimulation is impaired [27]. Interictal positron emission tomography (PET) and fMRI

studies using visual [28] and pain stimuli [29] have also supported the lack of habituation and cortical hyperexcitability

in migraine, suggesting alteration in the pain

modulatory circuits. Recent evidence has shown that recurrent attacks may lead to functional and structural changes in the brain, whereby episodic migraine may transform to its chronic form. Morphometric changes of cerebral gray and white matter in migraine have also been demonstrated by fMRI studies using voxel-based morphometry (VBM) and diffusion tensor imaging [30,31]. Consequently, although there is a continued debate about whether these changes are a cause or consequence of migraine, it is thought that they may

explain the abnormal central pain processing and sensory hypervigilance in migraine.

The brain_gut axis

A growing body of evidence has indicated that central and visceral hypersensitivity may play a crucial role in the pathophysiology of IBS [32]. This hypersensitivity has been primarily observed in the ENS, where low

thresholds to rectal mechanical distension [33,34], thermal stimuli [35] and electrical stimulation [36] were

determined, suggesting altered nociceptive processing and reduced pain inhibition. As with migraine, fMRI studies in IBS patients have demonstrated morphological changes in cerebral gray [37,38] and white matter [39], which have been reported to be key regions of the pain matrix, raising the question of whether these changes are a cause or consequence of IBS. An emerging model of a dysregulated brain_gut

axis is increasingly accepted as the pathophysiology of IBS, in which the nervous system can be seen as an integrated system consisting of the CNS, the peripheral nervous system (PNS), the ENS and the autonomic nervous system (ANS). The CNS continually receives and interprets information from the other three systems [19]. An environmental threat, for instance,

alarms the entire system through the ANS and produces a state of hypersensitivity for defense. It was

and immunological factors may increase sensitization in the ENS and brain_gut axis [19]. Serotonin (5-HT), a crucial CNS neurotransmitter, is abundant in the gastrointestinal (GI) system and acts as a mediator between the CNS and ENS. Studies showed that smooth muscle relaxation, contraction and visceral

sensations are governed by different 5-HT subtypes in the GI tract [20]. Both 5-HT agonists and antagonists

have been proved to be effective in reducing the abnormal motility and visceral hypersensitivity in IBS

patients [40]. Serotonin also plays a crucial role in migraine; for instance, sumatriptan (a 5-HT1D _agonist)

is one of the major medications for treating migraine.

Limitations of the current study

The current study was subject to several limitations that are worth mentioning. First, there was a concern regarding the validity of the migraine and IBD diagnoses. Nevertheless, the wide use of the ICHD-2 and

Rome criteria for coding migraine and IBD diagnoses in Taiwan, as well as the routine sample check of medical records by the NHI Bureau with regard to health

insurance claims might have improved the accuracy of the coding. Secondly, although a universal healthcare system exists in Taiwan, there is a potential bias that migraineurs with more frequent medical-care-seeking behavior may have a higher chance of being diagnosed with IBS compared with other migraineurs. Thirdly, the NHIRD does not provide detailed information on patients regarding their lifestyle, habits, body mass index, physical activity, socioeconomic status and family history that may also contribute as potential confounding factors. Fourthly, the records of the NHI

claims primarily serve the purpose of administrative billing and do not undergo verification for scientific purposes. The patients could not be contacted directly to obtain additional information on their use of medications because of the anonymity assured by the identification numbers. Finally, patients with migraine or

IBS who have not used healthcare services are not included in the study population.

Conclusion

This nationwide population-based cohort study revealed that migraine patients _ particularly the young _ exhibited an increased risk of developing IBS. Considerable evidence is emerging to support a unified concept of these functional pain disorders, in which a genetically sensitive nervous system becomes hypervigilant to stimuli and develops structural and functional changes over time to give rise to ubiquitous diseases. Screening patients with migraine for IBS and vice versa is necessary to provide an effective therapy toward understanding these disorders.