Association of inflammatory bowel disease with asthma risk: A nationwide cohort study.

Association of inflammatory bowel

disease with asthma

risk: A nationwide cohort study

Yi-Hao Peng, M.Sc.,1,2 Wei-Chih Liao, M.D.,3 Ching-Hua Su, B.S.,4 Hsuan-Ju

Chen, M.Sc.,5

Te-Chun Hsia, M.D.,2,3 Chia-Chen Chu, M.Sc.,1,2 Chin-Jung Liu, M.Sc.,1,2 and

Chia-Hung Kao, M.D.6,7

I

tract and comprise two major disorders: Crohn’s

disease (CD) and ulcerative colitis (UC). IBDs are considered to result from an exaggerated inflammatory

immune response in the gastrointestinal tract and are associated with various extraintestinal manifestations, including those in the respiratory system.1–4 Although

previous epidemiologic studies indicated that the incidence of IBD is relatively higher in Western countries

than in the Asian–Pacific region, its incidence and prevalence are increasing in various geographic regions worldwide, including Taiwan,5,6 which indicates

that the growing burden of IBD is at the global level. Asthma is a chronic inflammatory disorder that presents as a reversible airway obstruction, and the concept

of its pathophysiology has focused on overlapping interactions of smooth muscle dysfunction, airway inflammation,

and airway remodeling.7,8 This disorder

affects _300 million people of all ages worldwide, and its prevalence has increased considerably in the past 20 years.9,10 Previous studies proposed that, in addition to

genetic and environmental risk factors, and occupational exposure, various comorbid conditions, such as

with asthma in adults,11–13 which implies that

the nature of asthma pathogenesis is multifactorial. Several studies addressed the association of IBD and asthma14–16; however, some of these studies used a

case-control research design with a limited sample size, which rendered it difficult to generalize the study results.14 Furthermore, most studies were conducted in

Western countries, where the incidences and prevalence differ considerably from those in Asian populations.

14–16 Thus far, evidence regarding increased asthma risk after an IBD diagnosis is

scant, and no

study has examined the longitudinal frequency of asthma development in patients with IBD in an Asian population. Therefore, we conducted this nationwide population-based study by using data from the Taiwan National Health Insurance Research Database (NHIRD) to test the hypothesis that patients with IBD are at an increased subsequent risk of asthma.

METHODS Data Source

The National Health Insurance (NHI) program in

Taiwan is a universal health insurance program, implemented in March 1995, in which nearly 99% of the

population of Taiwan is enrolled. This study used the Longitudinal Health Insurance Database 2000 (LHID2000), which is a subset of NHIRD, and contains all of the

reimbursement data of the NHI with the identification numbers of all enrollees encrypted, transformed, and maintained by Taiwan’s National Health Research Institutes. This data base contains information on 1 million

randomly selected enrollees who were beneficiaries from 1996 to 2011, and the data contain no

statistically significant difference in the sex distribution between the beneficiaries in the LHID2000 and those in the entire NHI database.17 The NHIRD contains comprehensive

outpatient and inpatient information, such

as demographic data; dates of visits; International Classification of Diseases, Ninth Revision, Clinical

Modification (ICD-9-CM) diagnostic codes; and complete prescription details documented from 1996 to

2011. This study was approved by the institutional review board of China Medical University (CMU-REC-101-012).

Study Population

Based on the LHID2000, we identified 8482 cases that involved patients with newly diagnosed IBD, including both CD (ICD-9-CM codes 555.0 –555.2, 555.9) and UC (ICD-9-CM code 556), who presented for medical care from 2000 to 2005, as eligible study patients. We excluded patients younger than 20 years of age (n _ 2675), those whose asthma was already diagnosed at baseline (n _ 546), and those with incomplete information (sex or age) (n _ 1) from the analysis. The date of

the first diagnosis of IBD was considered the index date. Finally, the remaining 5260 patients with newly diagnosed IBD were included in the IBD cohort. For the comparison cohort, four enrollees for every patient in the IBD cohort were randomly selected from beneficiaries with no history of asthma and were frequency

matched by age (per 5 years, e.g., 20–24, 25–29, 30–34, and so on), sex, and index year. Finally, 5260 and

21,040 patients were included as the IBD and non-IBD cohorts, respectively.

Covariates

The relevant demographic factors were sex and age. Age was categorized into four levels: 20–34, 35–49, 50–64, and _65 years. Patients with a history of rhinitis (ICD-9 code 477, 472.0), chronic sinusitis (ICD-9

code 473), atopic dermatitis (ICD-9-CM code 691), and chronic obstructive pulmonary disease (COPD) (ICD-9-CM code 496) before the index date were identified as having comorbidities.

Main Outcome

The primary outcome measure was asthma (ICD-9-CM code 493), which was determined by linking records based on outpatient and inpatient claims data

in the NHIRD. To ensure the accuracy of the asthma diagnosis, only those patients with asthma who had been treated by using inhaled corticosteroids, systemic corticosteroids, or inhaled short-acting _2 agonists (SABA) were selected. Follow-up person-years were from the index date until December 2011 or until onset of asthma, death, or withdrawal from the NHI program.

Statistical Analysis

The distribution of demographic data and comorbidities (age, sex, rhinitis, chronic sinusitis, atopic dermatitis, and COPD) were compared between the two

cohorts. Differences were examined by using the _2 test

for the categorical variables and the Student’s t-test for the continuous variables. Kaplan-Meier analysis was used to plot the cumulative incidence of asthma, and a log-rank test was used to test the differences between the two cohorts. Cox proportional hazard regression models were used to assess the risk of asthma and asthma-associated risk factors, and the models were adjusted for sex, age, rhinitis, chronic sinusitis, atopic dermatitis, and COPD. We also performed sex-, age-, and comorbidity-stratified analysis to investigate the association between IBD and asthma. Adjusted hazard ratios (HR) and their 95% confidence intervals (CI) were estimated. SAS version 9.3 (SAS Institute, Cary, NC) was used for the data analyses, two-sided tests were performed, and p _ 0.05 was considered statistically significant.

RESULTS

The IBD and non-IBD cohorts comprised 5260 patients with newly diagnosed IBD and 21,040 age- and

sex-matched patients without IBD, respectively, ascertained from the LHID2000 from 2000 to 2005. The age

and sex distributions were identical in both groups. Compared with patients without IBD, patients with IBD had a higher prevalence of rhinitis (13.44% versus 8.14%), chronic sinusitis (1.86% versus 1.10%), and

[SD]) follow-up periods was

8.61 _ 2.54 years) in the IBD cohort and 8.52 _ 2.58 years in the non-IBD cohort.

The Kaplan-Meier analysis showed that the risk of asthma increased during the follow-up period in both cohorts and that the cumulative incidences of asthma were significantly higher in the IBD cohort than that in the non-IBD cohort (Fig. 1). Patients with IBD had a higher incidence density rate of asthma than those without IBD (7.04 versus 4.50 per 1000 person-years), with an adjusted HR of 1.50 (95% CI, 1.32–1.71) after adjusting for age, sex, and comorbidities (Table 2). In addition, a multivariate Cox proportional hazard analysis revealed that asthma was independently associated with female sex (1.18, 1.05–1.33), in the age group of 50–64 years and 65 years and older (2.18, 1.81–2.62 and 3.84, 3.20–4.62, respectively) as well as having rhinitis (1.91, 1.62–2.25), and COPD (2.94, 2.21–3.91). Sex-specific analysis showed that the incidence density rates of asthma in women and men with IBD were 7.13 and 6.94 per 1000 person-years, respectively, which were higher than those in the non-IBD cohort (4.82 and 4.13 per 1000 person-years, respectively). In addition, compared with the non-IBD cohort, women exhibited a 1.42-fold (adjusted HR 1.42 [95% CI, 1.20– 1.70]) higher risk of developing asthma, whereas men exhibited a 1.60-fold (1.60, 95% CI, 1.32–1.94) higher risk. The age-stratified results showed that the adjusted HR of asthma was the highest in the youngest group (i.e., 20–34 years) in the IBD cohort than in the non-IBD cohort (1.83, 1.33–2.50). Compared with patients in the non-IBD and with no comorbidity, patients in the IBD with no comorbidity (including rhinitis, chronic sinusitis, atopic dermatitis, and COPD) exhibited a 1.56-fold increased risk of developing asthma (Table 3). Further analysis according to the two major forms of IBD showed that patients with both UC and CD exhibited an increased risk of developing asthma compared with

the non-IBD cohort (1.46, 1.03–2.07, and 1.50, 1.31–1.72, respectively) (Table 4).

DISCUSSION

In this large population-based cohort study, we observed that patients with IBD exhibited a 1.50-fold risk of asthma development compared with that of the general population in the subsequent 12 years, after adjusting for age, sex, and comorbidities. Although the prevalence of rhinitis, chronic sinusitis,

and atopic dermatitis were significantly higher in patients with IBD than in the general population, the risk of asthma remained significantly higher in patients with IBD after adjusting for these covariates. A

growing body of evidence has indicated that IBD is

not only associated with chronic gastrointestinal disorders, but it is also frequently associated with various

extraintestinal manifestations, including respiratory and autoimmune disorders, e.g., multiple

sclerosis.1– 4,18 Historically, the relationship between

IBD and pulmonary disease was first demonstrated nearly 40 years ago in a case series by Kraft et al.,19

who reported six patients who had developed unex- plained severe pulmonary disease after their IBD

diagnosis.

More recently, Ceyhan et al.14 reported a higher

prevalence of respiratory symptoms, allergic symptoms, and abnormal lung function test results in 30

patients with IBD (compared with the controls). Haapamaki

et al.15 showed that the prevalence of asthma

was significantly higher in the 2381 patients with IBD than in a control group in Finland. In a population-based cohort study that involved 8072 patients with IBD and 41,815 controls in Canada, Bernstein et

al.16reported a higher prevalence of asthma in both

patients with CD and those with UC. Our data were consistent with the aforementioned studies, and, to our knowledge, this study was the first populationbased study that involved an Asian population,

which indicated that IBD increases the future risk of asthma in such populations.

Although the pathophysiology that underlies the association between IBD and asthma remains unclear,

there are several possible explanations. First, embryonically, both the intestine and the lungs originate from

the primitive foregut, which implies that they share some common structural features: a luminal structure that contains goblet cells, submucosal glands, and more importantly, both bronchus-associated lymphoid tissue and gut-associated lymphoid tissue play crucial roles in host mucosal defense. Therefore, previous

studies proposed that the similarities of the inflammatory and immune components of these organs explain

the overlap in pathophysiology in intestinal mucosal and lung disease.20,21 Second, tumor-necrosis factor _,

which is predominantly expressed by mast cells in the ileum wall of patients with IBD, is responsible for the tissue change and symptoms associated with IBD,22–24

is a target molecule in the immune-modulatory therapy for IBD,25 and is proposed to play a role in the

pathogenesis of asthma. When tumor necrosis factor _ is administered via inhalation in humans, bronchial hyperresponsiveness and sputum inflammatory cell influx can be induced in patients with mild asthma.26

Finally, the vascular endothelial growth factor (VEGF) pathogenesis of IBD by promoting intestinal angiogenesis and inflammation.

27,28 VEGF has also been shown to be increased

by human pulmonary cells stimulated with Dermatophagoides species extract and Dermatophagoides

species, and contribute to the pathogenesis of

asthma.29 Moreover, according to Kanazawa and Yoshikawa30

the VEGF levels in induced sputum and

airway vascular permeability index of 12 patients with UC and with asthma were significantly higher than in 15 patients with UC and without asthma,

which indicates that VEGF might mediate the development of asthma in patients with UC.

The stratified analysis conducted in this study

showed that the incidence of asthma progressively increased with age in both cohorts. However, the risk of

asthma progressively decreased with age. We speculated that this occurred because the comorbidities of

IBD increased with age; therefore, the adjusted HR appeared higher in younger participants than in older participants.

The prevalence of rhinitis, chronic sinusitis, and atopic dermatitis was significantly higher in the IBD

cohort than in the control cohort. The relationship between chronic sinonasal disease and IBD has been

reported by Book et al.31 in a cross-sectional study in

which 47.5% of the patients with IBD reported a history of chronic rhinosinusitis, chronic rhinitis, or chronic sinonasal symptoms. Although participants with no IBD were not included as controls in that study, the prevalence of chronic rhinosinusitis was estimated to be 14% in the general population of the United States,32

and the prevalence of allergic rhinitis was between 7.8% and 34.3% in the Chinese population,33 all of

which seems lower than those observed in the previous study.30 In addition, Myrelid et al.34 reported that

atopic manifestation (asthma, allergic rhinitis, and ec-zema) were significantly more frequent in patients

with CD. However, Boneberger et al.35 reported that

the prevalence of atopic dermatitis and allergic rhinitis was not significantly higher in patients with UC than in the control group. To date, evidence on whether

rhinitis, chronic sinusitis, or atopic dermatitis is associated with IBD is inconclusive, and future study is

warranted to clarify this issue.

The strength of this study is its nationwide population-based cohort design, in which the sample was

highly representative of the general population. However, several limitations should be considered when

interpreting our findings. First, all data retrieved from the NHIRD were de-identified secondary data; therefore,

detailed patient information, such as smoking

habits, body mass index, occupational exposure, and a family history of systemic diseases, was not available. Because these may be risk factors or comorbidities of asthma, this could have biased the study results. Second, the data regarding each participant’s clinical information, such as imaging results, pathologic findings,

serum laboratory data, and lung function tests,

were unavailable. Third, evidence from a retrospective cohort study is generally weaker than that from randomized trials because of potential bias related to the

adjustments for confounders. Despite our meticulous study design, biases that result from unknown confounding variables may exist to under- or overestimate

the relationship between IBD and asthma risk. Nevertheless, the NHIRD was derived from the NHI program,

which is operated by the single payer, the government of Taiwan. All insurance claims must be

reviewed and audited by medical specialists for the purpose of reimbursement. Therefore, the diagnoses of IBD and asthma should be reliable, and our findings regarding the relationship between IBD and asthma should also be reliable because of the validity of the database, long follow-up period, and large sample. In conclusion, our study showed that Asian patients with IBD exhibited a higher subsequent risk of asthma than that of the general population. This increased risk of asthma is significantly higher in both sexes, in all age groups older than 20 years, and in both patients with CD and those with UC compared with the general population. We suggest that clinicians be aware that IBD is a potential risk factor of asthma. Further study that explores the underlying pathophysiology between IBD and asthma is warranted.