Dry eye disease (DED) is a common and chronic ocular disorder. It is a multifactorial disease of the ocular surface and tear glands that results in long-term discomfort (Lemp et al., 2007; Lemp, 2008). DED is an immune- and inflammation-related disease (Stevenson, Chauhan & Dana, 2012; Wei & Asbell, 2014). DED is generally diagnosed according to its symptoms, without united criteria (Lemp et al., 2007). In several population-based studies, DED has been linked to numerous other medical issues including cardiovascular, neurological, rheumatological, endocrine, gastrointestinal, and mental diseases (Ahn et al., 2014; Galor et al., 2011; Uchino et al., 2011; Van Der Vaart et al., 2015; Vehof et al., 2014; Wang et al., 2012).
Asthma is also a common and chronic disease in adults and children, involving airway inflammation, obstruction, and hyperresponsiveness. Allergen-specific type 2 CD4+ T-helper cells and related cytokines mediate the inflammatory process, and several cells, including eosinophils, mast cells, macrophages, and epithelial cells, play essential roles in the pathogenesis of asthma (Locksley, 2010). Patients with asthma often have other allergic conditions including allergic rhinitis, allergic conjunctivitis, and atopic dermatitis (Ledford & Lockey, 2013; Neto et al., 2010). Several ophthalmologists have reported the association of DED with a high risk of allergy or asthma (Chia et al., 2003; Moss, Klein & Klein, 2008; Paulsen et al., 2014; Wang et al., 2012); however, the risk of DED in patients with asthma or allergy has not been investigated. The present study analyzed the possible factors associated with DED in asthma patients.
Materials and Methods
The data used in this study were obtained from the National Health Insurance (NHI) Research Database of Taiwan. The NHI program, initiated in Taiwan in 1995, currently provides medical care to more than 98% of the total population of 23 million. Taiwan’s National Health Research Institute (NHRI) provides a representative database of 1 million randomly sampled patients from all NHI enrollees for research purposes. The identification information of every individual is encrypted before making the data publicly available for research purposes.
In this study, we retrospectively identified all patients with asthma from among the one million randomly sampled patients in the NHRI representative database between 1998 and 2010 by using the International Classification of Diseases, Ninth Revision, Clinical Modifications (ICD-9-CM) code 493. To ensure the validity of diagnoses, only those patients who received at least three asthma diagnoses as outpatients or one as inpatients were included in the analysis. Patients who were less than 18-years-old in 1997 were excluded from this study. The control group was randomly selected from the remainder of the NHRI representative database. Every patient with asthma had four sex-and age-matched controls according to birth year.
Dry eye disease was defined as three consecutive diagnoses of DED (ICD-9-CM code 375.15, tear film insufficiency, unspecified) and eye lubricant prescriptions. Patients with Sjogren syndrome (ICD-9-CM code 710.2) were excluded from the study.
Allergic comorbidities and asthma-related medications
Allergic comorbidities, including allergic rhinitis (ICD-9-CM code 477), atopic dermatitis (ICD-9-CM codes 691.0 and 691.8), and acute urticaria (ICD-9-CM code 708), were defined as at least three diagnoses during the observation period. The prescription of asthma-related medications (leukotriene receptor antagonists (LTRAs), oral antihistamines, and inhaled corticosteroids) were defined as an outpatient prescription for more than 180 days during the observation period. The odds ratios (ORs) for allergic comorbidities and asthma-related medications were adjusted for sex, age, income, and residence area.
All statistical analyses were performed using SAS (version 9.2; SAS Institute, Cary, NC, USA). We used the chi-squared test and t-test to compare the differences between asthma and the matched control groups. We assessed the risk factors for DED among patients with asthma by using ORs accompanying 95% confidence intervals (CIs); p < 0.05 were considered significant.
The design of this study was reviewed and approved by the Institutional Review Board of the Taichung Veteran General Hospital, Taiwan. (Approval number: IRB TCVGH No: CE16141B).
In total, data of 206,145 people were sampled in this study with 41,229 and 164,916 people comprising the asthma group and non-asthma groups, respectively. The distribution of sex and age for both groups is summarized in Table 1. During the observation period (1998–2010), 6.35% of patients received diagnosis of DED in the asthma group compared with 4.92% of patients in the non-asthma group (p < 0.0001). Patients with DED were younger in the asthma group than in the non-asthma group; the age distributions of patients with DED in the asthma and non-asthma groups were significantly different, as presented in Table 2.
|Total (n = 206,145)||Asthma (n = 41,229)||Non-asthma (n = 164,916)|
|Age in 1997 (years old)|
|Mean (range)||49.56 (18–91.64)||49.56 (18–91.43)||49.56 (18–91.64)||0.967|
|Total (n = 206,145)||Asthma (n = 41,229)||Non-Asthma (n = 164,916)|
|Age distribution of DED|
|18 ≤ age < 40||489||4.56||157||5.99||332||4.1||<0.0001|
|40 ≤ age < 65||3,877||36.15||1,052||40.17||2,825||34.85|
|Age ≥ 65||6,360||59.3||1,410||53.84||4,950||61.06|
|Age of DED (years old)|
We further analyzed the distribution and risk factors for DED in the asthma group (shown in Table 3). In the asthma group, female patients were more likely to develop DED than were male patients (OR = 1.70, 95% CI [1.57–1.85]). After adjustment for sex, age, income, urbanization, and allergic comorbidities, we found that patients with allergic rhinitis (adjusted OR = 1.58, 95% CI [1.46–1.72]) and urticaria (adjusted OR = 1.25, 95% CI [1.12–1.38]) were associated with DED, but not patients with atopic dermatitis (adjusted OR = 1.17, 95% CI [0.98–1.4]). Patients with asthma and DED had more prescriptions for LTRAs (adjusted OR = 1.29, 95% CI [1.01–1.64]), oral antihistamines (adjusted OR = 2.02, 95% CI [1.84–2.21]), and inhaled corticosteroids (adjusted OR = 1.19, 95% CI [1.04–1.36]) for more than 180 days.
|Drug used more than 180 days|
Table 4 presents the distribution and risk factors for DED in the non-asthma group. Female patients were more likely to develop DED than were male patients (OR = 1.77, 95% CI [1.69–1.86]). We found that allergic rhinitis, urticaria, and atopic dermatitis were associated with DED after adjustment for sex, age, income, urbanization, and allergic comorbidities.
In this study, we found that patients in the asthma group were more likely to have DED than were those in the control group. Several allergic comorbidities and medications were associated with DED among both asthma and non-asthma patients; these findings have rarely, if ever, been discussed in previous studies.
The prevalence of DED varies from 5.5% to 33.7% (Gayton, 2009; Han et al., 2011; Um et al., 2014) between different populations, primarily affecting women, and elderly people (Lin et al., 2003; McCarty et al., 1998; Moss, Klein & Klein, 2000). The mechanisms underlying the relationships between DED and several reportedly associated medical conditions remain unclear. The percentage of DED in our study is similar to that in other studies.
In the Blue Mountains Eye Study conducted by Chia et al. (2003), patients with more than one of the four DED symptoms were more likely (OR = 1.6) to have asthma (dry eye, grittiness, itchiness, and discomfort). In a nationwide population-based study, Wang et al. (2012) found that patients with DED are associated with asthma (6.18% vs. 5.01%, p < 0.001). In a 10-year cohort study, Moss, Klein & Klein (2008) found that elderly patients with an allergy history had an increased incidence of DED (25.2% vs. 20.4%, p = 0.006). In this study, we found a similar association between asthma and DED.
In the present study, we found an even higher association of DED in patients with asthma with other allergic comorbidities; this finding has not been reported before although the association of single allergic diseases with DED has been reported in a few studies. Dogru et al. (2016) found that children with allergic diseases had a shorter tear film break-up time than did healthy children, which may lead to DED in the future. A short tear film break-up time leads to the disruption of the protective barrier against possible airborne pollutants or allergens, which may further induce or aggravate allergic conjunctivitis. Yenigun et al. (2016) have described a significant association between DED and allergic rhinitis using a positive skin-prick test. Apart from these studies, the relationship between allergic symptoms and DED has rarely been discussed. Atopic dermatitis was once believed to be the first presentation of the classical “atopic march.” However, nowadays, abnormalities of the epidermal structure and function are considered crucial pathophysiological peculiarities of atopic dermatitis rather than only secondary effects of immunological mechanisms (Weidinger & Novak, 2016). This observation may explain our findings that atopic dermatitis is associated with DED in the asthma group, but the association is not as prominent after adjustment for the other two comorbidities.
In the Beaver Dam Offspring Study conducted by Paulsen et al. (2014), a history of allergies and the use of medications, including antihistamines and steroids, were found to be risk factors of DED. However, the authors did not further discuss the mechanisms underlying the association between DED and those medications. In this study, we specifically analyzed the medication associated with asthma, which has rarely been discussed in the literature. LTRAs and inhaled corticosteroids are common asthma controllers. We found that patients with asthma with prescriptions for these three types of medication for more than 180 days were associated with DED, after adjustment for sex, age, income, and urbanization. Due to the limitation of this study, determining whether these medications increased the risk of DED or whether patients with more severe asthma symptoms consumed more medication and had more frequent DED was difficult. Oral antihistamines are not regularly used in treatment of asthma, but they are widely prescribed for other allergic symptoms. The relationship between antihistamines and DED has been described in several studies (Galor et al., 2011; Moss, Klein & Klein, 2008; Paulsen et al., 2014), possibly because of the mucosa-drying effect of the antihistamines. In this study, we found a similar association between antihistamines and DED.
Although the exact mechanisms underlying both DED and asthma are not completely clear, we believe that some correlations may exist between the pathogenesis of the two diseases. Inflammation is now hypothesized to be the core mechanism underlying DED. Inflammatory cells, cytokines, and chemokines were detected on the ocular surface and in tears of patients with DED (Wei & Asbell, 2014). By contrast, asthma is triggered by exposure to allergens, infection, or irritants. Innate and adaptive cells, including eosinophils, mast cells, lymphocytes, neutrophils, and monocytes, are activated to produce cytokines and chemokines and enhance inflammation and airway hyperresponsiveness.
This study has limitations. This is a retrospective study using the database provided by NHRI; therefore, the patients’ personal history, environmental exposure, and severity of asthma or DED were not available in the database. We used the ICD-9-CM diagnoses codes and medication prescription to define DED, asthma, and allergic comorbidities. For validation, we defined DED as three consensus diagnoses of DED and eye lubricant prescription to confirm the diagnosis (Wang et al., 2012). We defined medication usage as prescription for more than 180 days during the study period; however, the amount of each type of medication was not calculated as a defined daily dose in our study. Several environmental factors, such as smoking and exposure to certain pollutants, can be important risk factors for both asthma and DED, but they cannot be analyzed in the study.
In conclusion, our study reveals that patients with asthma were more prone to DED than were the matched controls. In the asthma group, female patients, as well as patients with allergic comorbidities, including allergic rhinitis, atopic dermatitis, and urticaria, were more likely to have DED than were those without comorbidities. The use of oral antihistamines, LTRAs and inhaled corticosteroids was associated with DED. Asthma is a chronic disease that requires long-term follow-up; the results of our study suggest that in clinical practice, physicians should pay attention to DED particularly in patients with a high risk of DED.