|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From Pulmonary and Critical Care Medicine (Drs. Dixon, Kaminsky, and Irvin), University of Vermont, Burlington, VT; and Department of Medicine (Drs. Wise and Shade) and Bloomberg School of Public Health (Dr. Holbrook), Johns Hopkins University, Baltimore, MD.
Correspondence to: Anne E. Dixon, MD, FCCP, Pulmonary and Critical Care Medicine, Patrick 204, Fletcher Allen Health Care, 111 Colchester Ave, Burlington, VT 05401; e-mail: anne.dixon{at}vtmednet.org
Abstract
Background: Allergic rhinitis and sinusitis are frequently associated with asthma. The purpose of this study was to determine the impact of self-reported allergic rhinitis and sinusitis on lower airway disease in a large cohort of participants with well-characterized asthma.
Methods: A cohort study of participants in two trials of the American Lung Association-Asthma Clinical Research Centers: 2,031 asthmatics in the Safety of Inactivated Influenza Vaccine in Asthma in Adults and Children (SIIVA) trial and 488 asthmatics in the Effectiveness of Low Dose Theophylline as Add-on Treatment in Asthma (LODO) trial. At baseline, participants reported the presence of allergic rhinitis and sinusitis, and then lung function and asthma control were measured. During the trials, participants were monitored for asthma exacerbations.
Results: More than 70% of participants reported either allergic rhinitis or sinusitis. Sinusitis was more common in female patients (odds ratio, 1.46 [SIIVA]), those with gastroesophageal reflux disease (odds ratio, 2.21 [SIIVA]), and those of white race (odds ratio, 1.53 [SIIVA]). Similar associations were seen for allergic rhinitis. LODO participants with allergic rhinitis and sinusitis had increased asthma symptoms and a trend toward more sleep disturbance. Participants with allergic rhinitis had higher baseline lung function than those without allergic rhinitis measured by peak flow (91.2% vs 95.8% in the SIIVA trial). Participants with sinusitis had similar lung function to those without sinusitis. Participants with and without allergic rhinitis had similar exacerbation rates. In the LODO trial only, participants with sinusitis had increased asthma exacerbations (5.68 per patient per year vs 3.72 per patient per year).
Conclusion: Allergic rhinitis and sinusitis are associated with more severe asthmatic symptoms and, in patients with poorly controlled asthma, more exacerbations but are not associated with low lung function.
Key Words: allergic rhinitis asthma gastroesophageal reflux disease pulmonary function tests sinusitis
Allergic rhinitis is a risk factor for sinusitis,1 and asthmatics are known to be at increased risk for sinusitis.23 Nasal and sinus symptoms are clearly very common in asthma, and some studies34 have suggested that sinusitis is a risk factor for severe asthma, whereas others56 have not. Similarly, there are reports78 that allergic rhinitis is associated with more severe asthma, but another study9 has found the opposite.
The roles of allergic rhinitis and sinusitis in asthma are controversial, and one possible explanation is that they have differential effects based on asthma disease severity. The American Lung Association-Asthma Clinical Research Centers (ALA-ACRC) conducted two clinical trials: one trial that enrolled a general population sample of asthmatics, most with mild disease, and a second trial that enrolled participants with poorly controlled asthma. Both trials acquired information on self-report of allergic rhinitis and sinusitis, asthma symptoms, pulmonary function, and exacerbation frequency by asthma diary. These data sets provided a unique opportunity to examine the relationship of allergic rhinitis and sinusitis to asthma in two well-characterized cohorts of patients with different levels of asthma severity.
Materials and Methods
Participants
We evaluated baseline cross-sectional data obtained from two cohorts of patients enrolled in the ALA-ACRC Safety of Inactivated Influenza Vaccine in Asthma in Adults and Children (SIIVA) and Effectiveness of Low Dose Theophylline as Add-on Treatment in Asthma (LODO) trials. The primary results of the SIIVA trial have been published10; those of the LODO trial have been submitted for publication. In brief, the SIIVA trial was a multicenter, double-blind, placebo-controlled, crossover study to determine the safety of the influenza vaccine in asthmatic participants. The LODO trial was a randomized, double-blind, placebo-controlled trial of low-dose theophylline vs montelukast vs placebo as add-on therapy for poorly controlled asthma. Both trials were approved by the institutional review boards of all participating institutions, and informed consent was obtained from all participants.
The SIIVA trial enrolled 2,031 participants
3 years old with a history of physician-diagnosed asthma. Participants were not selected based on the severity of their asthma. Asthma severity was rated by lung function with peak flow and according to symptoms by the asthma symptom utility index (ASUI).11 In addition, 995 participants underwent spirometry (without bronchodilator). During the trial, exacerbations within 14 days of an injection (placebo or influenza vaccine) were measured. These were defined as any one of the following: (1) a decrease of at least 30% in the peak expiratory flow rate (PEFR); (2) an increase in the daily use of rescue medication (eg, four or more puffs of albuterol above baseline); (3) an increase in, or addition of, systemic corticosteroids for asthma; or (4) an unscheduled health-care visit for asthma.
The LODO cohort comprised 488 participants
15 years old with poorly controlled asthma (regardless of baseline treatment regimen) as measured by a score of
1.5 on the Juniper asthma control questionnaire. Participants who had smoked within the last 6 months or who had a > 20 pack-year smoking history were excluded. At baseline, asthma severity was rated according to symptoms by the ASUI and according to lung function by peak flow and spirometry (before and after bronchodilator). Quality of life12 and sleep disturbance1314 were assessed at baseline according to validated questionnaires. Severity of symptoms related to cough was assessed by the question pertaining to cough from the Asthma Quality of Life Questionnaire (AQLQ).12 During the trial, participants were randomized to one of three treatment groups (theophylline, montelukast, or placebo) and monitored over 6 months for exacerbations, defined as described above for the SIIVA trial.
Classification of Disease of the Nose and Sinuses
The diagnosis of disease of the nose and sinuses was determined by response to a structured questionnaire, as is common practice in other studies, such as the European Community Respiratory Health Surveys1516 and the National Health and Nutrition Examination Survey.17 The presence of sinusitis was defined as patient report of sinusitis. The presence of allergic rhinitis was defined as a patient report of allergic rhinitis or hay fever.
Statistical Analysis
Descriptive statistics were used to describe the study population, followed by bivariate and multivariate analyses using logistic regression to explore the association between demographic and health characteristics and the presence of allergic rhinitis and sinusitis. Variables significant at p < 0.05 by univariate analysis were considered for inclusion in the final model.
We compared symptom scores and exacerbation rates in LODO participants with and without allergic rhinitis and sinusitis using the Wilcoxon rank-sum test. We compared the proportion of SIIVA participants with an exacerbation after injection by Pearson
2 test. We investigated whether there was an interaction between baseline asthma severity and either allergic rhinitis or sinusitis on asthma symptoms and exacerbations by creating an interaction term using baseline percentage of predicted peak flow < 80% predicted, and presence of allergic rhinitis and sinusitis, respectively. We pooled data from the SIIVA and LODO trials and examined the effect of these interaction terms on the ASUI symptom score by regression techniques using baseline peak flow as a covariate.
We studied asthma severity by lung function measures in participants with and without both allergic rhinitis and sinusitis by univariate and multivariate regression. All analyses were performed using statistical software (STATA 8; StataCorp; College Station, TX).
Results
Baseline Demographics
The SIIVA trial enrolled 2,031 participants (age range, 3 to 83 years). The distribution of race and age is shown in Table 1
. The LODO trial enrolled 488 participants
15 years old; age and race distribution are shown in Table 1. The SIIVA cohort included significantly more children: 38% of the participants were < 20 years old, compared with 10% in the LODO trial.
|
|
|
15 years old. Overall, patients
15 years old with allergic rhinitis in the SIIVA trial still had similar symptom scores to those without allergic rhinitis (data not shown).
|
Interaction Between Allergic Rhinitis and Sinusitis, and Asthma-Symptom Severity
We found that the effect of both allergic rhinitis and sinusitis on symptoms depended on baseline asthma severity. Using pooled data from the LODO and SIIVA trials, the presence of allergic rhinitis was associated with a decrease in the ASUI of 0.02 (n = 2,270, p = 0.002). We then included an interaction term for baseline peak flow < 80% predicted and the presence of allergic rhinitis (using baseline peak flow as a covariate); the presence of allergic rhinitis in those with peak flow < 80% predicted was associated with a decrease in the ASUI of 0.04 points (p for interaction term = 0.02). This indicates a more significant effect of allergic rhinitis on symptoms for patients with a combination of lower baseline lung function and allergic rhinitis, ie, those with more severe asthma.
Similarly, using pooled data from the LODO and SIIVA trials, the presence of sinusitis was associated with a decrease in the ASUI of 0.03 (n = 2,269, p < 0.001). We then included an interaction term for baseline peak flow < 80% predicted and the presence of sinusitis (using baseline peak flow as a covariate); the presence of sinusitis in those with peak flow < 80% predicted was associated with a decrease in the ASUI of 0.06 points (p for interaction term = 0.002). This indicates a more significant effect of sinusitis on symptoms for patients with a combination of lower baseline lung function and sinusitis, ie, those with more severe asthma.
Severity of Lung Disease in Participants With Rhinosinusitis
Participants with allergic rhinitis had higher lung function measured by peak flow than those without allergic rhinitis (Table 4
). FEV1 was also higher in SIIVA subjects with allergic rhinitis, and a similar trend was observed in LODO subjects. Similar trends toward higher FEV1 and peak flow were seen in participants with sinusitis in both the LODO and SIIVA trials, but these were not statistically significant (Table 4).
|
15 years old (data not shown).
|
In our study, self-reported allergic rhinitis and sinusitis were very common in asthma. GERD, female gender, and white race were particularly associated with both allergic rhinitis and sinusitis. Both allergic rhinitis and sinusitis were associated with more severe symptoms in those with more severe asthma at baseline; in one cohort, sinusitis (but not allergic rhinitis) was associated with increased exacerbations. Participants with allergic rhinitis and sinusitis tended to have higher lung function than those without allergic rhinitis and sinusitis.
We analyzed data separately from the two cohorts because the trials recruited different types of asthmatics. Indeed, SIIVA participants had fewer severe symptoms and better lung function at baseline. We studied both allergic rhinitis and sinusitis, as we wanted to determine if these disease processes had similar demographic associations and a similar effect on asthma; our findings are consistent with the concept that allergic rhinitis and sinusitis are part of a disease continuum.
The main limitation of our analysis is that cross-sectional prevalence data reported by participants may not accurately reflect upper airway disease activity assessed by more objective measures such as sinus CT. Reports of the correlation between sinus symptoms and actual sinus CT abnormalities vary, with one study18 reporting that 66% of subjects with sinus symptoms had an abnormal sinus CT finding, and another study19 (specifically of asthmatics) reporting significant sinus CT abnormalities in 80% of participants with nasal symptoms. However self-report is commonly used in large studies in which more invasive measures to document these diseases would be prohibitively expensive (such as the National Health Interview Survey and the European Community Respiratory Health Study20), and the rates of allergic rhinitis and sinusitis that we report are similar to those previously described in the literature (reviewed by de Benedictis and Bush21).
Demographic factors associated with allergic rhinitis and sinusitis in asthma have not previously been reported in a large cohort of asthmatics. We found that female gender was associated with an increased risk of allergic rhinitis and sinusitis; this is consistent with data from the National Health Interview Survey.22 White race was also associated with both allergic rhinitis and sinusitis; similarly, Chen et al23 found that among asthmatics, white patients were more likely to have allergic rhinitis than African Americans. Sinusitis has previously been associated with active smoking17; however, given the small number of smokers in our asthmatic population, we had limited power to detect a small increased risk. We also found that GERD was associated with both allergic rhinitis and sinusitis, as has previously been reported.2425
Symptom scores were increased in LODO participants with both allergic rhinitis and sinusitis; in particular, symptoms related to cough tended to be higher. Similarly, Bousquet et al26 reported that participants with seasonal allergic rhinitis and asthma had increased cough, and in the general population allergic rhinitis and sinusitis are associated with chronic cough.2728 We also found a trend toward increased sleep disturbance in LODO participants with allergic rhinitis and sinusitis. This has also been reported previously in nonasthmatics.29 Nasal obstruction likely contributes to sleep disturbance through increased airflow resistance. However, we did not find more severe symptoms in SIIVA participants with allergic rhinitis. As SIIVA participants had milder asthma, we pooled from the LODO and SIIVA trials. Both allergic rhinitis and sinusitis had a more pronounced effect on symptoms in those with more severe asthma at baseline. This suggests that allergic rhinitis and sinusitis have a differential impact on patients depending on the baseline level of asthma severity: they may be a minor irritation to people with mild asthma, but they substantially add to the burden of disease in patients with more severe asthma.
We found similar exacerbations in SIIVA participants with and without allergic rhinitis and sinusitis, but in LODO participants we found that sinusitis (but not allergic rhinitis) was associated with increased exacerbations. Studies730 have typically found increased exacerbations in patients with allergic rhinitis. Some but not all studies831 show that treatment of allergic rhinitis decreases asthma exacerbations. Our data suggest that although allergic rhinitis and sinusitis are common in asthmatics with all levels of disease severity, they contribute to poor asthma control in terms of symptoms and exacerbations only in those with more severe asthma.
There was a trend toward better lung function in participants with allergic rhinitis and sinusitis; this has been reported in other studies.26 The observation that allergic rhinitis and sinusitis did not affect lung function is consistent with results from experimental models: sinusitis in rabbits does not alter baseline lung function but does increase airway responsiveness,32 and nasal provocation in humans increases bronchial hyperreactivity without affecting baseline lung function.33 The link between the upper and lower airway is thought to be related to a common mechanism inciting disease in both places, possibly a postnasal drip mechanism.32 But this dissociation between symptoms and exacerbations, and lung function suggests that changes in lung function may be mediated through a different pathophysiologic pathway than the process associated with inflammation of the nose and sinuses. It is noteworthy that although remodeling is apparent in the lower airway of asthmatics, it is not so readily apparent in the nasal mucosa.34 In other words, lower airway disease has distinct features that may lead to dissociation in some measures of disease severity between the upper and lower airways.
Our findings suggest that self-reported allergic rhinitis and sinusitis are present in the vast majority of asthmatics. They contribute to increased symptoms and sleep disturbance in patients with poorly controlled asthma, and may also contribute toward exacerbations. Additionally, it seems reasonable to postulate that treatment of allergic rhinitis and sinusitis in all asthmatics may improve symptoms attributed to asthma through mechanisms unrelated to improvement in pulmonary function.
Footnotes
Abbreviations: ALA-ACRC = American Lung Association-Asthma Clinical Research Centers; AQLQ = Asthma Quality of Life Questionnaire; ASUI = asthma symptom utility index; ESS = Epworth sleepiness scale; GERD = gastroesophageal reflux disease; LODO = Effectiveness of Low Dose Theophylline as Add-on Treatment in Asthma; PEFR = peak expiratory flow rate; PSQI = Pittsburgh sleep quality index; SIIVA = Safety of Inactivated Influenza Vaccine in Asthma in Adults and Children
None of the authors have any conflicts of interest in the content of this article.
Funding was provided by the American Lung Association and grant K23 RR019965.
Received for publication December 20, 2005. Accepted for publication February 15, 2006.
References
This article has been cited by other articles:
![]() |
G. Rolla, G. Guida, E. Heffler, I. Badiu, L. Bommarito, A. De Stefani, A. Usai, D. Cosseddu, F. Nebiolo, and C. Bucca Diagnostic Classification of Persistent Rhinitis and Its Relationship to Exhaled Nitric Oxide and Asthma: A Clinical Study of a Consecutive Series of Patients Chest, May 1, 2007; 131(5): 1345 - 1352. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |