Published ahead of print on September 25, 2003, doi:10.1164/rccm.200308-1092OC
© 2004 American Thoracic Society Induced Sputum Inflammatory Mediator Concentrations in Chronic CoughDepartment of Respiratory Medicine, Institute for Lung Health, Glenfield Hospital, Leicester, United Kingdom Correspondence and requests for reprints should be addressed to Surinder Birring, M.R.C.P., Department of Respiratory Medicine, Institute for Lung Health, Glenfield Hospital, Leicester LE3 9QP, UK. E-mail: sb134{at}le.ac.uk
Previous studies have shown evidence of airway inflammation in patients with chronic cough and have suggested that the cough may be due to release of tussive mediators and activation of afferent sensory nerve endings. We measured the concentration of various proinflammatory and tussive mediators in induced sputum supernatants from 20 patients with cough variant asthma or eosinophilic bronchitis, 20 patients with nonasthmatic chronic cough, 22 patients with idiopathic chronic cough, and 18 healthy control subjects. We measured histamine, cysteinyl-leukotrienes, prostanoids (prostaglandin D2 and prostaglandin E2), and interleukin-8 by enzyme immunoassay. The median sputum histamine concentrations were significantly higher in patients with idiopathic chronic cough (8.0 ng/ml) and cough variant asthma/eosinophilic bronchitis (10.2 ng/ml) than in normal subjects (2.6 ng/ml; 95% confidence interval of difference from idiopathic chronic cough, 0.8 to 25.8 [p = 0.009] and 95% confidence interval of difference from cough variant asthma/eosinophilic bronchitis, 1.1 to 20.1 [p = 0.01]). Median sputum prostaglandin D2 and prostaglandin E2 concentrations were significantly higher in all categories of chronic cough. Our findings support the view that there is release of inflammatory and tussive mediators in patients with chronic cough and suggest that there might be similarities in the mechanism of cough in a diverse range of conditions.
Key Words: chronic cough • histamine • interleukin-8 • prostaglandins • sputum Cough is a common presenting symptom in both general practice and respiratory clinics. Most cases are mild and self-limiting, but some patients have persistent troublesome cough and require specialist review. Persistent cough is thought to be due to one or more of the following: asthma, eosinophilic bronchitis, gastroesophageal reflux, or rhinitis, with postnasal drip in the majority of cases (1, 2). However, in up to 20% of patients, the cough remains unexplained even after extensive investigation and treatment trials (3). Several studies employing different techniques to assess airway inflammation have shown that airway inflammation is present in patients with chronic cough. Various abnormalities have been detected depending on the category of patient and method used to assess airway inflammation, including sputum neutrophilia and an increased concentration of interleukin-8 in induced sputum (4); bronchoalveolar lavage and bronchial biopsy lymphocytosis in idiopathic chronic cough (5–7); and sputum, bronchoalveolar lavage, and bronchial biopsy eosinophilia in asthma and eosinophilic bronchitis (2, 8, 9). The mechanism of cough associated with these different patterns of airway inflammation is unclear, but one possibility is release of protussive mediators and activation of sensory nerve endings (10). Mast cell–derived mediators may be particularly important because there is evidence of increased mast cell numbers in the superficial airway of patients with idiopathic chronic cough and eosinophilic bronchitis (6, 11). In this study we tested the hypothesis that chronic cough syndromes are associated with an increased concentration of mast cell–derived tussive mediators by measuring induced sputum histamine and prostaglandin (PG) D2 concentrations in patients with cough variant asthma or eosinophilic bronchitis, nonasthmatic explained chronic cough, idiopathic chronic cough, and in healthy subjects. We also measured induced sputum concentrations of other potentially relevant inflammatory mediators: PGE2, cysteinyl-leukotrienes (LTC4, LTD4, and LTE4), and interleukin-8 (IL-8). Some of the results of these studies have been previously reported in the form of an abstract (12).
Subjects The study population consisted of 22 patients with idiopathic chronic cough, 20 with cough variant asthma or eosinophilic bronchitis, 20 with nonasthmatic explained chronic cough, and 18 normal control subjects. Patients were recruited randomly as a representative sample of the population attending cough clinics in 2000 and 2001. The referrals received by the clinic at Glenfield Hospital (Leicester, UK) from primary and secondary care are largely confined to a population of 970,000 within Leicestershire. Normal control subjects were recruited from healthy volunteers responding to local advertising. Investigations were performed according to a standardized algorithm (2). We defined idiopathic chronic cough as a cough lasting more than 3 weeks in association with normal clinical examination (including ear, nose, and throat), normal chest X-ray and computed tomography scan, normal lung function tests, negative methacholine inhalation test (provocative concentration causing a 20% fall in FEV1 [PC20FEV1] > 8 mg/ml), normal peak expiratory flow (PEF) variability, normal sputum eosinophil count, and no pathological gastroesophageal reflux or evidence of temporal association between cough and gastroesophageal reflux on 24-hour esophageal pH monitoring. Patients with idiopathic chronic cough had extensive negative treatment trials including trials of inhaled and systemic corticosteroids and treatment for gastroesophageal reflux according to current guidelines (13). Subjects with eosinophilic bronchitis had an isolated chronic cough, no symptoms suggesting variable airflow obstruction, normal spirometric values, normal PEF variability, a methacholine PC20FEV1 > 16 mg/ml, normal chest radiograph, and a sputum eosinophilia (greater than 3% nonsquamous cell). Subjects with cough variant asthma had an isolated cough and objective evidence of variable airflow obstruction, as indicated by one or more of the following: (1) methacholine airway hyperresponsiveness (PC20FEV1 < 8 mg/ml); (2) greater than 15% improvement in FEV1 10 minutes after administration of 200 µg of albuterol; or (3) PEF variability (greater than 20% maximum within-day amplitude from twice daily PEF measurements over 14 days). Subjects with nonasthmatic chronic cough were those in whom there was an identifiable cause for their cough (other than asthma or eosinophilic bronchitis) and improvement in cough after specific treatment of this. The protocol for investigation and treatment, and criteria for accepting diagnosis, were as previously described (2, 13). Normal subjects were asymptomatic and had normal spirometry and methacholine PC20FEV1 > 16 mg/ml. No subjects had received corticosteroids or other specific treatment for the condition causing cough for at least 6 weeks before the study. All subjects gave full informed consent to participate. The current study overlapped with another study (14) and 5 normal subjects, 12 patients with idiopathic chronic cough, and 4 patients with explained chronic cough who met the criteria for both studies participated in these studies after giving separate informed consent. The protocol for this study was approved by the Leicestershire Research Ethics Committee.
Protocol and Clinical Measurement
Mediator Measurements
Statistical Analysis
The subject characteristics are as shown in Table 1 . All subjects were nonsmokers. Patients with cough were recruited from a population (n = 236) in which the primary causes of cough were as follows: cough variant asthma, 39 (17%); gastroesophageal reflux, 35 (15%); rhinitis, 29 (12%); eosinophilic bronchitis, 17 (7%); idiopathic, 54 (23%); postviral, 17 (7%); bronchiectasis, 14 (6%); chronic bronchitis, 10 (4%); enlarged tonsils, 8 (3%); pulmonary fibrosis, 7 (3%); angiotensin-converting enzyme inhibitor cough, 4 (2%); sarcoidosis, 1 (0.5%); and bronchial tumor, 1 (0.5%). The causes of cough in the nonasthmatic chronic cough group were as follows: gastroesophageal reflux (4), rhinitis (5), postviral (5), chronic bronchitis (3), angiotensin-converting enzyme inhibitor cough (1), pulmonary fibrosis (1), and bronchiectasis (1). Patients with idiopathic chronic cough had a significantly higher prevalence of organ-specific autoimmune disease than normal subjects and patients with explained cough (Table 1). The sputum eosinophil counts and exhaled nitric oxide concentrations were significantly higher in patients with cough variant asthma/eosinophilic bronchitis than in patients with idiopathic chronic cough, nonasthmatic chronic cough, and normal subjects (Table 2) . There were no other differences in sputum leukocyte differential cell counts between groups (Table 2).
The sputum supernatant mediator concentrations were as shown in Table 3 and Figure 1 . There were significant differences in sputum histamine concentrations between the four groups (p = 0.017; Table 3 and Figure 1). The median sputum histamine concentrations were significantly higher in patients with idiopathic chronic cough (8.0 ng/ml) than in normal subjects (2.6 ng/ml; 95% confidence interval of difference, 0.8–25.8; p = 0.009) and approached significance when compared with nonasthmatic chronic cough (3.3 ng/ml; 95% confidence interval of difference, -0.5 to 19.6; p = 0.07). Median sputum histamine concentration was also significantly higher in patients with cough variant asthma/eosinophilic bronchitis (10.2 ng/ml) compared with normal subjects (95% confidence interval of difference, 1.1–20.5; p = 0.01). There were significant differences in PGD2, PGE2, and cysteinyl-leukotrienes, but not IL-8, between the four groups (p = 0.008, p < 0.001, and p = 0.01, respectively; Table 3 and Figure 1). Median sputum PGD2 and PGE2 concentrations were significantly higher in all cough groups compared with normal subjects (Table 3 and Figure 1). Median sputum cysteinyl-leukotriene concentration was significantly higher in patients with cough variant asthma/eosinophilic bronchitis than in normal subjects (Table 3 and Figure 1). Within the nonasthmatic chronic cough group, there were no important differences in sputum mediator profile between different diagnostic groups (Figure 1).
This is the first study to investigate the concentrations of protussive mediators in induced sputum from patients with chronic cough. We found increased histamine concentration in patients with idiopathic chronic cough and cough variant asthma/eosinophilic bronchitis compared with normal subjects. Sputum PGD2 and PGE2 concentrations were raised in all subjects with chronic cough. Our findings are consistent with the hypothesis that chronic cough is associated with abnormal release of inflammatory and potentially protussive mediators within the airway. We chose to estimate airway mediator production using induced sputum supernatant because the procedure is noninvasive, simple (18), safe (19), and responsive (20), and previous studies have shown that sputum mediator concentrations are significantly higher than in bronchoalveolar lavage (21, 22) and can be measured repeatably (18, 23). Our primary focus was to investigate the inflammatory mediator release in explained chronic cough and idiopathic chronic cough. Because cough variant asthma and eosinophilic bronchitis have characteristic and largely similar inflammatory cell profiles (15) we further subdivided the explained chronic cough group into cough variant asthma or eosinophilic bronchitis and nonasthmatic explained chronic cough. We have measured a wide spectrum of mediators with the potential to activate the cough reflex by a variety of mechanisms, with a bias toward mast cell–derived mediators, because of evidence particularly implicating this cell type in the genesis of chronic cough. We chose to investigate mast cell activation by sputum supernatant histamine and PGD2 assays because, unlike alternative measures such as sputum mast cell counts and markers of mast cell degranulation, these are well-validated techniques (18). Furthermore, we reasoned that mediator concentrations would better reflect the presence and activation of mast cells within superficial airway structures than mast cell numbers in sputum. Histamine and PGD2 produced by mast cells are smooth muscle contractile agonists (22) and, along with PGE2, may increase cough reflex sensitivity by a direct effect on cough sensory receptors (10). Cysteinyl-leukotrienes produced by eosinophils and mast cells cause bronchoconstriction and increase mucus production and vascular permeability (24). We also measured interleukin-8, a cytokine associated with neutrophilic inflammation, because an earlier study showed that sputum IL-8 concentration is increased in patients with nonasthmatic chronic cough (4). We found increased sputum histamine concentration in idiopathic chronic cough, to a degree similar to that seen in patients with cough variant asthma and eosinophilic bronchitis. The elevation in histamine concentration in combination with PGD2 is highly suggestive of mast cell activation because basophils, which also produce histamine, do not produce PGD2 (25). Thus mast cell activation with release of mediators appears to be a feature of both idiopathic chronic cough and cough due to asthma/eosinophilic bronchitis. In support of this view, increased mast cell numbers have been reported in bronchoalveolar lavage fluid from patients with idiopathic chronic cough (6), and in bronchial brushings from patients with eosinophilic bronchitis (11) compared with normal subjects. An important role for histamine and other mast cell-derived mediators in the pathogenesis of cough is also supported by studies showing that high-dose antihistamines improve distilled water-induced cough in patients with idiopathic chronic cough (26) and cause a marked decrease in cough due to seasonal atopic asthma (27). Our findings of raised sputum histamine and PGD2 concentrations in patients with cough variant asthma contrast with previous findings in non–cough-predominant asthma (15). A plausible explanation for this discrepancy that is worth further investigation is that there are differences in the site of release of histamine, with localization of mast cells within the airway smooth muscle important in the genesis of bronchoconstriction and airway hyperresponsiveness (9), and localization to sensory nerve endings important in cough (15). Thus there may be parallels between mechanism of cough and the mechanism of itch, concerning which a role for mast cells localizing to sensory nerve endings in the skin is well established (28). Sputum PGE2 concentrations were elevated in all patients with cough. Although we cannot exclude the possibility that this is due to the mechanical effects of coughing per se because it was seen in all cough groups, it is possible that this mediator release is related to the enhanced cough reflex sensitivity seen in these subjects because inhaled PGE2 results in heightened cough reflex sensitivity to capsaicin (10). Our findings do not confirm previous reports of increased sputum neutrophil cell numbers or sputum IL-8 concentrations in idiopathic chronic cough (4). This difference seen in the earlier study may have been due to a young control population, because data suggest that sputum neutrophil cell count increases with age (29). We did find increased exhaled nitric oxide levels in patients with cough variant asthma/eosinophilic bronchitis, findings that are in keeping with Chatkin and coworkers (30), who showed that exhaled nitric oxide levels were significantly higher in patients with chronic cough due to asthma than in those with chronic cough due to causes other than asthma. Bronchoscopy studies have shown evidence of lymphocytic airway and perhaps alveolar inflammation in patients with idiopathic chronic cough (5, 14). The absence of an increase in lymphocyte count in induced sputum cannot necessarily be taken as evidence against these earlier findings because the proportion of lymphocytes in these samples is low, so they may not reflect lymphocytic airway inflammation well (31). The nonasthmatic chronic cough group contained patients with various conditions that may be associated with different patterns and degrees of airway inflammation. There were no differences in mediator profile, suggesting perhaps that there is a common final pathway in the genesis of chronic cough in these diverse conditions although our power to identify differences in mediators was low. Further studies are required to address this question in more detail. A significant proportion (23%) of the patients in the population with chronic cough referred to our clinic had idiopathic chronic cough. Our findings provide some insight into the mechanisms of cough and airway inflammation. In common with earlier reports and case control studies (14, 32), patients with idiopathic chronic cough were predominantly female and there was a high incidence of organ-specific autoimmune disease. The association with organ-specific autoimmune disease and bronchoalveolar lymphocytosis in patients with idiopathic chronic cough has led us to speculate that cough is due to homing of activated lymphocytes into the pulmonary compartment from primary sites of autoimmune inflammation (14). The current study supports the view that idiopathic chronic cough is associated with airway inflammation and organ-specific autoimmune disease. It provides new evidence that the final mechanism of cough might have similarities to cough in better defined conditions such as asthma and eosinophilic bronchitis, with involvement of mast cell–derived mediators. Further work is required to determine whether other protussive mediators are involved and to determine the mechanism of the associated cough.
The authors thank the subjects who participated in the study, and Mrs. Susan McKenna, Mrs. Beverley Hargadon, Natalie Neale, Dhiraj D. Vara, and the Department of Respiratory Physiology for assistance in the clinical characterization of some of the patients.
Supported by the British Lung Foundation and University Hospitals of Leicester NHS Trust. S.S.B. is a British Lung Foundation Clinical Research Fellow. Conflict of Interest Statement: S.S.B. has no declared conflict of interest; D.P. has no declared conflict of interest; C.E.B. has no declared conflict of interest; P.B. has no declared conflict of interest; A.J.W. has no declared conflict of interest; I.D.P. has no declared conflict of interest. Received in original form August 6, 2003; accepted in final form September 24, 2003
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
2 tests were used to make between-group comparisons on the prevalence of autoimmune disease. A value of p < 0.05 was taken as being statistically significant. 
and IL-8, in induced sputum of non-asthmatic patients with chronic dry cough. Thorax 1999;54:234–237.