Chronic bronchitis in COPD patients is associated with increased risk of exacerbations: a cross-sectional multicentre study

ORIGINAL PAPER Chronic bronchitis in COPD patients is associated with increased risk of exacerbations: a cross-sectional multicentre study J. L. Corhay, 1 W. Vincken, 2 M. Schlesser, 3 P. Bossuyt, 4 J. Imschoot 4 Linked Comment: Burgel. Int J Clin Pract 2013; 67: 1218 9. 1 Respiratory Department, Centre HospitalierUniversitaire du Sart-Tilman, Liege, Belgium 2 Respiratory Department, Universitair Ziekenhuis Brussel, Brussels, Belgium 3 Respiratory Department, Centre Hospitalier du Luxembourg, Luxembourg, Luxembourg 4 Medical Department, Takeda Belgium, Brussels, Belgium Correspondence to: Jean L. Corhay, Department of Pneumology, CHU Sart-Tilman B-35, Liege 4000, Belgium Tel.: 0032 43668888 Fax: 0032 43668846 Email: jlcorhay@chu.ulg.ac.be Disclosures Jean L. Corhay has received consultancy honoraria from AstraZeneca, Boehringer Ingelheim/Pfizer, GlaxoSmithKline, Novartis and Takeda. Walter Vincken received consultancy honoraria from AstraZeneca, Boehringer Ingelheim/Pfizer, GlaxoSmithKline, Novartis, Mundipharma, MEDA and Takeda. Pascale Bossuyt and Johan Imschoot are employees of Takeda Belgium. SUMMARY Background and aims: Chronic bronchitis (CB) in chronic obstructive pulmonary disease (COPD) patients is associated with increased mortality, frequent exacerbations and faster disease progression. This study investigates the prevalence of CB in a large population of COPD patients to identify features associated with CB. Methods: Cross-sectional multicentre study in patients with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages 2 4 from Belgium and Luxembourg. Results: The 974 patients included were on average 67.8 9.6 years old; 72% were male, FEV 1 was 52.5 15.8% of predicted. The prevalence of CB was 64% (622/974). In patients with CB, the number of pack-years smoked and the prevalence of chronic respiratory failure, cachexia and skeletal muscle wasting were significantly higher, whereas FEV 1 and FEV 1 /VC were lower. The prevalence of CB increased with GOLD stage and was higher in patients with emphysema and those exposed to occupational risk factors. The CB group had more exacerbations, a higher percentage of patients with frequent exacerbations (37.3% vs. 14.2% of patients; p < 0.0001), increased COPD-related, non-intensive care unit hospitalisations and all-cause hospitalisation rates. In multiple logistic regression analysis, frequent exacerbation was the most important independent variable associated with CB, followed by current smoking, chronic respiratory failure, COPD duration and age. Conclusions: CB prevalence in GOLD stage 2 4 COPD patients is high. CB is related to current tobacco smoking, and prevalence increases with COPD severity and duration, emphysema and age. CB could be the hallmark of a subtype of COPD easy to identify in clinical practice, associated with increased disease severity and increased risk of exacerbation. Introduction What s known Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease with variable clinical presentation. Chronic bronchitis is present in a considerable percentage of COPD patients, but prevalence figures reported vary widely. What s new The present study analyses the prevalence of chronic bronchitis and its association with disease and patient characteristics in a cross-sectional multicentre study in a large population of Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage 2 4 COPD patients. Chronic bronchitis could be the hallmark of a subtype of COPD associated with increased disease severity and increased risk of exacerbation. Chronic obstructive pulmonary disease (COPD) is a major health problem and a leading cause of morbidity and mortality in the adult population worldwide. Increasing evidence indicates that COPD is a heterogeneous disease (1) and that different subtypes or phenotypes may contribute to a variable extent to the severity of the disease. Chronic bronchitis (CB) is a clinically and epidemiologically useful entity, defined as the presence of cough and sputum production for at least 3 months in each of 2 consecutive years. Chronic cough and sputum production are symptoms of mucus hypersecretion in proximal airways. CB is a well-known feature present in a considerable percentage of COPD patients, although it was not included in the definition of COPD in any version of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) report (2). Chronic bronchitis is considered a major manifestation of COPD, but the prevalence of CB in COPD varies widely. Lower prevalence was found in populationbased studies (14 30%) (3,4) and higher prevalence in clinical-based studies (27.3 74.1%) (1,5 10). It is suggested that CB is not an innocent manifestation in patients with COPD, as epidemiologic studies observed that CB in COPD patients is associated with increased mortality (11 13), frequent exacerbations including those requiring hospitalisation (4,7,9,10), worse respiratory symptoms (4,7), disease progression (14), physical activity limitation (4), and worse general health status (1,4,7). However, in the large observational evaluation of COPD longitudin- 1294. doi: 10.1111/ijcp.12248

Chronic bronchitis in COPD patients 1295 ally to identify predictive surrogate endpoints (ECLIPSE) study, CB was not associated with an accelerated rate of decline in forced expiratory volume in 1 second (FEV 1 ) (8), leaving the impact of CB on the disease course of COPD still controversial, despite some biological evidence of a distinct inflammatory cell profile in the airways of COPD patients with CB (15). CB associated with COPD could be a therapeutic target (16), reminding that early studies that validated inhaled corticoids in COPD selected patients on the presence of symptoms of CB (17,18) and that recent studies validated phosphodiesterase-4 inhibitor in preventing the development of exacerbation in COPD patients with CB (2). To address the uncertainty regarding the prevalence of CB in COPD patients, we analysed the prevalence of CB and its association with different disease and patient characteristics in a large population of GOLD stage 2 4 COPD patients from Belgium and Luxembourg. Methods Population and study design In this cross-sectional multicentre study, we investigated the prevalence of CB and its association with disease and patient characteristics, exacerbations and COPD treatment in ambulant patients with moderate-to-severe COPD. Between October 2010 and April 2011, pneumologists from 16 pneumology departments in Belgium and Luxembourg (8 from the Flemish and 7 from the French speaking part of Belgium, 1 from Luxembourg) enrolled consecutive COPD outpatients fulfilling the study s eligibility criteria. To be eligible, patients had to present with COPD of stages 2 4 according to the GOLD diagnostic criteria (2). The study protocol was approved by the ethics committee of Liege (B 70720109208) which acted as the central ethics committee for Belgium, and by the local ethics committees of all participating centres. Written informed consent was obtained from all patients prior to or at inclusion. Data collection Data were collected using a structured questionnaire available in Dutch, French and German (Online supplemental Material/Appendix) in a single outpatient visit. The standardised questionnaire covered demographic data, COPD characteristics (duration, severity, presence of CB and emphysema), spirometric data, presence of pulmonary risk factors and comorbidities, exacerbations and hospitalisations over the last 12 months, and current COPD maintenance treatment. Demographic data included age, gender, height and weight. Severity of pulmonary obstruction was assessed using the GOLD staging criteria (2). Spirometric data comprised postbronchodilator FEV 1 (% predicted and absolute values) and postbronchodilator FEV 1 /VC (vital capacity, %). If available, data of three spirometric evaluations were collected; the most recent spirometry and spirometric evaluations performed about 6 months (interval of 4 8 months) and 12 months (interval of 10 14 months) before inclusion. The presence of emphysema was appreciated by the chest physician and radiologist, and based on their interpretation of CT scan of the thorax and/or on transfer factor of lung for carbon monoxide (TLCO) measurement. Chronic bronchitis was defined as cough and sputum production for at least 3 months in each of two consecutive years, in the absence of other causes of chronic cough (19). Smoking history included smoking status and number of pack-years. To investigate occupational risk factors, the exposure to mineral, organic or metallic dusts, fumes and vapours was assessed. The number and severity of exacerbations over the last 12 months was recorded. Exacerbations were graded as mild when the increased severity of respiratory symptoms could be controlled by the patient with an increase in the usual medication, as moderate when requiring treatment with systemic corticosteroids and/or antibiotics and as severe if hospitalisation or a visit to the emergency department was required. The number of exacerbations was determined according to the patient s self-reported events (for mild and moderate exacerbations), and as noted in the medical file (for moderate and severe exacerbations). Additionally, the number of COPDrelated visits to the emergency department, visits to the emergency department without hospitalisation, COPD-related hospitalisation with or without stay in the intensive care unit and the overall number of hospitalisations (including those because of COPD) over the last year were recorded. Comorbidities were diagnosed by the chest physician according to personal clinical judgement and/or recorded from the data file of the patient. Comorbidities recorded included respiratory diseases (allergy, asthma and chronic respiratory failure), cardiovascular diseases (ischaemic heart disease, heart failure and hypertension), osteoporosis, skeletal muscle wasting, cachexia, diabetes mellitus, depression, anaemia, gastroesophageal reflux disease and lung cancer. The COPD treatment parameters comprised current COPD treatment as well as long-term oxygen therapy, influenza vaccination and smoking cessation therapy over the last 12 months.

1296 Chronic bronchitis in COPD patients Determination of sample size As the reported prevalence of CB in COPD patients varies between 50% and 80%, a study population of approximately 1000 COPD patients is needed in order to have at least 251 patients in the group without CB. A sample size of 251 patients per group is necessary for a two-sided statistical test to detect a difference of 0.25 exacerbations/patient/year between patients with and without CB with at least 95% power, at a significance level of alpha = 0.05 and with an assumed common standard deviation of 1.0. An earlier study performed in France by Burgel et al. (10) with a similar design and end-points compared patients with and without CB in a cohort of 433 subjects. With a sample size of about 1000 patients it would be possible to identify additional features associated with CB in COPD patients. Statistical analysis Patients were only included in the statistical analysis if the following minimal data were available: gender and age; GOLD stage; presence or absence of CB; number and severity of exacerbations over the last 12 months. Quantitative variables are reported as the mean standard deviation or median (5th and 95th percentiles); categorical variables are presented as percentages. Percentages are based on the total number of patients in the analysis set, i.e. missing values are also tabulated including percentages unless otherwise specified. Number and percentage of missing observations are provided for all parameters. Differences between patients with and without CB were analysed using Fisher s exact test or the v 2 test for binary variables, Mann Whitney U-test for ordinal variables, and unpaired t-tests for normally distributed variables. CB prevalence was calculated using the normal approximation and presented as point estimate with the corresponding two-sided 95% confidence interval. All variables associated with the CB subtype with a p < 0.10 in univariate analysis were included in a multivariate logistic regression model to identify variables independently associated with CB. Statistical analysis was performed with Microsoft Excel version 2007 and Medcalc version 8.2.1.0 software (MedCalc Software, Mariakerke, Belgium). A p-value of 0.05 was considered statistically significant. Results Population characteristics and prevalence of chronic bronchitis Data from 976 subjects were collected, 625 from the Flemish part and 311 patients from the French part of Belgium, and 38 patients from Luxembourg. Two patients were excluded from the analysis because the minimal criteria for analysis were not met. Thus, data analysis was performed for 974 subjects. The characteristics of the study population are summarised in Table 1. Patients in this study were on average 67.8 9.6 years old, 72% were male and mean FEV 1 was 52.5 15.8% of predicted. The prevalence of CB was 64% (CB present in 622/974 patients). Comparison of COPD patients with and without chronic bronchitis No significant difference was observed between patients with and without CB for age, gender, smoking status, body mass index (BMI) (Table 1) and prevalence of most comorbidities (asthma, allergy, cardiovascular disease, osteoporosis, anaemia, lung cancer, depression and diabetes) (Table 2). Chronic respiratory failure, cachexia and skeletal muscle wasting, however, were more frequently present in the group with CB. The number of pack-years and COPD duration were higher, and both FEV 1 % predicted and FEV 1 /VC% were lower in patients with CB. The proportion of patients with CB increased with GOLD stage (Figure 1) and the prevalence of CB was higher in patients with emphysema and in patients exposed to occupational risk factors. Chronic bronchitis and COPD exacerbations The overall number of exacerbations per patient per year was significantly higher in patients with CB (2.08 2.78 vs. 1.05 1.71; p < 0.001), as were the numbers of mild (0.7 1.9 vs. 0.3 0.9; p = 0.002), moderate (1.1 1.4 vs. 0.5 0.9; p < 0.0001) and severe (0.3 0.8 vs. 0.2 0.6; p = 0.0185) exacerbations per patient per year (Table 3). The percentage of patients with frequent (moderate or severe) exacerbations (defined as two or more per patient per year) was higher in the CB group than in the group without CB (37.3% vs. 14.2% of patients; p < 0.0001). The percentage of patients who experienced mild, moderate or severe exacerbations over the last year was higher in the CB group (Figure 2). More hospitalisations and more COPD-related hospitalisations without stay at the intensive care unit were observed for COPD patients with CB (Table 4). COPD treatment in patients with and without chronic bronchitis Chronic obstructive pulmonary disease patients with CB are treated more often with theophyllines, antibiotics and smoking cessation therapy, but less frequently used long-acting muscarinic antagonist (LAMA) monotherapy than COPD patients without

Chronic bronchitis in COPD patients 1297 Table 1 Population characteristics and comparison of chronic obstructive pulmonary disease patients with (CB+) and without (CB ) chronic bronchitis Total CB+ CB N = 974 N = 622 N = 352 p-value Age (years) 67.75 9.63 68.16 9.43 67.02 9.93 0.0768* Gender (% female) 28.0 27.5 29.0 0.6733 Disease duration (years) 7.31 6.07 7.83 6.12 6.37 5.87 0.0003* FEV1 (% predicted) 52.47 15.83 50.8 15.9 55.3 15.2 <0.0001* FEV1/VC (%) 52.32 11.46 51.5 11.8 53.7 10.8 0.0057* Smoking status, % (n) Never smoked 5.85 (57) 4.98 (31) 7.39 (26) 0.097 Former smoker 66.22 (645) 63.99 (398) 70.17 (247) Current smoker 27.82 (271) 30.87 (192) 22.44 (79) Unknown 0.10 (1) 0.16 (1) 0.00 (0) Number of pack years, % (n) <10 years 1.97 (18) 1.53 (9) 2.76 (9) 0.043* 10 20 years 8.52 (78) 8.64 (51) 8.28 (27) >20 years 88.54 (811) 88.81 (524) 88.04 (287) Unknown 0.98 (9) 1.02 (6) 0.92 (3) Body mass index (kg/m²) 26.05 5.40 25.96 5.41 26.21 5.38 0.4877 Emphysema, % (n) Present 60.47 (589) 64.79 (403) 52.84 (186) <0.0001* Absent 38.09 (371) 33.12 (206) 46.88 (165) Unknown 1.44 (14) 2.09 (13) 0.28 (1) Occupational risk factors, % (n) Present 40.66 (396) 41.64 (259) 38.92 (137) 0.025* Absent 59.03 (575) 58.04 (361) 60.80 (214) Unknown 0.31 (3) 0.32 (2) 0.28 (1) *p < 0.05. Postbronchodilator. Refers to current and former smokers only. Patients were considered to have chronic bronchitis (CB+) if they noted chronic cough and phlegm production for 3 month/year over two consecutive years. CB, chronic bronchitis; FEV 1, forced expiratory volume in 1 second; VC, vital capacity. CB (Table 5). The percentage of COPD patients treated with systemic corticoids tended to be higher in the CB group, but this trend did not reach statistical significance. Identification of variables independently associated with chronic bronchitis In a multiple logistic regression analysis, we found that frequent exacerbation was the most important independent variable associated with CB, followed by current smoking. Other variables significantly associated with CB were chronic respiratory failure, duration of COPD and age (Table 6). Gender, BMI, allergy, asthma, GOLD stage and comorbidities were tested but not included in the logistic regression model. Discussion This observational study in a large population of COPD patients found that the prevalence of CB in COPD patients is high (64%) and increases with the severity of obstruction. In our study, the presence of CB was associated with risk factors such as smoking status, duration of COPD, age, and also with features such as chronic respiratory failure and a history of frequent (moderate or severe) COPD exacerbations. Among the systemic comorbidities investigated, cachexia and skeletal muscle wasting were more frequent in COPD patients with CB. The prevalence of CB in our study population of COPD patients is higher than that reported in most previous studies (3 9). Differences in the prevalence of CB between studies could result from the definition of CB used as well as from differences between the study populations as suggested recently by Burgel (20). Indeed, we had a selected population of symptomatic patients with moderate to very severe COPD consulting the outpatient pneumology clinic. The mean age and the proportion of current smokers were also high in our COPD cohort. It is well known that the risk of CB increases with age and tobacco consumption (21). In a recent large Danish co-twin

1298 Chronic bronchitis in COPD patients Table 2 Presence of concomitant diseases in chronic obstructive pulmonary disease patients with (CB+) and without (CB ) chronic bronchitis CB+ CB N = 622 N = 352 p-value N # Respiratory diseases Allergy 8.9 (55) 12.5 (44) 0.0675 952 Asthma 10.8 (67) 13.1 (46) 0.2963 957 Chronic respiratory failure 18.3 (114) 9.4 (33) 0.0002* 960 Cardiovascular diseases Ischaemic heart disease 22.1 (137) 19.9 (70) 0.5144 956 Heart failure 13.5 (84) 11.9 (42) 0.6061 953 Hypertension 39.5 (246) 39.8 (140) 0.9740 953 Other relevant disease Osteoporosis 17.7 (110) 17.0 (60) 0.4776 804 Skeletal muscle wasting 22.8 (142) 13.4 (47) 0.0004* 959 Cachexia 12.7 (79) 6.8 (24) 0.0052* 964 Diabetes mellitus 12.7 (79) 10.5 (37) 0.3342 965 Depression 19.1 (119) 14.2 (50) 0.0636 953 Anaemia 6.4 (40) 5.1 (18) 0.6445 842 Gastroesophageal reflux 21.2 (132) 19.0 (67) 0.4232 941 Lung cancer 6.6 (41) 9.4 (33) 0.1650 949 *p < 0.05. N #, number of patients with available data; CB, chronic bronchitis. % of Patients 80% 60% 40% 20% 0% CB CB+ p < 0.0001 II III IV GOLD Stages Figure 1 Distribution of global initiative for chronic obstructive lung disease stages and chronic bronchitis in our chronic obstructive pulmonary disease population. CB, chronic bronchitis control analysis, the risk of CB was confirmed to increase among smokers in a dose-dependent manner and in older subjects (22). Moreover, in a French study with the same methodology and a comparable proportion of smokers, the prevalence of CB was similar (74%) (10). We found an increased prevalence of emphysema in our CB subgroup, in contrast to the COPDGene study, where there was no difference in the percentage of patients with emphysema between COPD groups with and without CB (7). This could be explained by the fact that in our study, emphysema was reported retrospectively by the appreciation of chest physician and radiologist based on their interpretation of CT scan of the thorax and/or TLCO, whereas the COPDGene study prospectively assessed Table 3 Chronic obstructive pulmonary disease exacerbations in patients with (CB+) and without chronic bronchitis (CB ) CB+ CB N = 622 N = 352 p-value All 2.08 2.78 1.05 1.71 <0.001* Mild 0.7 1.9 0.3 0.9 0.002* Moderate 1.1 1.4 0.5 0.9 <0.0001* Severe 0.3 0.8 0.2 0.6 0.0185* *p < 0.05. The number of exacerbations per patient per year is expressed as mean SD. CB, chronic bronchitis. the degree of emphysema by CT scan of the thorax and the measurement of Hounsfields units. We also found that COPD patients with CB had a worse pulmonary function, and that prevalence of CB increased with GOLD stage. These findings are consistent with those reported in both unselected and selected populations of COPD (1,4), although in the COPDGene study, no difference in pulmonary function was observed between COPD groups with or without CB (7). Chronic obstructive pulmonary disease is well known to be associated with systemic manifestations and comorbidities (23). We found that in our COPD patients, among all examined systemic manifestations

Chronic bronchitis in COPD patients 1299 Figure 2 Percentage of patients with mild, moderate, severe and frequent exacerbations of chronic obstructive pulmonary disease. Frequent exacerbations: at least two moderate and/or severe exacerbations per year. CB, chronic bronchitis Table 4 Hospitalisations in chronic obstructive pulmonary disease (COPD) patients with (CB+) and without chronic bronchitis (CB ) CB+ CB N = 622 N = 352 p-value N # Hospital visits related to COPD Emergency department without hospitalisation 0.04 0.24 0.03 0.25 0.3445 968 Hospitalisation without stay at ICU 0.28 0.75 0.15 0.43 0.0027* 969 Hospitalisation with stay at ICU 0.05 0.26 0.03 0.16 0.1247 967 All-cause hospitalisations 0.52 0.91 0.38 0.83 0.0185** 964 *p < 0.01, **p < 0.05. The number of hospitalisations per patient per year is expressed as mean SD. N #, number of patients with available data; CB, chronic bronchitis; ICU, intensive care unit. and comorbidities, cachexia and skeletal muscle wasting were more frequently encountered in the CB than in the non-cb group. Moreover, hospitalisations not related to COPD were more frequent in the CB group. Of interest is the fact that prevalence of anaemia (5%), a recently recognised comorbidity in COPD (24), is similar between the COPD patients with and without CB. Our study clearly shows that CB in COPD patients is associated with a higher risk of frequent exacerbations, including severe exacerbations requiring hospitalisation. This confirms the findings of Burgel (10) in a French multicentre study including 433 COPD patients. However, in the epidemiological ECLIPSE study, the CB subtype in COPD was not associated with an increase in moderate and severe exacerbations, neither in the year prior to study entry (1) nor during the first year of follow-up (25). Reasons for these discrepancies in association with CB and COPD exacerbations remain to be established. A larger percentage of COPD patients with CB are using oral theophyllines, preventive antibiotics and smoking cessation therapy. This could be explained by the fact that the CB subpopulation has more obstruction and exacerbations, and contains a greater proportion of current smokers. Moreover, the fact that smoking cessation therapy is more frequent in the group of chronic bronchitis could hint to a poorer quality of life (1,4,7) pushing patients to stop smoking. Inhaled therapies did not differ between COPD patients with and without CB, except that those with CB less often used LAMA monotherapy. That is surprising because LAMA is well known to produce benefits on symptoms and exacerbations rates (26). Burgel et al. did not find differences in inhaled treatments between both groups of COPD patients (10). In contrast, in the COPDGene study the CBgroup reported a greater use of short-acting bronchodilators and lesser use of combination inhaled steroid/long-acting b-agonists, but the use of LAMA was similar in both subgroups (7). Additional epidemiological studies focusing on the treatment of COPD are necessary to clarify these differences. We also would like to underline that approximately 75%

1300 Chronic bronchitis in COPD patients Table 5 Current chronic obstructive pulmonary disease treatment in patients with (CB+) and without chronic bronchitis (CB ) Total CB+ CB N = 974 N = 622 N = 352 p-value No treatment 1.4 (14) 1.8 (11) 0.9 (3) 0.4009 Inhaled therapy LABA only 6.8 (66) 6.6 (41) 7.1 (25) 0.791 LAMA only 18.1 (176) 14.5 (90) 24.4 (86) <0.0001* LABA + ICS 75.2 (732) 76.7 (477) 72.44 (255) 0.1433 LABA + ICS + LAMA 60.1 (585) 58 (361) 63.6 (224) 0.0891 Oral therapies Theophylline 17.5 (187) 22.0 (137) 14.2 (50) 0.003* Corticosteroids 8.3 (81) 9.6 (60) 6 (21) 0.0529 Mucolytics 20.2 (197) 20.0 (124) 20.7 (73) 0.8034 Antibiotics 4.0 (39) 5.6 (35) 1.1 (4) 0.0003* Other therapies Oxygen 10.5 (102) 10.1 (63) 11.1 (39) 0.6637 Smoking cessation therapy 18.3 (178) 21.5 (134) 12.5 (44) 0.0004* Influenza vaccination 71.3 (694) 71.7 (446) 70.4 (248) 0.7126 *p < 0.05. Over the last 12 months. Data are expressed as percentage of patients treated (number of patients). CB, chronic bronchitis; ICS, inhaled corticosteroids; LABA, long-acting b 2 -agonist; LAMA, long-acting muscarinic antagonist. Table 6 Factors associated with the presence of chronic bronchitis in chronic obstructive pulmonary disease (COPD) patients Variable b SE Estimated OR 95% CI p-value Frequent exacerbations 0.60 0.09 3.33 3.31 4.79 < 0.0001* Current smoker 0.41 0.09 2.26 1.58 3.22 < 0.0001* Chronic respiratory failure 0.31 0.12 1.88 1.19 2.97 0.0071* Duration of COPD 0.03 0.01 1.03 1.01 1.06 0.0199* Age (years) 0.02 0.01 1.02 1.00 1.03 0.0370* Emphysema 0.14 0.08 1.31 0.98 1.76 0.0726 *p < 0.05. Multivariate logistic regression with chronic bronchitis as the dependent variable. Variables included in this model were those associated with chronic bronchitis with p < 0.10 in univariate analysis. b, estimated coefficient; OR, odds ratio; SE, standard error; CI, confidence interval. of patients were receiving ICS in both the CB and the non-cb group. However, ICS is currently recommended in severe COPD (< 50% predicted) and frequent exacerbations. As the lack of chronic bronchitis is associated with clearly less exacerbation, this points to an overuse of ICS in the group without CB. This is important as there is now a clear recognition of side effects of ICS in COPD (namely pneumonia) revealed by recent trials or observational studies (2,27). Several limitations of the present study must be taken into account. First, the allocation of a patient to the CB group is based on the perception of the symptoms by the patient himself, which may be subject to a recall or interpretation bias. Second, data were obtained cross-sectionally, and the number of exacerbations and their severity were determined according to medical files and patient recollections. In particular the data collected from the patients may be incomplete, as Seemungal found that about 50% of acute exacerbations of COPD are not reported by patients (28). Thus, mild exacerbations could be underestimated in the present study. However, moderate and especially severe exacerbations were evaluated objectively based on medical records, and both were more frequent in the CB group.

Chronic bronchitis in COPD patients 1301 Finally, our study helps to understand the prevalence of CB in an unselected COPD population and suggest that CB in COPD is possibly associated with worse outcomes. COPD with CB could represent a subtype of COPD that is easy to identify in clinical practice and appears to be associated with increased disease severity and with a high risk of exacerbations. Acknowledgements This research was supported by Takeda. The authors acknowledge the contribution of Veerle Persy (HuginMugin Research) as an independent medical writer for editing and writing assistance, funded by Takeda Belgium. Author contributions Corhay JL, Vincken W, Schlesser M, Bossuyt P and Imschoot J participated in the conception and design of the study, in the acquisition of data, in the analysis and interpretation of data. Corhay JL drafted the article; all authors participated in revising it critically for important intellectual content. All authors gave their final approval on the version to be submitted. References 1 Agusti A, Calverley P, Celli B et al. Characterisation of COPD heterogeneity in the ECLIPSE cohort. Respir Res 2010;11:122 35. 2 From the Global Strategy for the Prevention of Chronic Obstructive Pulmonary Disease. Global Initiative for Chronic Obstructive Lung Disease (GOLD) [Internet]. 2013. http://www.goldcopd. org/ (accessed July 2013). 3 Lu M, Yao W, Zhong N et al. Chronic obstructive pulmonary disease in the absence of chronic bronchitis in China. Respirology 2010; 15: 1072 8. 4 Montes de Oca M, Halbert RJ, Lopez MV et al. The chronic bronchitis phenotype in subjects with and without COPD: the PLATINO study. Eur Respir J 2012; 40: 28 36. 5 Kanner RE, Anthonisen NR, Connett JE. Lower respiratory illnesses promote FEV(1) decline in current smokers but not ex-smokers with mild chronic obstructive pulmonary disease: results from the lung health study. Am J Respir Crit Care Med 2001; 164: 358 64. 6 Marsh SE, Travers J, Weatherall M et al. Proportional classifications of COPD phenotypes. Thorax 2008; 63: 761 7. 7 Kim V, Han MK, Vance GB et al. The chronic bronchitic phenotype of COPD: an analysis of the COPDGene Study. Chest 2011; 140: 626 33. 8 Vestbo J, Edwards LD, Scanlon PD et al. Changes in forced expiratory volume in 1 second over time in COPD. N Engl J Med 2011; 365: 1184 92. 9 Miravitlles M, Guerrero T, Mayordomo C, Sanchez-Agudo L, Nicolau F, Segu JL. Factors associated with increased risk of exacerbation and hospital admission in a cohort of ambulatory COPD patients: a multiple logistic regression analysis. The EOLO Study Group. Respiration 2000; 67: 495 501. 10 Burgel P-R, Nesme-Meyer P, Chanez P et al. Cough and sputum production are associated with frequent exacerbations and hospitalizations in COPD subjects. Chest 2009; 135: 975 82. 11 Prescott E, Lange P, Vestbo J. Chronic mucus hypersecretion in COPD and death from pulmonary infection. Eur Respir J 1995; 8: 1333 8. 12 Pelkonen M, Notkola I-L, Nissinen A, Tukiainen H, Koskela H. Thirty-year cumulative incidence of chronic bronchitis and COPD in relation to 30-year pulmonary function and 40-year mortality: a follow-up in middle-aged rural men. Chest 2006; 130: 1129 37. 13 Ekberg-Aronsson M, Pehrsson K, Nilsson J-A, Nilsson PM, L ofdahl C-G. Mortality in GOLD stages of COPD and its dependence on symptoms of chronic bronchitis. Respir Res 2005;6: 98. 14 Vestbo J, Prescott E, Lange P. Association of chronic mucus hypersecretion with FEV1 decline and chronic obstructive pulmonary disease morbidity. Copenhagen City Heart Study Group. Am J Respir Crit Care Med 1996; 153: 1530 5. 15 Snoeck-Stroband JB, Lapperre TS, Gosman MME et al. Chronic bronchitis sub-phenotype within COPD: inflammation in sputum and biopsies. Eur Respir J 2008; 31: 70 7. 16 Kim V, Criner GJ. Chronic bronchitis and chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2013; 187: 228 37. 17 Dompeling E, Van Schayck CP, Molema J, Folgering H, Van Grunsven PM, Van Weel C. Inhaled beclomethasone improves the course of asthma and COPD. Eur Respir J 1992; 5: 945 52. 18 Paggiaro PL, Dahle R, Bakran I, Frith L, Hollingworth K, Efthimiou J. Multicentre randomised placebo-controlled trial of inhaled fluticasone propionate in patients with chronic obstructive pulmonary disease. Lancet 1998; 351: 773 80. 19 Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. American Thoracic Society. Am J Respir Crit Care Med 1995; 152(5 Pt 2): S77 121. 20 Burgel P-R. Chronic cough and sputum production: a clinical COPD phenotype? Eur Respir J 2012; 40: 4 6. 21 Montnemery P, Adelroth E, Heuman K et al. Prevalence of obstructive lung diseases and respiratory symptoms in southern Sweden. Respir Med 1998; 92: 1337 45. 22 Meteran H, Thomsen SF, Harmsen L, Kyvik KO, Skytthe A, Backer V. Risk of chronic bronchitis in twin pairs discordant for smoking. Lung 2012; 190: 557 61. 23 Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. Eur Respir J 2009; 33: 1165 85. 24 Portillo K, Martinez-Rivera C, Ruiz-Manzano J. Anaemia in chronic obstructive pulmonary disease. Does it really matter? Int J Clin Pract 2013; 67: 558 65. 25 Hurst JR, Vestbo J, Anzueto A et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010; 363: 1128 38. 26 Tashkin DP, Celli B, Senn S et al. A 4-year trial of tiotropium in chronic obstructive pulmonary disease. N Engl J Med 2008; 359: 1543 54. 27 Nannini LJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta(2)-agonist in one inhaler versus long-acting beta(2)-agonists for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2012; 9:CD006829. doi: 10.1002/ 14651858.CD006829.pub2. 28 Seemungal TA, Donaldson GC, Paul EA, Bestall JC, Jeffries DJ, Wedzicha JA. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998; 157(5 Pt 1): 1418 22. Paper received May 2013, accepted July 2013