Prognosis of COPD patients requiring frequent hospitalization: Role of airway infection

Transcription

1 Respiratory Medicine (2010) 104, 840e848 available at journal homepage: Prognosis of COPD patients requiring frequent hospitalization: Role of airway infection Feliu Renom a, *, Aina Yáñez b, Margarita Garau c, Mateu Rubí a, Maria-José Centeno a, Maria-Teresa Gorriz a, Magdalena Medinas a, Ferran Ramis a, Joan B. Soriano b,d,àlvar Agustí d,e a Respiratory Department, Hospital Joan March, Carretera Palma-Soller km 12,5, Bunyola, Balearic Island, Spain b Program of Epidemiology & Clinical Research, Fundació Caubet-CIMERA, Bunyola, Balearic Island, Spain c Microbiology Department, Fundació Hospital Son Llàtzer, Palma, Balearic Island, Spain d CIBER de Enfermedades Respiratorias (CIBERES), Spain e Institut Clínic del Torax, Hospital Clínic, Barcelona, Spain Received 20 October 2009; accepted 21 December 2009 Available online 27 January 2010 KEYWORDS Chronic obstructive; Pulmonary disease; Mortality; Airway infection; Microbiology; BODE index Summary Rationale: A subgroup of patients with chronic obstructive pulmonary disease require frequent hospitalization because of exacerbations of the disease. We hypothesized that airway infection by non-usual pathogens is a major factor driving hospitalization needs in these patients. Objectives: 1) To describe the clinical and functional characteristics of a cohort of COPD patients requiring 2 hospitalizations per year; 2) to determine prospectively their microbiological pattern during exacerbations; and, 3) to analyze the prognostic value of several clinical, functional and microbiological variables with respect to hospitalizations and mortality. Methods: Open cohort study of 116 COPD patients who had been hospitalized at least twice during the last 12 months. Patients were followed for an average of 21 months. Measurements and main results: Clinical data, forced spirometry and 6 min walking distance were determined, and the BODE index was calculated, at the time of inclusion in the study. During follow-up, sputum culture was obtained during exacerbations, and hospitalization and mortality were collected every two months. Mean age was 71 yrs, and 94% of patients were male. Main findings show that: 1) not all patients had severe disease according to either the degree of airflow limitation or the BODE index; 2) non-usual pathogens, mainly Pseudomonas aeruginosa, other gram-negative non-fermentative rods and Enterobacteriaceae, were isolated among 71.1% of the sputum obtained during exacerbations; and, 3) these pathogens were associated with poor prognosis and frequent hospitalization. * Corresponding author. Tel.: þ ; fax: þ address: frenom@gesma.caib.es (F. Renom) /$ - see front matter ª 2009 Elsevier Ltd. All rights reserved. doi: /j.rmed

2 Role of airway infection 841 Conclusions: Airway infection by non-usual pathogens appears to be a key driver of frequent hospitalizations and mortality in COPD. ª 2009 Elsevier Ltd. All rights reserved. Introduction Patients with chronic obstructive pulmonary disease (COPD) often present with episodes of exacerbation (ECOPD) during the course of their disease. These episodes are important events in the natural history of COPD because they accelerate the decline of lung function, worsen the prognosis of the patient 1,2 and impose a significant economic burden on society. 3e5 Although the mechanisms of ECOPD are likely to be multi-factorial, it is widely accepted that airway bacterial infection plays a key pathogenic role. 6 Besides, chronic airway colonization is recognized today as an important driver of COPD progression. 7 Not all COPD patients, however, have airway bacterial colonization, and the frequency of ECOPD also varies greatly among them. Thus, whereas some patients never present ECOPD, others exacerbate several times every year. Among the latter, there is a subgroup of patients who are hospitalized repeatedly. 8 The clinical, functional and microbiological characteristics of these patients are poorly understood. We hypothesized that airway infection by non-usual pathogens is a major factor driving hospitalization requirements in these latter subgroup of COPD patients. In 2002, we instituted in our hospital an integrated, multidisciplinary program aimed at providing clinical, rehabilitation and educational support to patients with advanced COPD (Online supplement Table E1), like others have done before successfully. 9e12 Therapeutic strategies were established by severity levels in stable COPD, and aiming to identify and manage ECOPD, according to GOLD guidelines. 13 Because one of the possible criteria for inclusion in our program was that the patient had required hospitalization because of ECOPD two or more times during the last 12 months, this program offered us the opportunity to characterize and follow-up this particular sub-population of COPD patients. Accordingly, in this study we aimed: (1) to describe their clinical and functional characteristics at the time of inclusion in our program; (2) to determine prospectively the microbiological pattern of airway infection during ECOPD; and, (3) to analyze the prognostic value of several clinical, functional and microbiological variables with respect to the frequency of hospitalizations and mortality. Methods Patients From October 2002 to December 2006, we included in our program, after obtained informed consent, 202 patients with advanced respiratory disease, of whom 140 required two or more respiratory hospitalizations during the last 12 months. Of these, 116 were diagnosed of COPD according to the GOLD guidelines. 13 All patients with diagnoses other than COPD were excluded of our study (Fig. 1). Design of the study This is an open cohort observational study. At the time of inclusion in our program, clinical data were obtained in stable phase (that is at least one month after the last ECOPD). Spirometry was obtained in all subjects and in 107 of them, the 6 min walking distance (6MWD) was also determined. Patients were then followed-up during an average of 21 months (range, 1e51 months). During this period, patients were visited every two months, or more often if clinically required or by patient request. The study protocol was approved by our Hospital Research Committee. Measurements Dyspnea was quantified using the modified Medical Research Council (mmrc) questionnaire. 14 Co-morbidities were assessed using the Charlson index. 15 Spirometry was measured according to standard international recommendations. 16 Reference values correspond to a Mediterranean population. 17 The 6-min walking distance (6MWD) was determined following 2002 ATS guidelines. 18 The BODE index was calculated according to Celli et al. 19 Culture of spontaneous sputum was obtained in all ECOPD episodes that fulfilled GOLD criteria, 13 occurred during follow-up and, according to the criteria of Anthonisen, 20 were of potential infectious origin, whether the patient was hospitalized or not. Culture of sputum was not required in stable state. The quality of sputum samples was assessed using the scoring system of Murray and Washington. 21 Sputum samples were processed 22 and bacterial colonies were identified by standard methods 23 in a central laboratory. All isolates were tested for antibiotic susceptibility following the CLSI (previously NCCLS) recommendations. 24 Bacterial isolates were grouped in two categories: usual pathogens, including Haemophilus influenzae, Streptococcus pneumoniae, and Moraxella catarrhalis 6,7,25e29 ; and non-usual pathogens, including Pseudomonas aeruginosa and other non-fermentative gram-negative rods (Stenotrophomonas maltophilia and Achromobacter xylosoxidans), Enterobacteriaceae (Klebsiella pneumoniae, Escherichia coli, Proteus mirabilis, Enterobacter cloacae, Serratia marcescens and Morganella morganii), Corynebacterium spp. (Corynebacterium striatum in almost all isolates), methicillin-resistant Staphyloccocus aureus (MRSA) and Aspergillus spp. Antibiotics used to treat each bacteria are listed in the online supplement (Table E2). For the analysis, patients were classified in four different groups: 1) no pathogen (no pathogen isolates

3 842 F. Renom et al. Patients in the continuity of care for chronic advanced respiratory patients program(resc), from October 2002 to December patients Two or more hospitalizations in the previous year One or no hospitalization in the pr evious year COPD : 116 Other : 24: COPD : 46 Other : 16: Asthma: 8 Asthma: 6 Bronchiectasis: 7 Bronchiectasis: 5 Sequelae post - TB: 4 Sequelae post - TB: 0 Ventilatory disorders: 5 Ventilatory disorders: 3 Inters titial pulmonary disease: 3 Interstitial pulm. dis.: 1 Not identified: 1 Not identified: 1 Inclusion: Functional assessment in stable phase Control every two months (at the least) Sputum culture in all COPD exacerbations Outcome at 2 4 months : Lost: 18 (15,5%) Exitus: 35 (30,2%) Completers: 63 (54,3%) Outcome at 51 months: Lost: 24 (20,7%) Exitus: 43 (37,1%) Completers: 49 (42,2%) Figure 1 Flow chart of selections of patients. during follow-up); 2) usual pathogen (one or more usual pathogens isolates); 3) non-usual pathogen (one or more non-usual pathogens isolates) and 4) both (one or more usual and non-usual pathogens isolates). Statistical analysis Results are presented as mean standard deviation, and range or percentage as required. Continuous variables are analyzed using the Student s t-test and ANOVA, whereas categorical variables are analyzed using the c 2 test. Nonparametric test (ManneWhitney U test, KruskaleWallis) were used for variables that were not normally distributed. A KaplaneMeier survival analysis was performed to analyze mortality. The unadjusted association of predictors with mortality was estimated with a bivariate semiparametric Cox model. Due to in our cohort only a few patients were classified in the first two quartiles of the BODE index (Q1 and Q2), both were pooled together for the survival analysis and were used as the reference category. Those predictors found to be significantly associated with mortality in the bivariate analysis were assessed in an extension of the multivariate Cox s proportional hazard model with timedependent co-variates. 30,31 Hazard ratios and 95% confidence intervals were calculated. Sputum culture was introduced as a time-dependent co-variate, which allowed to include the microbiological data collected at different follow-up visits. Thus, any changes in the patient s exposures (i.e., changes in microbiological pattern) during follow-up was taken into account. A p value lower than 0.05 was considered statistically significant. All statistical analysis were performed using SPSS v.15.0 and STATA v.8.0. Results Clinical and functional characteristics Table 1 presents the main clinical and functional characteristics of the 116 patients studied. Most of the patients were male, mean age was 71 yrs and the mean of

4 Role of airway infection 843 Table 1 Clinical and functional characteristics of patients (n Z 116, unless noted otherwise). Mean SD Range No of previous e8 hospitalizations/yr Age, yrs e90 Males, % 109 (94%) e Smoking history, pack-year e160 Current smokers, % 13 (11.2%) e Charlson index e5 BMI, kg/m e46.8 Dyspnea score, mmrc e4 FEV 1 /FVC, % e67 FEV 1, % reference e77 GOLD I, n (%) 0 (0) e GOLD II, n (%) 19 (16.4) e GOLD III, n (%) 60 (51.7) e GOLD IV, n (%) 37 (31.9) e 6MWD, meters (n Z 107) e500 BODE index (n Z 107) e9 BODE Q1, n (%) 8 (7.5%) e BODE Q2, n (%) 25 (23.4%) e BODE Q3, n (%) 39 (36,4%) e BODE Q4, n (%) 35 (32,7%) e BMI: body mass index; mmrc: modified medical research council questionnaire; FEV 1 : forced expiratory volume in the 1st second; 6MWD: 6 min walking distance; BODE index according to Celli et al. 19 hospitalizations because of ECOPD in the previous year was , (range 2e8 hospitalizations). The mean FEV 1 % was and the mean BODE index was and its quartile distribution revealed that 33 individuals (31%) were only mildly impaired (Q1 or Q2). Pattern of airway infection during ECOPD Table 2 shows the number (and percentage) of patients with positive sputum culture during the ECOPD episodes that occurred during follow-up, as well as the number (and percentage) of isolated pathogens during such episodes. Eighty of the 116 patients (69%) presented at least one positive sputum culture during follow-up, whereas this was negative in 20 patients (17%) and could not be obtained in 16 patients (14%). Often, patients presented different isolates in different cultures and/or more than one pathogen in the same culture (58 sputum cultures [18,2%] were polymicrobial) (more information in the Online Supplement, Figure E1). Overall, the total number of positive sputum cultures was 318, in which 377 pathogens of 15 different species were isolated. Non-usual pathogens were isolated about three times (71.1%) more often than usual ones. P. aeruginosa was the non-usual pathogen most often isolated (27.3%); it was identified in 46 of the 116 patients studied (39.7%). When we analyze the association between functional status and microbiological results, we observed that FEV 1 % was not associated with microbiological group (data not shown). We recorded in all participants the previous use of home long-term oxygen therapy (LTOT) and non-invasive mechanical ventilation: 48.7% of our patients were using LTOT. Within this group non-usual germs were isolated in 52.5% during follow-up, while in the non-ltot group they were similarly isolated in 55.2% (p Z 0.767). Only 1.7% of included patients needed continuous positive airways pressure (CPAP) and 3.4% bi-level positive airways pressure (BiPAP). However using the multidimensional BODE index we observed that the percentage of patients in whom non-usual pathogens was recovered increased with BODE severity (p Z 0.021) (Fig. 2). In fact, in patients with the highest BODE index values (Q3 and Q4), non-usual pathogens were isolated, alone or in combination with other germs, in 96.2% and 91.7%, respectively, of the positive sputum cultures obtained. Hospitalizations and mortality Because our patients were included in a specific healthcare program, the mean number of hospitalizations due to ECOPD fell significantly from 3.2 in the year before entering in the program to 1.4, and 1.3 at 12 and 24 months of follow-up (p < 0.001) (Online supplement, Figure E2). Table 3 compares several clinical and functional baseline data and microbiological results in patients with or without 2 or more hospitalizations per year due to ECOPD during followup. The distribution of bacterial isolates in sputum cultures was significantly different in both groups (p Z 0.003). In fact, sputum culture was sterile in 70.3% of those patients requiring less than 2 hospitalizations during follow-up, compared to 23.1% in the other group. Age was significantly lower in those with two or more hospitalizations ( vs yrs; p Z 0.032). The remaining Table 2 Number (and %) of patients with positive culture results, and number (and %) of isolates identified during follow-up. Patients f N Z 80 Pathogen isolates N Z 377 Usual pathogen (28.9%) H. influenzae 30 (37.5%) 39 (10.3%) S. pneumoniae 29 (36.3%) 42 (11.1%) M. catarrhalis 24 (30%) 28 (7.4%) Non-usual pathogen (71.1%) P. aeruginosa 46 (57.5%) 103 (27.3%) Other non-fermentative 22 (27.5%) 49 (13%) gram-negative rods g Enterobacteriaceae h 21 (26.3%) 32 (8.5%) Corynebacterium spp. 20 (25.0%) 34 (9%) MRSA i 15 (18.8%) 41 (10.9%) Aspergillus spp. 8 (10.0%) 9 (2.4%) f Figures do not add up to total because patients may have more than one pathogen isolated. g Includes Stenotrophomonas maltophilia and Achromobacter xylsoxidans. h Includes Klebsiella pneumoniae, Escherichia coli, Proteus mirabilis, Enterobacter cloacae, Serratia marcescens and Morganella morganii. i Meticillin-resistant Staphylococcus aureus.

5 844 F. Renom et al. 100% 90% 80% 12,50% 12,50% 28% 48,70% 25,70% 70% Percentage of patients 60% 50% 40% 30% 37,50% 24% 24% 17,90% 2,60% 37,20% 5,70% Both Non-usual pathogen Usual Pathogen No pathogen p= % 10% 37,50% 24% 30,80% 31,40% 0% Q1 Q2 Q3 Q4 N=8 N=25 N=39 N=35 BODE quartiles Percentage of patients with no infectious ECOPD and negative sputum culture (no pathogen), positive sputum culture for usual, non-usual, and both pathogens during follow-up, according to the definitions shown in text, in relation to the baseline BODE quartile. Figure 2 Distribution of microbiology by patient BODE quartiles. clinical and lung function variables shown in Table 3 were not different between groups. Sixty-four percent of patients were alive 24 months after their inclusion in the program. Table 4 presents the association between clinical, functional and microbiological variables and mortality at 24 months. The bivariate analysis Table 3 Comparison of several clinical, functional and microbiological variables in patients with or without frequent hospitalizations during follow-up. <2 hosp/year N Z 37 2 hosp/year N Z 13 p value Age, yr BMI, kg/m Current smoker 1 (2.1%) 3 (23.1%) Dyspnea, MMRC Charlson index FEV 1, % ref MWD, meters BODE index Sputum culture No pathogen 26 (70.3%) 3 (23.1%) Usual pathogen 4 (10.8%) 2 (15.4%) Non-usual 7 (18.9%) 5 (38.5%) pathogen Both 0 (0) 3 (23.1%) Results are shown as mean (SD), or percentage. Clinical and functional data were determined at baseline. Microbiology results correspond to those obtained during the first 6 months of follow-up and the number of hospitalizations was recorded during the next 12 months (only 50 patients had complete data for all variables at 18 month of follow-up). showed that BMI, 6MWT, BODE quartiles and time-dependent microbiological results were all significantly associated with mortality. In the multivariate analysis (Cox regression), BODE quartiles and time-dependent microbiological results remained independently associated with mortality (see adjusted Hazard Ratio (HR) in Table 4). Fig. 3 presents the crude (panel A), and adjusted for baseline BODE quartiles (panel B), survival curves according to microbiological results determined during the first sixmonths of follow-up. In both instances, the isolation of nonusual pathogens, alone or in combination with other germs, was associated with a statistically significant poorer prognosis. Discussion This study investigates a particularly severe group of COPD patients, those requiring frequent hospitalizations because of COPD exacerbations, whose clinical, functional and microbiological characteristics are poorly understood. Main results show that: (1) not all of them had severe disease; (2) sputum culture during ECOPD often identifies a wide range of non-usual pathogens; and, (3) the latter are associated with poorer prognosis. Previous studies Several previous studies have identified a number of factors associated with the need of hospitalization in COPD. These include advanced age, a history of three or more hospitalizations in the previous year, the presence of poor quality of life, severe dyspnea, depression and comorbidity, poor lung function (low FEV 1, low PaO 2, high PaCO2), presence of

6 Role of airway infection 845 Table 4 Crude and adjusted associations of clinical, functional and microbiologic variables and 24-month mortality Unadjusted HR (95% CI) Adjusted HR (95% CI) Age, yr (0.98e1.05) BMI, kg/m (0.86e0.99) Current smoker 0.81 (0.25e2.64) Dyspnea, mmrc 1.57 (0.96e2.5) Charlson index 0.98 (0.72e1.35) FEV 1, % ref 0.99 (0.96e1.01) 6MWD (change of 35 m) 0.82 (0.68e0.93) BODE quartiles (ref. category: quartile 1 þ 2) Quartile (0.37e3.60) Quartile (1.55e11.27) Sputum culture time dependent (ref. category: no pathogen) Usual pathogen 0.40 (0.05e3.06) Non-usual pathogen 3.88 (1.82e8.20) Both 3.05 (1.15e8.06) 1.02 (0.98e1.06) e NS e NS e e 0.78 (0.24e2.52) 2.83 (1.01e7.94) 0.62 (0.08e4.86) 3.36 (1.46e7.72) 3.82 (1.27e11.48) pulmonary arterial hypertension and under prescription of long-term oxygen therapy. 32e35 Only a few authors, however, have investigated specifically the population of COPD patients studied here, that is those actually requiring frequent hospitalizations (two or more per year). Decramer et al. 36 reported that this particular subgroup of patients showed impaired ventilatory and peripheral muscle function, a physiological aspect that was not explored in our study. In a retrospective study, Soler et al. 8 reported that patients hospitalized and/or attended at the emergency room twice or more during the previous year were older and had more severe lung function impairment. Finally, Cao et al. 37, conducted a cross-sectional survey of 186 patients with moderate to severe COPD and a history of two or more hospitalizations due to ECOPD during a one year period and they found that the severity of airflow limitation, the presence of psychosocial distress and the use of vaccination were all associated with frequent hospitalizations. None of these studies, however, analyzed the association of microbiological results and hospitalization requirements. Interpretation of findings Our microbiological findings in this particular subgroup of COPD patients are particularly relevant. On the one hand, they are markedly at variance with previous studies in ECOPD, where sputum culture is positive in about 50% of the cases for H. influenzae (20e30%), S. pneumoniae (10e15%) or M. catarrhalis (10e15%.) 6,7,25e29,38e41 By contrast, a much higher proportion of our patients (69%) showed a positive sputum culture during ECOPD at some point during follow-up and, importantly, the proportion of non-usual pathogens recovered (Table 2) was much higher (71%). This is particularly evident for P. aeruginosa which is normally recovered in 5e10% of ECOPD, 7 whereas it was isolated in 27.3% of our patients. On the other hand, the high yield of isolations of other non-fermentative gram-negative rods (S. maltophilia and A. xylosoxidans often recovered in patients with bronchiectasis and cystic fibrosis, 42,43 Enterobacteriaceae and S. aureus (whose role in ECOPD is not well established), 7 C. striatum (recently identified as a potentially pathogen in advanced COPD) 44,45 and some fungus (such as Aspergillus spp.) is also remarkable. Further, it is worth noting that one third of P. aeruginosa and all the C. striatum isolated were resistant to three or more antibiotic groups, that all the S. aureus isolated were methicillin-resistant, that infections by other non-fermentative gram-negative required specific antibiotics, and that Aspergillus spp. infection required longterm specific anti-fungal therapy. We found that sputum cultures often yielded more than one pathogen (18.2%), that the specific type of pathogen isolated often changed over time in the same patient, and that the BODE index (Fig. 2) were associated with the recovery of non-usual pathogens in the sputum culture. Finally, a unique characteristic of our study is that we followed the patients prospectively for up to 51 months. By doing so, several observations of interest arose. First, we found that airway sterility was associated with a significant reduction in the number of hospital admissions (Table 3), supporting a role for airway infection as a risk factor for

7 846 F. Renom et al. Cum Survival Cum Survival 1,0 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0,0 1,0 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 Panel A. Unadjusted survival curves Usual No pathogen Both Non - usual p< month Panel B. Adjusted for BODE quart iles survival curves Usual No pathogen Non - usual Both p< 0.05 females. Second, sputum cultures were obtained only during those ECOPD episodes considered of likely infectious origin, according to the criteria proposed by Anthonisen. 20 Thus, our study should be complemented by others investigating in these patients the role of airway colonization during clinical stability and other types of ECOPD episodes. Third, bronchiectasis were not specifically searched by high-resolution computerized tomography, but recent studies that include this tool, 47,48 found bronchiectasis as a frequent complication in patients diagnosed of moderate and severe COPD. Finally, this is not a pure observational study because our patients were included in a integral and multidisciplinary health-care program that reduce significantly the number of hospitalizations due to ECOPD. Furthermore, our group is focused in identification and treatment of pathogens involved in airway infection. This may limit the generalizability of our results. Yet, we believe that the fact that we have been able to identify several prognostic factors in these patients despite the potential beneficial effect of the program, actually strengthens the validity of our observations. We acknowledge that reporting additional information such as arterial blood gases either at inclusion or at time of exacerbations during follow-up, frequency of emergency visits during the study period, and disease duration, could have added relevant information about the characterization and prognostic factors related to these patients, but regrettably they were not collected systematically. In a population of COPD patients who require frequent hospitalizations airway infection by non-usual pathogens (specially P. aeruginosa) appears to be a key driver of mortality and hospitalization needs. Given the importance of these two clinical outcomes, we propose that screening and management of airway infection by non-usual pathogens should be considered in COPD patients who require frequent hospitalizations. 0,0 Conflict of interest Month None declared. Figure 3 KeM survival curves of patients with negative results, usual, non-usual, and both pathogens obtained during the first 6 months of follow-up. frequent hospitalizations. Second, as reported previously, 2 we found that mortality at two years was substantial (36%). Interestingly, however, we also found that it was significantly associated with the recovery of non-usual pathogens in the sputum culture, particularly P. aeruginosa (Table 4 and Fig. 3). This observation is in keeping with previous observations in patients with cystic fibrosis 46 but contrasts with a study by Murphy et al. 47 in COPD patients. In any case, it suggests that the therapeutic strategy in COPD patients with frequent hospitalizations, such as those studied here, has to consider, and eventually target, non-usual pathogens. Potential limitations Our study has some limitations that deserve comment. First, we studied mostly males. This is quite often the case in Spain, where females started to smoke much later. Our results, therefore, cannot be extrapolated readily to Acknowledgements Authors thank Maria Moranta (Fundació Caubet-Cimera) for help with data processing. CIBERES is an initiative of Instituto Salud Carlos III (Ministerio de Ciencia e Innovación). Supplementary information Supplementary data associated with this article can be found in the online version at doi: /j.rmed References 1. Connors Jr AF, Dawson NV, Thomas C, et al. Outcomes following acute exacerbation of severe chronic obstructive lung disease. The SUPPORT investigators (Study to Understand

8 Role of airway infection 847 Prognoses and Preferences for Outcomes and Risks of Treatments). Am J Respir Crit Care Med 1996;154:959e Soler-Cataluna JJ, Martinez-Garcia MA, Roman SP, et al. Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary disease. Thorax 2005;60:925e Miravitlles M, Murio C, Guerrero T, et al. Costs of chronic bronchitis and COPD: a 1-year follow-up study. Chest 2003; 123:784e Strassels SA, Smith DH, Sullivan SD, et al. The costs of treating COPD in the United States. Chest 2001;119:344e Sullivan SD, Ramsey SD, Lee TA. The economic burden of COPD. Chest 2000;117:5Se9S. 6. Papi A, Bellettato CM, Braccioni F, et al. Infections and airway inflammation in chronic obstructive pulmonary disease severe exacerbations. Am J Respir Crit Care Med 2006;173:1114e Sethi S, Murphy TF. Infection in the Pathogenesis and Course of Chronic Obstructive Pulmonary Disease. N Engl J Med 2008; 359:2355e Soler J, Sanchez L, Latorre M, et al. [The impact of COPD on hospital resources: the specific burden of COPD patients with high rates of hospitalization]. Arch Bronconeumol 2001;37: 375e Clini E, Vitacca M, Foglio K, et al. Long-term home care programmes may reduce hospital admissions in COPD with chronic hypercapnia. Eur Respir J 1996;9:1605e Guell R, Gonzalez A, Morante F, et al. [Better at home: a continuous health care program for patients with advanced chronic respiratory disease]. Arch Bronconeumol 1998;34: 541e Casas A, Troosters T, Garcia-Aymerich J, et al. Integrated care prevents hospitalisations for exacerbations in COPD patients. Eur Respir J 2006;28:123e Rea H, McAuley S, Stewart A, et al. A chronic disease management programme can reduce days in hospital for patients with chronic obstructive pulmonary disease. Intern Med J 2004;34:608e Rabe KF, Hurd S, Anzueto A, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007;176:532e Mahler DA, Wells CK. Evaluation of clinical methods for rating dyspnea. Chest 1988;93:580e Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373e Standardization of Spirometry, 1994 Update. American Thoracic Society. Am J Respir. Crit Care Med 1995;152:1107e Roca J, Sanchis J, Agusti-Vidal A, et al. Spirometric reference values from a Mediterranean population. Bull Eur Physiopathol Respir 1986;22:217e ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111e Celli BR, Cote CG, Marin JM, et al. The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med 2004;350: 1005e Anthonisen NR, Manfreda J, Warren CP, et al. Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 1987;106:196e Murray PR, Washington JA. Microscopic and bacteriologic analysis of expectorated sputum. Mayo Clin Proc 1975;50: 339e Murray PR, Baron EJ, Jorgensen JH, et al. Manual of clinical microbiology. 8th ed. Washington, DC: American Society for Microbiology; Barrow GI, Feltham RK. Cowan and Steel s manual for the identification of medical bacteria. 3rd ed. Cambridge: Cambridge University Press; Wayne PA. Methods for dilution antimicrobial susceptibility test for bacteria that grow aerobically. NCCLS approved standard M7-A6. NCCLS; Eller J, Ede A, Schaberg T, et al. Infective exacerbations of chronic bronchitis: relation between bacteriologic etiology and lung function. Chest 1998;113:1542e Groenewegen KH, Wouters EF. Bacterial infections in patients requiring admission for an acute exacerbation of COPD; a 1- year prospective study. Respir Med 2003;97:770e Miravitlles M. Exacerbations of chronic obstructive pulmonary disease: when are bacteria important? Eur Respir J Suppl 2002; 36:9se19s. 28. Rosell A, Monso E, Soler N, et al. Microbiologic determinants of exacerbation in chronic obstructive pulmonary disease. Arch Intern Med 2005;165:891e Soler N, Torres A, Ewig S, et al. Bronchial microbial patterns in severe exacerbations of chronic obstructive pulmonary disease (COPD) requiring mechanical ventilation. Am J Respir Crit Care Med 1998;157:1498e Hosmer DW, Lemeshow S. Extensions of the proportional hazards model. Applied Survival analysis: regression modeling of time to event data. New York: Jonh Wiley and Sons; Hosmer DW, Lemeshow S. Extensions of the proportional hazards model: Stata Texbook Examples. Date last updated; 2009 [accessed: ]. 32. Almagro P, Barreiro B, Ochoa de EA, et al. Risk factors for hospital readmission in patients with chronic obstructive pulmonary disease. Respiration 2006;73:311e Garcia-Aymerich J, Monso E, Marrades RM, et al. Risk factors for hospitalization for a chronic obstructive pulmonary disease exacerbation. EFRAM study. Am J Respir Crit Care Med 2001; 164:1002e Kessler R, Faller M, Fourgaut G, et al. Predictive factors of hospitalization for acute exacerbation in a series of 64 patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999;159:158e Miravitlles M, Guerrero T, Mayordomo C, et al. 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;67:495e Decramer M, Gosselink R, Troosters T, et al. Muscle weakness is related to utilization of health care resources in COPD patients. Eur Respir J 1997;10:417e Cao Z, Ong KC, Eng P, et al. Frequent hospital readmissions for acute exacerbation of COPD and their associated factors. Respirology 2006;11:188e Lin SH, Kuo PH, Hsueh PR, et al. Sputum bacteriology in hospitalized patients with acute exacerbation of chronic obstructive pulmonary disease in Taiwan with an emphasis on Klebsiella pneumoniae and Pseudomonas aeruginosa. Respirology 2007;12:81e Miravitlles M, Espinosa C, Fernandez-Laso E, et al. Relationship between bacterial flora in sputum and functional impairment in patients with acute exacerbations of COPD. Study Group of Bacterial Infection in COPD. Chest 1999;116:40e Nseir S, Di PC, Cavestri B, Jozefowicz E, et al. Multiple-drugresistant bacteria in patients with severe acute exacerbation of chronic obstructive pulmonary disease: prevalence, risk factors, and outcome. Crit Care Med 2006;34:2959e Patel IS, Seemungal TA, Wilks M, et al. Relationship between bacterial colonisation and the frequency, character, and severity of COPD exacerbations. Thorax 2002;57:759e Canton R, Cobos N, de Gracia J, et al. Antimicrobial therapy for pulmonary pathogenic colonisation and infection by Pseudomonas aeruginosa in cystic fibrosis patients. Clin Microbiol Infect 2005;11:690e703.

9 848 F. Renom et al. 43. Vendrell M. Diagnóstico y tratamiento de las bronquiectasias. Recomendaciones SEPAR. 51. Elsevier España SL; Otsuka Y, Ohkusu K, Kawamura Y, et al. Emergence of multidrug-resistant Corynebacterium striatum as a nosocomial pathogen in long-term hospitalized patients with underlying diseases. Diagn Microbiol Infect Dis 2006;54:109e Renom F, Garau M, Rubi M, et al. Nosocomial outbreak of Corynebacterium striatum infection in patients with chronic obstructive pulmonary disease. J Clin Microbiol 2007;45: 2064e Emerson J, Rosenfeld M, McNamara S, et al. Pseudomonas aeruginosa and other predictors of mortality and morbidity in young children with cystic fibrosis. Pediatr Pulmonol 2002;34: 91e Murphy TF, Brauer AL, Eschberger K, et al. Pseudomonas aeruginosa in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2008;177:853e Patel IS, Vlahos I, Wilkinson TM, et al. Bronchiectasis, exacerbation indices, and inflammation in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2004;170:400e7.