Chronic Pseudomonas aeruginosa Infection in Chronic Obstructive Pulmonary Disease

Transcription

1 MAJOR ARTICLE Chronic Pseudomonas aeruginosa Infection in Chronic Obstructive Pulmonary Disease Laura Martínez-Solano, 1,3 María D. Macia, 2 Alicia Fajardo, 1,3 Antonio Oliver, 2 and Jose L. Martinez 1,3 1 Departamento de Biotecnología Microbiana, Centro Nacional de Biotecnología (Consejo Superior de Investigaciones Cientificas), Cantoblanco, Madrid, 2 Servicio de Microbiologia and Unidad de Investigación, Hospital Son Dureta, Instituto Universitario de Investigación en Ciencias de la Salud, Palma de Mallorca, and 3 CIBER Epidemiología y Salud Pública, Spain (See the article by Murphy on pages ) Background. Pseudomonas aeruginosa infections are increasingly associated with acute exacerbations in chronic obstructive pulmonary disease (COPD). We aimed to determine whether an underlying chronic infection might be behind this process and to determine the epidemiological characteristics of the isolates involved, to implement useful protocols for preventing and treating these infections. Methods. P. aeruginosa isolates obtained from respiratory samples of 13 patients with COPD and from blood samples of 10 patients in intensive care units were investigated. In 8 patients with COPD, isolates were obtained during sequential exacerbation episodes. Five patients presented a single infection episode. Production of virulence determinants and genetic relationships were analyzed in all isolates. Results. Patients with COPD were usually infected with 1 P. aeruginosa clone that remained in the lung for years, without evidence of interpatient transmission. During chronic infection, each clone diversified, which led to the coexistence of isolates with different morphotypes and antibiotic susceptibility. Overall, P. aeruginosa evolved toward an increased mutation rate, increased antibiotic resistance, and reduced production of proteases. Isolates from samples of infected lungs tend to be less cytotoxic and motile and to produce more biofilm, compared with isolates from blood samples. Conclusion. These results provide the first evidence supporting the hypothesis that P. aeruginosa causes chronic infections in COPD, with patterns of infection and evolution that resemble those observed in cystic fibrosis. Experience gained from treating cystic fibrosis might be useful for implementing new procedures for the prevention, diagnosis, and treatment of infection due to P. aeruginosa in COPD. Patients with chronic obstructive pulmonary disease (COPD) often present with acute exacerbations (AE- COPDs) that have a major impact on the patients quality of life. These exacerbations constitute the main cause of mortality among patients affected by this disease [1]. Although the causes of AECOPD are not yet well understood, bacterial infections are thought to be involved in approximately one-half of cases [2, 3]. In-depth studies of the microbiology of the respiratory tracts of patients with COPD are, nevertheless, needed to clearly ascertain the role of infection in the prognosis of the Received 5 June 2008; accepted 8 August 2008; electronically published 6 November Reprints or correspondence: Dr. Jose L. Martinez, Departamento de Biotecnología Microbiana, Centro Nacional de Biotecnología, Darwin 3, Campus UAM, Cantoblanco, Madrid, Spain (jlmtnez@cnb.csic.es). Clinical Infectious Diseases 2008; 47: by the Infectious Diseases Society of America. All rights reserved /2008/ $15.00 DOI: / disease. Pseudomonas aeruginosa is beginning to be recognized as a relevant pathogen in COPD that is associated with an intense airway inflammation and poor prognosis for those with the disease [4, 5]. The prevalence of P. aeruginosa infection in AECOPD is estimated to be 4% but increases to as much as 13% among patients with advanced airway obstruction [6]. The implementation of adequate protocols to prevent and treat P. aeruginosa infection might, therefore, help to alleviate the mortality and morbidity associated with COPD. To reach that goal, more information is required about the population structure and the pathophysiological characteristics of P. aeruginosa that cause infection in the context of COPD. In this sense, we still have very little information on whether AECOPDs are mainly truly acute P. aeruginosa infections in a COPD background or whether, on the contrary, they reflect an acute exacerbation of a chronic infection process. The high prevalence of hypermutable strains recently 1526 CID 2008:47 (15 December) Martínez-Solano et al.

2 found in COPD, which is a typical marker of chronic infections but not of acute processes [7 9], would argue in favor of the latter possibility. P. aeruginosa has long been recognized to have a paramount relevance in the development, prognosis, and outcome of cystic fibrosis (CF), another severe disease that results in a chronic deterioration of pulmonary function. Although the causes of both diseases are completely different, it has been established that some symptoms, such as mucus hypersecretion or decreased mucus clearance, that contribute to morbidity are common to both diseases, and it has been suggested that the information obtained from the treatment of patients with CF may help to identify effective therapies and diagnostic tests for the management of COPD [10]. The usefulness of antibiotics during AECOPD remains controversial, mainly because only 40% 50% of exacerbations can be attributed to bacteria [2]. In the case of documented P. aeruginosa infection, the information derived from decades of treatment of patients with CF may be valuable for the development of useful therapies for the management of COPD. Although it has been discussed that AECOPD can be attributable to bacterial pathogens only when new strains are isolated during exacerbations [3], it should be noted that patients with CF experience acute exacerbations even though they are chronically infected [11], and the risks involved with these exacerbations are decreased with prophylactic antibiotic therapy [12]. Similarly, a link between chronic infection and AECOPD might be possible, and evaluation of this aspect of COPD is relevant for establishing novel therapeutic strategies. In the present study, we analyzed the population dynamics and quantified the production of virulence determinants in P. aeruginosa isolates from respiratory samples from patients with COPD and from blood samples from patients in the intensive care unit, with the aim of defining the pathophysiological characteristics of P. aeruginosa strains infecting patients with COPD, in comparison with isolates that produce acute infections. To our knowledge, our results provide the first indication that patients with COPD may experience chronic P. aeruginosa infections that remain between AECOPD episodes. The epidemiological and pathophysiological characteristics of the bacteria that produce these infections are very similar to those already found in patients with CF. Thus, diagnostic procedures, antibiotic therapies, and containment protocols successfully implemented for patients with CF might be useful for patients with COPD who have P. aeruginosa infection. PATIENTS, MATERIALS, AND METHODS Bacterial strains, growth conditions, and antibiotic susceptibility testing. The bacterial strains used in this study were obtained from the Hospital Son Dureta, Palma de Mallorca, Spain (table 1). The studied collection included strains isolated from samples from 13 patients with COPD who were admitted to the hospital with P. aeruginosa AECOPD. Eight of the patients with COPD presented documented sequential episodes of P. aeruginosa respiratory infections (strains named COPD ), and 5 presented a first episode of infection due to this microorganism (strains named L ). The quality of the respiratory samples was assessed by Gram staining before processing for culture, according to established recommendations [13]. All patients for whom the first P. aeruginosa isolate was included had at least 1 sputum culture negative for P. aeruginosa during the previous year. In addition, 10 strains isolated by culture of blood samples from 10 patients in the intensive care unit (strains named H ) were included for comparative purposes. We chose blood culture isolates because bacteremia (independent of the primary focus) is the best representative of an acute infection process. The primary focus of bacteremia in these patients included pneumonia (4 patients), wound infection (2), urinary tract infection (2), catheter-related infection (1), and infection of unknown origin (1). None of the patients had a primary hematological disease. P. aeruginosa strains were routinely grown in Luria-Bertani medium (Pronadisa). Motility assays (to assess swimming, swarming, and twitching) were performed as described elsewhere [14]. MICs were determined in Müller-Hinton (Pronadisa) agar plates with use of Etest strips (AB Biodisk), according to the manufacturers recommendations. Rifampin (300 mg/ml) resistance mutation frequencies were determined as described elsewhere [8]. Molecular epidemiology. The epidemiological relatedness of the strains was studied by PFGE [15]. Bacterial DNA embedded in agarose plugs, prepared as described elsewhere [7], was digested with use of the restriction enzyme SpeI. DNA separation was performed in a CHEF-DRIII apparatus (Bio- Rad) with 6 V/cm 2 for 26 h, with pulse times of 5 40 s. DNA macrorestriction patterns were interpreted in accordance with the criteria established by Tenover et al. [16]. Quantitative analysis of the production of virulence factors. For the biofilm assays, bacteria were grown in Falcon 3911 MicroTest III silicone flexible assay plates (Becton Dickinson). Biofilm formation was quantified, as described elsewhere [14], by staining the biofilms with Crystal Violet and determining their absorbance at 570 nm. Production of pyoverdine and pyocyanin was quantified as described elsewhere [17]. Bacterial strains were grown at 37 C in liquid broth containing 2% peptone, 1% potassium sulphate, and 0.14% magnesium chloride for 40 h for pyocyanin quantification or in Luria-Bertani broth for 24 h for pyoverdine quantification. Bacteria were pelleted by centrifugation. The amount of pyocyanin was evaluated by measuring the absorbance of the supernatants at 690 nm. The amount of pyoverdine was measured by fluorescence by exciting the supernatants at 400 nm and measuring the emission at 460 nm. Production of proteases was quantified as described Chronic P. aruginosa Infection in COPD CID 2008:47 (15 December) 1527

3 Table 1. Characteristics of clinical Pseudomonas aeruginosa isolates or strains. P. aeruginosa isolate a Isolation date PFGE genotype Date of first documented isolate Morphotype MIC, mg/ml Caz Cip Imp Mer Tob Col Mutation rate COPD1a 17 March 2003 A 1998 Mucoid Hypermutable COPD1b 19 November 2003 A 1998 Scv Hypermutable COPD1c 28 December 2004 A 1998 Mucoid Hypermutable COPD1d 30 March 2005 A 1998 Mucoid Hypermutable COPD2a 2 January 2004 B 2 January 2004 Normal Not hypermutable COPD2b 22 December 2004 C 2 January 2004 Normal Not hypermutable COPD2c 29 December 2004 C 2 January 2004 Melanin Hp Hypermutable COPD2d 8 May 2005 C 2 January 2004 Melanin Hp Hypermutable COPD3a 10 May 2003 D 1999 Scv Hypermutable COPD3b 2 January 2004 D 1999 Mucoid Hypermutable COPD3c 4 February 2004 D 1999 Mucoid Hypermutable COPD3d 19 June 2004 D 1999 SCV Hypermutable COPD4a 15 November 2004 E 15 November 2004 Normal Not hypermutable COPD4b 8 September 2005 E 15 November 2004 Normal Not hypermutable COPD5a 5 August 2005 F 5 August 2005 Metallic Not hypermutable COPD5b 5 August 2005 F 5 August 2005 Metallic Hypermutable COPD5c 29 September 05 F 5 August 2005 Normal Hypermutable COPD5d 1 February 06 F 5 August 2005 Normal Not hypermutable COPD5e 25 June 2007 F 5 August 2005 SCV Not hypermutable COPD6a 13 November 2003 G 13 November 2003 Metallic Not hypermutable COPD6b 14 June 2006 H 13 November 2003 Normal Not hypermutable COPD6c 10 November 2006 H 13 November 2003 Mucoid Not hypermutable COPD6d 25 May 2007 H 13 November 2003 Mucoid Not hypermutable COPD7a 25 November 2004 I 25 November 2004 Mucoid Not hypermutable COPD7b 27 March 2007 J 25 November 2004 Normal Not hypermutable COPD7c 8 July 2007 K 25 November 2004 Mucoid Not hypermutable COPD7d 8 July 2007 K 25 November 2004 SCV Not hypermutable COPD8a 25 July 2005 L 7 March 2005 Normal Not hypermutable COPD8b 29 December 2005 L 7 March 2005 SCV Hypermutable L1 14 November 2003 M 14 November 2003 Normal Not hypermutable L2 31 January 2004 N 31 January 2004 Normal Not hypermutable L3 9 January 2004 O 9 January 2004 Normal Not hypermutable L4 15 March 2004 P 15 March 2004 Normal Not hypermutable L5 8 July 2003 Q 8 July 2003 Normal Not hypermutable H1 4 November 2002 R 4 November 2002 Normal Not hypermutable H2 2 August 2003 S 2 August 2003 Normal Not hypermutable H3 22 March 2003 T 22 March 2003 Normal Not hypermutable H4 28 February 2003 U 28 February 2003 Normal Not hypermutable H5 10 April 2004 V 10 April 2004 Normal Not hypermutable H6 15 September 2004 W 15 September 2004 Normal Not hypermutable H7 3 April 2005 X 3 April 2005 Normal Not hypermutable H8 10 October 2005 Y 10 October 2005 Normal Not hypermutable H9 18 January 2006 Z 18 January 2006 Normal Not hypermutable H10 16 October 2006 Aa 16 October 2006 Normal Not hypermutable NOTE. Caz, ceftazidime; Cip, ciprofloxacin; Col, colistin; COPD, sequential isolates from patients with chronic obstructive pulmonary disease; H, isolates from blood culture; Imp, imipenem; L, P. aeruginosa strains involved in nonchronic lung infection; Melanin Hp, pyomelanin hyperproducer; Mer, meropenem; SCV, small colony variant; Tob, tobramycin. a Patients are differentiated by numbers. Different strains isolated from the same patient are specified with a letter. elsewhere [17]. Bacteria were grown overnight in Luria-Bertani broth at 37 C, were pelleted by centrifugation, and were tested for proteolytic activity in the culture supernatants. Caseinase activity was tested using azocasein as the substrate [17]. Elastolytic activity was determined using elastin Congo Red as the substrate [17]. Cytotoxicity against either the macrophage cell line J774 or the A549 bronchial epithelial cell line was quantified [18] with the Cytotoxicity Detection Kit (LDH; Roche) in accordance with the manufacturer s instructions. One hundred percent cytotoxicity was estimated by lysing noninfected cells with 2% (vol/vol) Triton X-100 (Sigma-Aldrich). To compare the production of virulence factors among all the strains, we divided the obtained values for each isolate by the mean value for all isolates. The log 2 of the obtained values was used for a hierarchically clustering analysis with use of software available at the Eisen Lab Web site [19] CID 2008:47 (15 December) Martínez-Solano et al.

4 RESULTS Molecular epidemiology, morphotype, and mutation rate. Serial P. aeruginosa isolates were recovered from 8 patients with COPD who experienced AECOPD. As shown in table 1, colonization of a single patient by different clones was not a frequent event, and even in such cases (patients COPD2, COPD6, and COPD7), once a clone was established, it remained in the lung of the patient for a long time. The clones were specific for each patient, which showed that, at least in this cohort, there was not interpatient transmission. P. aeruginosa that cause chronic CF infection diversify during in-host evolution, with different isolates from the same patient presenting different morphotypes and antibiotic susceptibilities [20]. As shown in table 1, our results indicate that the same situation can be observed in patients with COPD, including the isolation of small colony variants and mucoid morphotypes, which are typical of chronic infection associated with CF [21]. These morphotypes persisted in the patient, at least for the duration of our study (table 1). P. aeruginosa hypermutable strains are isolated from a high percentage of patients with CF who have chronic infections, in contrast to what is observed in acute infections [8, 9]. We observed a similar situation in patients with COPD (table 1), in which isolates from only chronic colonization exhibited hypermutable phenotypes. Antibiotic resistance. As a general trend, evolution toward lower levels of susceptibility to antibiotics was observed in all patients, although some fluctuations in these levels were eventually seen (compare isolates COPD1a and COPD1c in table 1). The coexistence of 2 isolates from the same P. aeruginosa clone but with presentation of different patterns of antibiotic susceptibility was observed for patient COPD5 (isolates COPD5a and COPD5b) and for patient COPD7 (isolates COPD7c and COPD7d). The highest MIC value for each antibiotic was always observed in isolates obtained at late stages of chronic colonization of patients with COPD. Virulence factors. Production of virulence factors was quantified as described in the Patients, Materials, and Methods section. A large variability in the level of production was observed in all types of isolates, irrespective of their origin. Nevertheless, most isolates from blood culture clustered independent from those from chronic COPD. Most isolates from nonchronic lung infection clustered close to some COPD isolates and independently, as well from isolates from blood culture (figure 1). The most evident trend was that P. aeruginosa strains chronically infecting patients with COPD tend to produce lower levels of caseinase, elastase, and pyocyanin, whereas pyoverdine production is more evenly distributed. COPD strains produced higher amounts of biofilm, and their cytotoxicity was impaired, Figure 1. Quantitative analysis of the expression of phenotypes relevant to the virulence of Pseudomonas aeruginosa in clinical isolates from respiratory samples of patients with chronic obstructive pulmonary disease (COPD) and blood samples from patients in the intensive care unit. The heat map of the clustering of the different isolates as the function of the production of the virulence determinants is shown. Green indicates that the production is lower than the mean; red indicates that the production is higher than the mean. Epith, epithelial; H, isolates from blood culture; L, P. aeruginosa strains involved in nonchronic lung infection; macro, macrophage. when compared with strains from blood cultures. The lower level of expression of virulence factors is less clear in the case of strains isolated from the lungs of patients without evidence of chronic infection (e.g., isolates L1, L2, and L5 in table 1). These strains are the first to be isolated from samples from these patients, and their phenotype, intermediate between COPD and H strains, likely reflects the pathophysiological char- Chronic P. aruginosa Infection in COPD CID 2008:47 (15 December) 1529

5 acteristics of P. aeruginosa in the earlier stages of evolution during COPD infection. Motility. Strains from chronic colonization exhibited overall decreased swimming, swarming, and twitching motility than did strains involved in nonchronic pulmonary infection and strains from patients in the intensive care unit (figure 2). As a general trend, swimming and twitching motility decreased over time during the evolution of chronic infection. It has been recently reported that swarming ability is associated with increased production of virulence factors [22]. In agreement with this statement, we found that strains showing a strong swarming ability also produced virulence factors at high levels. DISCUSSION Most patients with CF acquire chronic P. aeruginosa infections early in life; these infections afflict patients for decades and are responsible for much of the morbidity and mortality of people with this disease [21, 23]. Most of these infections are clonal [24], and infections are acquired independently by each patient, presumably from diverse environmental reservoirs [25]. It has been suggested that there are not strong differences in the genetic structure or in the physiology of the different P. aeruginosa strains [15, 26] and that the same factors are used to infect humans, plants, and nematodes [27, 28]. Nevertheless, the level of expression of the different virulence determinants is different for bacteria producing acute (the most clear example is blood infection) or chronic (e.g., CF) infections, and it has been reported that genes associated with acute infection and chronic persistence in P. aeruginosa are reciprocally regulated [29]. The main factors contributing to the pathogenicity of P. aeruginosa [30] are the production of quorum-sensing regulated virulence factors including proteases, pyoverdine, and pyocyanin and type III secretion linked cytotoxicity. Motility and biofilm formation are also relevant for surface colonization. As a consequence of long-lasting infection in CF-affected lungs, P. aeruginosa follows a characteristic pattern of evolution defined by selection of hypermutable strains [8], increase in antibiotic resistance, and strong diversification [20]. The most commonly Figure 2. Motility of Pseudomonas aeruginosa clinical isolates from respiratory samples of patients with chronic obstructive pulmonary disease (COPD) and blood samples from patients in the intensive care unit. Swarming, swimming, and twitching motilities were measured. For either swimming or twitching motility, all pictures are at the same scale. The motility in the swarming assay had a diameter range of!0.5 to 124 cm. Therefore, only pictures of those strains with motility 12-cm diameter are shown. In all cases, the first column indicates swimming motility, the second indicates twitching, and the third indicates swarming motility. Some patients were infected with 2 clones. The clones that were different are highlighted with a square. *Strains with very high swarming motility (shown at a scale one-half that of other images of swarming). H, isolates from blood culture; L, P. aeruginosa strains involved in nonchronic lung infection; LM, low-level motility (diameter,!2 cm; images not shown); NM, no motility in this assay (diameter,!0.5 cm; images not shown) CID 2008:47 (15 December) Martínez-Solano et al.

6 observed trends include a reduction in the expression of proteases [31] and in the level of cytotoxicity against mammal cells [32] and an increase in biofilm production [33]. Although CF and COPD have completely different causal origins, they share some relevant symptoms, and bacterial colonization may be relevant for the outcome of both pulmonary diseases. The role of bacterial infection in the development of COPD is still under debate, although it has been reported that infection with P. aeruginosa is a relevant element in the mortality and morbidity of patients with AECOPD [34]. One difference between patients with CF and patients with COPD is that infections occur later in life in the latter. Because of this, the observation time for patients with COPD in our study is decades shorter than that for patients with CF. The elements that govern P. aeruginosa infection in patients with COPD in our cohort are similar to those reported in patients with CF. First, usually a single clone infects each patient, and there is no evidence of interpatient transmission of epidemic clones. Second, there is a trend in strains involved in chronic COPD infections toward an increase in mutation rate and antibiotic resistance, whereas the cytotoxicity and the motility, as well as the levels of expression of pyocyanin and proteases, decrease a feature that could promote long-term survival in lungs of patients with CF [35]. The isolates from the lung, where they grow attached to surfaces, tend to produce higher amounts of biofilm than do those found in blood culture samples. Finally, radiative evolution of P. aeruginosa is observed in chronically infected patients with COPD. Although the isolates from the same patient belonged to the same clone, different morphotypes were obtained, as were fluctuations in the susceptibility to antibiotics. For instance, patient COPD1 had 3 clonal mucoid isolates, and in spite of selective pressure because of antibiotic therapy, the second isolate was more susceptible to some antibiotics than was the first one. The most suitable hypothesis to explain this evolution is the coexistence of strains with different phenotypes (that belong to the same clone) in each patient. An example of this situation was observed in patient COPD5, for whom 2 isolates (COPD5a and COPD5b) were obtained on the same day and had different phenotypes of antibiotic susceptibilities, although the isolates belonged to the same clone. Together with the fact that hypermutator P. aeruginosa strains are frequently isolated from patients with COPD, all these data strongly suggest that the infecting process of P. aeruginosa is likely to be similar in CF and COPD. Therefore, we can learn from decades of experience in treating patients with CF to implement better procedures for the bacteriological diagnosis and treatment of infections in patients with COPD. As a general practice, we suggest analysis of the phenotype of 11 strain in patients with P. aeruginosa infection associated with AECOPD. A mixture of strains with different levels of antibiotic susceptibility is possible, and information for only 1 of these isolates may incorrectly qualify the infection as antibiotic susceptible (leading to incorrect antibiotic choices) a scenario similar to that for CF isolates [20]. It is possible that a lack of information about the actual levels of antibiotic susceptibility in the complex bacterial population that may infect patients with COPD might be the cause of failures recorded with empirical antibiotic therapy [36]. The data presented in this article are still preliminary, and more-extensive studies are required to make definitive statements. However, the comparison of the chronic infection process of P. aeruginosa in patients with COPD with the process extensively studied in CF might provide some clues for the management of this process. The pattern of infection due to P. aeruginosa in patients with CF is generally similar to that described here in countries with at-home antibiotic treatment of P. aeruginosa infection. In these conditions, interpatient transmission is rare. However, in other countries with a history of hospital-based antibiotic treatment, it has been reported that a few successful clones are responsible for the majority of infections [37]. In this case, the frequent patient-to-patient contact in hospitals or at holiday camps [38] allows selection of bacterial clones that can establish themselves as evolutive successful (epidemic) lineages in hospitals. Because AECOPDs are usually treated in hospitals, preventive containment measures should be considered, to avoid the possibility of cross-infection by these types of P. aeruginosa strains among patients with COPD. The implementation of protocols currently used for the treatment of patients with CF might also be considered as an alternative in the treatment of COPD. This might include maintenance of antimicrobial therapy that reduces the deterioration progression of lung function of chronically infected patients with CF [12, 39] or aggressive treatment of P. aeruginosa lung colonization before the typical markers of P. aeruginosa chronic infection (e.g., hypermutation) are selected, because once the chronic infection is established, eradication becomes almost impossible [39, 40]. Our study supports the hypothesis that, at least in this patient cohort and within the studied time period, P. aeruginosa may produce chronic infection in patients with COPD. The observed infection process, with each patient infected by 1 bacterial clone that persists for years and with a specific pattern of evolution during infection, resembles the situation with CF. To confirm these findings, more studies are needed, with a larger number of patients and longer observation times, to implement better strategies for prevention and treatment of infections in patients with COPD. Acknowledgments We thank Carolina Alvarez-Ortega for English proofreading. Financial support. Ministerio de Educación y Ciencia (to A.F.), Eu- Chronic P. aruginosa Infection in COPD CID 2008:47 (15 December) 1531

7 ropean Union (LSHM-CT and LSHM-CT to J.L.M.), Ministerio de Educación y Ciencia (BIO to J.L.M.), and Ministerio de Sanidad y Consumo (SAF , REIPI C03/14, and REIPI RD06/008 to A.O.). Potential conflicts of interest. All authors: no conflicts. References 1. Tan WC. Factors associated with outcomes of acute exacerbations of chronic obstructive pulmonary disease. COPD 2004; 1: Puhan MA, Vollenweider D, Latshang T, Steurer J, Steurer-Stey C. Exacerbations of chronic obstructive pulmonary disease: when are antibiotics indicated? A systematic review. Respir Res 2007; 8: Sethi S, Evans N, Grant BJ, Murphy TF. New strains of bacteria and exacerbations of chronic obstructive pulmonary disease. N Engl J Med 2002; 347: Murphy TF, Brauer AL, Eschberger K, et al. Pseudomonas aeruginosa in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2008; 177: Lin SH, Kuo PH, Hsueh PR, Yang PC, Kuo SH. 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: Lieberman D, Lieberman D. Pseudomonal infections in patients with COPD: epidemiology and management. Am J Respir Med 2003;2: Macia MD, Blanquer D, Togores B, Sauleda J, Perez JL, Oliver A. Hypermutation is a key factor in development of multiple-antimicrobial resistance in Pseudomonas aeruginosa strains causing chronic lung infections. Antimicrob Agents Chemother 2005; 49: Oliver A, Canton R, Campo P, Baquero F, Blazquez J. High frequency of hypermutable Pseudomonas aeruginosa in cystic fibrosis lung infection. Science 2000; 288: Gutierrez O, Juan C, Perez JL, Oliver A. Lack of association between hypermutation and antibiotic resistance development in Pseudomonas aeruginosa isolates from intensive care unit patients. Antimicrob Agents Chemother 2004; 48: Henke MO, Shah SA, Rubin BK. The role of airway secretions in COPD clinical applications. COPD 2005; 2: Goss CH, Burns JL. Exacerbations in cystic fibrosis. 1. Epidemiology and pathogenesis. Thorax 2007; 62: Cheer SM, Waugh J, Noble S. Inhaled tobramycin (TOBI): a review of its use in the management of Pseudomonas aeruginosa infections in patients with cystic fibrosis. Drugs 2003; 63: Thomson RB. Specimen collection, transport, and precessing: bacteriology. 9th ed. Washington, DC: ASM Press, Linares JF, Gustafsson I, Baquero F, Martinez JL. Antibiotics as intermicrobial signaling agents instead of weapons. Proc Natl Acad Sci USA2006; 103: Morales G, Wiehlmann L, Gudowius P, et al. Structure of Pseudomonas aeruginosa populations analyzed by single nucleotide polymorphism and pulsed-field gel electrophoresis genotyping. J Bacteriol 2004; 186: Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33: Sanchez P, Linares JF, Ruiz-Diez B, et al. Fitness of in vitro selected Pseudomonas aeruginosa nalb and nfxb multidrug resistant mutants. J Antimicrob Chemother 2002; 50: Linares JF, Lopez JA, Camafeita E, Albar JP, Rojo F, Martinez JL. Overexpression of the multidrug efflux pumps MexCD-OprJ and MexEF-OprN is associated with a reduction of type III secretion in Pseudomonas aeruginosa. J Bacteriol 2005; 187: Eisen Lab. Available at: Accessed 23 October Foweraker JE, Laughton CR, Brown DF, Bilton D. Phenotypic variability of Pseudomonas aeruginosa in sputa from patients with acute infective exacerbation of cystic fibrosis and its impact on the validity of antimicrobial susceptibility testing. J Antimicrob Chemother 2005; 55: Lyczak JB, Cannon CL, Pier GB. Lung infections associated with cystic fibrosis. Clin Microbiol Rev 2002; 15: Overhage J, Bains M, Brazas MD, Hancock RE. Swarming of Pseudomonas aeruginosa is a complex adaptation leading to increased production of virulence factors and antibiotic resistance. J Bacteriol 2008; 190: Govan JR, Deretic V. Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev 1996; 60: Struelens MJ, Schwam V, Deplano A, Baran D. Genome macrorestriction analysis of diversity and variability of Pseudomonas aeruginosa strains infecting cystic fibrosis patients. J Clin Microbiol 1993; 31: Romling U, Wingender J, Muller H, Tummler B. A major Pseudomonas aeruginosa clone common to patients and aquatic habitats. Appl Environ Microbiol 1994; 60: Alonso A, Rojo F, Martinez JL. Environmental and clinical isolates of Pseudomonas aeruginosa show pathogenic and biodegradative properties irrespective of their origin. Environ Microbiol 1999;1: Navas A, Cobas G, Talavera M, Ayala JA, Lopez JA, Martinez JL. Experimental validation of Haldane s hypothesis on the role of infection as an evolutionary force for Metazoans. Proc Natl Acad Sci USA2007; 104: Rahme LG, Ausubel FM, Cao H, et al. Plants and animals share functionally common bacterial virulence factors. Proc Natl Acad Sci USA2000; 97: Goodman AL, Kulasekara B, Rietsch A, Boyd D, Smith RS, Lory S. A signaling network reciprocally regulates genes associated with acute infection and chronic persistence in Pseudomonas aeruginosa. Dev Cell 2004; 7: Kipnis E, Sawa T, Wiener-Kronish J. Targeting mechanisms of Pseudomonas aeruginosa pathogenesis. Med Mal Infect 2006; 36: Burke V, Robinson JO, Richardson CJ, Bundell CS. Longitudinal studies of virulence factors of Pseudomonas aeruginosa in cystic fibrosis. Pathology 1991; 23: Jain M, Ramirez D, Seshadri R, et al. Type III secretion phenotypes of Pseudomonas aeruginosa strains change during infection of individuals with cystic fibrosis. J Clin Microbiol 2004; 42: Hoiby N. Understanding bacterial biofilms in patients with cystic fibrosis: current and innovative approaches to potential therapies. J Cyst Fibros 2002; 1: Miravitlles M, Espinosa C, Fernandez-Laso E, Martos JA, Maldonado JA, Gallego M. 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: Tingpej P, Smith L, Rose B, et al. Phenotypic characterization of clonal and nonclonal Pseudomonas aeruginosa strains isolated from lungs of adults with cystic fibrosis. J Clin Microbiol 2007;45: Huchon GJ, Gialdroni-Grassi G, Leophonte P, Manresa F, Schaberg T, Woodhead M. Initial antibiotic therapy for lower respiratory tract infection in the community: a European survey. Eur Respir J 1996;9: Jelsbak L, Johansen HK, Frost AL, et al. Molecular epidemiology and dynamics of Pseudomonas aeruginosa populations in lungs of cystic fibrosis patients. Infect Immun 2007; 75: CID 2008:47 (15 December) Martínez-Solano et al.

8 38. Ojeniyi B, Frederiksen B, Hoiby N. Pseudomonas aeruginosa crossinfection among patients with cystic fibrosis during a winter camp. Pediatr Pulmonol 2000; 29: 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: Ratjen F, Doring G, Nikolaizik WH. Effect of inhaled tobramycin on early Pseudomonas aeruginosa colonisation in patients with cystic fibrosis. Lancet 2001; 358: Chronic P. aruginosa Infection in COPD CID 2008:47 (15 December) 1533