Acute exacerbations of chronic obstructive pulmonary disease: etiological bacterial pathogens and antibiotic resistance in Upper Egypt

Context Previous data on etiologic bacteria in acute exacerbations of chronic obstructive pulmonary disease (AECOPD) in Upper Egypt are limited. Aim The aim of this study was to identify the causative bacteria in AECOPD and to determine the antibiotic resistance patterns for AECOPD in Upper Egypt. Settings and design The study design was a prospective one and was conducted in a University Hospital. Materials and methods Patients who were admitted in Assiut University Hospital with AECOPD were prospectively enrolled. Sputum specimens were investigated using culture. Susceptibilities of the isolated bacterial strains to different antibiotics were determined using the disk diffusion method. Results During 18 months, 156 patients who experienced 218 AECOPD were enrolled. A significant bacterial growth was found in 77% of patients during 81% of exacerbations. The most commonly detected bacteria were Haemophilus influenzae (18%), Streptococcus pneumoniae (15%), and Klebsiella pneumoniae (14%). The majority of the isolated strains showed high resistance rates to most groups of antibiotics; 63% of the isolated strains were multidrug resistant, 29% were extensively drug resistant, and 5% were pandrug resistant. High resistance rates were observed against penicillins and cephalosporins, moderate rates against fluoroquinolones, and lowest rates against the carbapenems. All gram-positive bacteria were sensitive to linezolid. Increased severity of chronic obstructive pulmonary disease was related to increased prevalence of antibiotic resistance. Conclusion The predominant bacterial pathogens for AECOPD in Upper Egypt are H. influenzae, S. pneumoniae, and K. pneumoniae. Bacterial resistance rates were the highest against penicillins and cephalosporins, moderate against fluoroquinolones, and least against carbapenems. Increased severity of chronic obstructive pulmonary disease is related to an increased prevalence of antibiotic resistance.


Introduction
Chronic obstructive pulmonary disease (COPD) is a progressive inflammatory airway disorder characterized by airflow limitation due to a mixture of small airway disease, chronic lung inflammation, parenchymal destruction, and increased airway responsiveness [1]. The chronic course of COPD is often accompanied by acute exacerbations [acute exacerbations of chronic obstructive pulmonary disease (AECOPD)], mainly due to increased inflammation. Patients prone to frequent AECOPD have impaired health status, reduced physical activity levels, increased lower airway bacterial colonization, and accelerated lung function decline [1,2]. Thus, the management of exacerbations with prompt diagnosis and effective treatment should be a major goal in COPD [2].
The majority of COPD exacerbations are caused by infections of the tracheobronchial tree [3]. A key characteristic of airway inflammation in COPD is the persistent presence of bacteria in the lower airways. The most commonly isolated bacteria in the lower respiratory tract of COPD patients were Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus pneumoniae, with growing evidence of the significance of Pseudomonas aeruginosa infections in severe COPD disease [4]. Previous data on infection exacerbations of COPD in Upper Egypt are limited [5]. Therefore, the present study was conducted to identify the causative bacteria in AECOPD and to determine the antibiotic susceptibility and resistance patterns among these pathogens at Assiut University Hospital, Upper Egypt.

Study population
This prospective study aimed to identify the causative bacteria, antibiotic sensitivity, and antibiotic resistance in hospitalized patients due to AECOPD in Upper Egypt. All patients previously diagnosed for COPD and admitted with AECOPD in the Department of Chest Diseases, Assiut University Hospital, during the period between September 2014 and March 2016, were candidates for inclusion in the study. COPD diagnosis was based on medical history, clinical presentation, and pulmonary function tests recorded previously to the current admission and was defined according to the Global Initiative for Chronic Obstructive Lung Disease guidelines [2]. AECOPD was defined using the same guidelines. Patients comprised those admitted both at the ward and higher care units (ICU and intermediate care unit), excluding those who underwent invasive mechanical ventilation. Exclusion criteria were diagnosis of another acute respiratory condition (e.g. pneumonia, pneumothorax, or pulmonary embolism), history of respiratory disorders other than COPD (e.g. asthma, bronchiectasis, pulmonary fibrosis, and pulmonary tuberculosis), or inability and/or unwillingness to cooperate for the study.
The study was explained to the participants, and all participants signed an informed consent form that was approved by the Institutional Ethical Committee.

Clinical and functional assessment
All patients underwent thorough clinical examination, pulmonary function testing (spirometry), arterial blood gases evaluation, and radiological evaluation. Chest radiographies were performed for all patients on admission. Computed tomographic scans or computed tomography pulmonary angiography was performed when appropriate or dictated by the clinical condition. Smoking index was calculated as the product of tobacco use (years) and the average number of cigarettes smoked per day/20 (one pack has 20 cigarettes) [6]. Lung volumes (forced expiratory volume in first second, and forced vital capacity) were obtained and calculated during the hospitalization due to the AECOPD. A spirometer (Zan 300; Sensor Medics MGA USB, Oberthulba, Germany) was used. Reversibility was expressed as a percentage of the predicted forced expiratory volume in first second values. Lung volumes were measured before and 10 min after administration of 400 μg of salbutamol. The maneuvers performed and the reference values used were those of the international recommendations [7]. Severity of airway obstruction was defined using Global Initiative for Chronic Obstructive Lung Disease guidelines and criteria as mild, moderate, severe, or very severe disease (stages I-IV, respectively).

Laboratory tests
Venous blood samples were obtained from patients for performing relevant laboratory investigations: complete blood count, blood glucose level, renal function tests, and erythrocyte sedimentation rate.

Bacteriological diagnosis
From each patient, a sputum sample was collected within 24 h of hospital admission, before antimicrobial therapy onset. Valid early morning sputum samples were collected into sterile cups from patients through effective coughing, sometimes assisted by physiotherapy to obtain lung secretions as described previously [8]. Samples were transported directly to the Microbiology and Immunology Department, Faculty of Medicine, Assiut University, where the bacteriological analyses were performed. Sputum samples were considered suitable for culture, with the presence of more than 25 polymorphonuclear leukocytes and less than 10 squamous epithelial cells per magnification field.

Identification of the causative bacterial strains
The samples were examined microscopically after staining with Gram's stain and cultured directly on nutrient, blood, chocolate, mannitol salt, bile esculin, CHROM agar, MacConkey's, and eosin methylene blue agar plates. The cultured plates were incubated aerobically at 37°C for 24-48 h. Blood and chocolate agar plates were incubated at 35-36°C with 5% CO 2 for 48 h for isolation of S. pneumoniae, H. influenzae, and Moraxella catarrhalis strains. Bacterial isolates were identified on the basis of colonial morphology, Gram staining, and standard biochemical reactions according to Bergey's Manual of Systematic Bacteriology [9].

Antibiotic susceptibility testing
Susceptibilities of the isolated bacterial strains were determined to penicillins (amoxicillin and amoxicillin-clavulanic acid), cephalosporins (ceftriaxone, cefepime, ciprofloxacin, and cefotaxime), fluoroquinolones (levofloxacin and ofloxacin) (Bioanalysis, Turkey). In addition, susceptibilities of gram-positive bacterial strains were tested against other penicillins (oxacillin and methicillin), macrolides (erythromycin), glycopeptides (vancomycin and teicoplanin), and oxazolidinones (linezolid). Susceptibilities of gram-negative stains were tested also against aminoglycosides (gentamicin and amikacin) and carbapenems (imipenem and meropenem). The test was performed using the disk diffusion method as recommended by the Clinical and Laboratory Standards Institute guidelines [10]. The results were interpreted as susceptible, intermediate, or resistant.
Multidrug resistance (MDR) was defined as an acquired nonsusceptibility to at least one agent in three or more antimicrobial categories; extensive drug-resistance (XDR) was defined as nonsusceptibility to at least one agent in all but two or fewer antimicrobial categories; and pandrugresistance (PDR) bacteria was defined as nonsusceptibility to all agents in all antimicrobial categories [11].

Statistical analysis
Statistical analysis was performed using the SPSS 19.0 software for Windows (SPSS Inc., Chicago, Illinois, USA). Data were presented as mean and SD or number and percentage, as appropriate. A P value less than 0.05 was considered statistically significant.

Demographic and clinical patients' data
During the study period, 156 patients with a total number of 218 attacks of AECOPD were enrolled prospectively into the current study. There were 144 (92%) male and 12 (8%) female patients, with a mean age of 57.6±8 years (range 40-72 years). Of the 156 patients, 69 (44%) patients had very severe (stage IV) COPD, 45 (29%) had severe (stage III) COPD, 32 (21%) had moderate, and 10 (6%) had mild disease. The duration of hospital stay ranged from 3 to 42 days (mean±SD, 14±9 days). It was significantly longer in patients with very severe versus those with severe, moderate, and mild COPD (analysis of variance, P=0.022, 0.001, and 0.000, respectively) and in patients with severe COPD versus moderate and mild COPD (analysis of variance, P=0.002 and 0.000, respectively). Table 1 shows the detailed demographic, clinical, and bacteriologic data of the enrolled patients.

Antibiotic susceptibility patterns
High resistance rates were observed among the isolated bacterial strains against most groups of antibiotics; 115 (65%) of the isolated strains were MDR, 52 (29%) were XDR, and nine (5%) were PDR (Table 2). Most isolates were resistant to amoxicillin, amoxicillin/clavulanic acid, cephalosporins (except ciprofloxacin), and ofloxacin. About half the isolates were resistant to ciprofloxacin and levofloxacin. Among the gram-positive bacteria, resistance rates were high against the penicillin group and erythromycin, whereas resistance to vancomycin, and teicoplanin ranged from 54 to 61%. All isolated gram-positive bacteria were sensitive to linezolid. For gram-negative isolates, the resistance rates to the aminoglycosides group ranged from high level to gentamicin to a slightly lower level to amikacin. Only few isolates showed resistance to the carbapenem group that belonged to H. influenzae (13%) and E. coli (11%), whereas K. pneumoniae, P. aeruginosa, and M. catarrahlis were completely susceptible to the carbapenem group ( Table 2).
The patterns of antibiotic resistance in different COPD stages are shown in Fig. 2. The advancements in COPD stage was related to the increased prevalence of antibiotic-resistant bacterial strains. Patients with stage IV disease had 72, 30, and 5 exacerbation attacks with MDR, XDR, and PDR organisms, respectively. However, patients with stage II disease had 14, 4, and 2 exacerbation attacks with MDR, XDR, and PDR organisms, respectively (Fig. 2).

Discussion
This prospective study aimed to identify the causative bacteria, antibiotic sensitivity, and antibiotic resistance in hospitalized patients due to AECOPD in Upper Egypt. The chronic course of COPD is often accompanied by acute exacerbations (AECOPD), mainly due to increased inflammation. Morbidity and mortality in COPD patients are, for the most part, related to those acute exacerbations [2]. Published data had shown that the majority of AECOPD (50-60%) are due to respiratory tract infection, the minority (10%) are due to environmental factors, whereas in nearly 30% of cases, the etiology remains unclear [12]. COPD exacerbations are frequently triggered by upper respiratory tract infections, and these are more common in the winter months, when there are more respiratory viral infections in the community [13]. It is also possible that patients are more susceptible to exacerbations in the winter months, as lung function in COPD patients shows small but significant decreases with reduction in outdoor temperature during the winter months [13]. However, data from previous studies showed that between 25 and 50% of COPD patients have lower airway colonization by bacteria, especially noncapsulated H. influenzae, S. pneumoniae, and M. catarrhalis. This colonization has been related to the severity of COPD and cigarette smoking [14].
The presence of bacteria in the lower airways of COPD patients implies a breach of host defense mechanisms, which sets up a vicious cycle of epithelial cell damage, impaired mucociliary clearance, mucus hypersecretion, increased submucosal vascular leakage, and inflammatory cell infiltration. The airway bacterial load in the stable state is associated with airway inflammatory markers, and thus increased bacterial colonization is associated with greater airway inflammation [15]. The presence of bacterial colonization in COPD patients may have an influence both on the exacerbation frequency and on the character and severity of exacerbations [13,16]. Fortunately, at AECOPD, there is an increased chance of detecting bacteria, especially if the exacerbation is associated with the presence of purulent sputum. With antibiotic therapy, bacterial load and airway inflammation decreases, and the rate of resolution of the airway inflammatory changes is related to the clearance of bacteria from the sputum [17].
In the current study, most of the participated patients were old aged, suffering from severe (29% of patients) or very severe (44%) COPD, had associated cardiopulmonary or systemic comorbidities, and were critically ill and required admission in the ICU (41%) or the intermediate care unit (30%). Moreover, our data showed that 87% of the enrolled patients were smokers, which reflected the effect of current smoking as a risk factor for severe exacerbations. Smoking perpetuates an ongoing inflammatory response that leads to airway narrowing and hyperactivity, and hence patients become more prone to infection exacerbation attacks [18]. Moreover, we found that exsmokers (10% of patients) had experienced exacerbation attacks, which imply that smoking cessation was too late and the disease progression continued even after smoking cessation. The duration of hospital stay was significantly longer in patients with severe and very severe COPD versus those with mild or moderate disease, which corresponds to previous reports [19]. Shorter durations of exacerbations were a predictor of success of treatment, whereas longer durations were a predictor of need for ventilatory support and poor outcome of the disease [20]. Antibiotic resistance patterns in relation to different COPD stages. Significant bacterial growth was detected in 176 exacerbation attacks. MDR, multidrug resistance; PDR, pandrug-resistance; XDR, extensive drug-resistance.
A significant bacterial growth was found in 77% of patients during 176 of 218 (81%) exacerbation attacks, and a total of 176 bacterial strains were isolated. The most predominantly encountered strains were H. influenzae, S. pneumoniae, and K. pneumoniae isolated in 31 (18%), 26 (15%), and 24 (14%) attacks, respectively. H. influenzae was the most common bacteria detected in our study. This is in accordance with previous works in Egypt [5] and other countries [21,22]. Strains of H. influenzae stimulate mucus hypersecretion and inhibit ciliary beat frequency. Furthermore, they can cause direct epithelial damage and their endotoxin increase epithelial expression of the proinflammatory cytokines, thus providing potential mechanisms to upregulate the process of inflammation in COPD [12]. S. pneumoniae strains were detected in 15% of AECOPD in this work. A previous documentation has shown that airway colonization with S. pneumoniae increases the risk for a first COPD exacerbation [23]. A substantial number (9%) of our patients had P. aeruginosa that was lower than that reported previously (15%) by other studies [24]. COPD was considered as an independent factor in the isolation of MRSA in the ICU [25]. MRSA were detected in 11% of our participated patients. Although A. baumannii is a major pathogen in nosocomial infections, communityacquired acinetobacter infections are of an increasing concern because they mainly affect patients with certain comorbidities such as COPD [26]. A. baumannii strains were found in 7% of AECOPD in our work.
In disagreement with previously reported data, [27] found a relationship between the severity of COPD and the type of isolated bacterial strains. Our results found no significant association between the type of bacterial isolates and severity of COPD. This difference may be due to different demographic data and the small sample size of our study. A larger sample size is required to prove these findings. However, we observed that the number of exacerbation attacks, and hence the number of bacterial isolates, was proportionately related to the severity of COPD. Thus, the number of isolates were 107 (61%), 49 (28%), and 20 (11%), in COPD stages IV, III, and II, respectively. No bacterial pathogens were isolated in stage I.
To obtain high susceptibilities to antimicrobial agents, we tested the susceptibilities of the isolated bacterial strains to major groups of antibiotics that have effect against both gramnegative and gram-positive bacteria. Our findings demonstrated high resistance rates among the isolated bacterial strains to different groups of antibiotics; 115 (65%) of the isolated strains were MDR, 52 (29%) were XDR, and 9 (5%) were PDR. Most isolates were resistant to amoxicillin, amoxicillin/clavulanic acid, cephalosporins (except ciprofloxacin), and ofloxacin. This finding is similar to that obtained previously in Upper Egypt [5]. About half the isolates in the current study were resistant to ciprofloxacin, and levofloxacin. A similar rate of sensitivity to ciprofloxacin was also observed in a previous data from India [28]. Previous studies evidenced the high bacteriological eradication rate in AECOPD patients when treated with levofloxacin [29].
All S. aureus strains in our COPD patients were oxacillin and methicillin resistant, which was similar to that reported in previous reports [27,30]. This high prevalence of MRSA in the current study should be an alarm for the increasing prevalence of MRSA among hospitalized patients in our locality. This coincides with the recent report by Borg et al. [31], who observed that the prevalence of MRSA in invasive isolates from blood cultures from nine hospitals in Egypt was 52%. This should alarm the local health authorities to take the proper infection control measures against increasing prevalence of MRSA. All isolated gram-positive bacteria in this work were sensitive to linezolid, the first commercially available oxazolidinone antibiotic. Similarly, linezolid was active against gram-positive isolates in previous studies in the UK [32].
In comparison with other members of the aminoglycoside group used in this study, resistance rate to amikacin was slightly lower among the gramnegative isolates. This is similar to previous reports from Egypt [5,33]. Sensitivity of our gram-negative bacteria was at the highest level to the carbapenem group, which was similar to previous studies from Egypt [33] and China [34]. Again, our data revealed that MDR bacteria were isolated at a rate of 65%, which is higher than that reported in previous reports [33,35]. Emergence of resistance to multiple antimicrobial agents in pathogenic bacteria has become a significant public health threat as there are fewer, or even sometimes no, effective antimicrobial agents available for infections caused by these bacteria. Gram-positive and gram-negative bacteria are both affected by the emergence and rise of antimicrobial resistance [11]. The situation is compounded by crossresistance within and between classes of antibacterial agents, which further limits treatment options [36]. Our data showed that the advancements in COPD stage was related to the increased prevalence of antibiotic-resistant bacterial strains. This finding highlights the importance of smoking cessation as an effective measure for preventing progression of the disease.
Finally, it would be appropriate if 'follow-up' sputum cultures are carried out before discharging the patient, to get an idea about possible resistant organisms, keeping these data in the patient's file for future use. However, in the current study we could not perform that due to financial concerns.
At the end, our results have many similarities to and differences from other studies. Continued surveillance, particularly based on local data, is obviously needed to clarify the problems of antimicrobial resistance and to prevent further spread of such resistance. Data from this study can be very useful. A master antibiogram for our region would allow tertiary care institutions to consider resistance patterns in hospitals referring patients and to select appropriate antimicrobial therapy or change drugs in nonresponding patients. Implementing continued local surveillance programs for antibiotic resistance is essentially important. Moreover, further local studies should be carried out to elucidate the mechanisms of resistance of different pathogens in AECOPD. Judicious use of antimicrobials is essential to prevent the emergence of resistant and/or MDR bacteria in AECOPD.

Conclusion
H. influenzae, S. pneumoniae, and K. pneumoniae are the leading bacterial pathogens in patients with AECOPD in Upper Egypt. Our bacteriological profiles highlighted the role of other pathogens, including E. coli and MRSA, in AECOPD. The isolated bacterial strains were characterized by high resistance rates to most groups of antimicrobials. Sensitivity was high to linezolid and the carbapenem group. More advanced COPD stage is related to increased prevalence of antibiotic-resistant bacterial strains. Further local studies are warranted.

Financial support and sponsorship
Nil.

Conflicts of interest
There was no conflicts of interest.