Some Pneumococcal Serotypes Are More Frequently Associated with Relapses of Acute Exacerbations in COPD Patients

Objectives To analyze the role of the capsular type in pneumococci causing relapse and reinfection episodes of acute exacerbation in COPD patients. Methods A total of 79 patients with 116 recurrent episodes of acute exacerbations caused by S. pneumoniae were included into this study (1995–2010). A relapse episode was considered when two consecutive episodes were caused by the same strain (identical serotype and genotype); otherwise it was considered reinfection. Antimicrobial susceptibility testing (microdilution), serotyping (PCR, Quellung) and molecular typing (PFGE/MLST) were performed. Results Among 116 recurrent episodes, 81 (69.8%) were reinfections, caused by the acquisition of a new pneumococcus, and 35 (30.2%) were relapses, caused by a pre-existing strain. Four serotypes (9V, 19F, 15A and 11A) caused the majority (60.0%) of relapses. When serotypes causing relapses and reinfection were compared, only two serotypes were associated with relapses: 9V (OR 8.0; 95% CI, 1.34–85.59) and 19F (OR 16.1; 95% CI, 1.84–767.20). Pneumococci isolated from relapses were more resistant to antimicrobials than those isolated from the reinfection episodes: penicillin (74.3% vs. 34.6%, p<0.001), ciprofloxacin (25.7% vs. 9.9%, p<0.027), levofloxacin (22.9% vs. 7.4%, p = 0.029), and co-trimoxazole (54.3% vs. 25.9%, p<0.001). Conclusions Although the acquisition of a new S. pneumoniae strain was the most frequent cause of recurrences, a third of the recurrent episodes were caused by a pre-existing strain. These relapse episodes were mainly caused by serotypes 9V and 19F, suggesting an important role for capsular type.


Introduction
Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality in developed countries [1]. Approximately 50% of acute exacerbation episodes of COPD are caused by bacterial pathogens, mainly Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis [2]. The development of an acute exacerbation episode caused by S. pneumoniae is thought to be associated with the acquisition of a new strain, although scarce information is available [3].
Capsular type, the principal pneumococcal virulence factor, had been related to the ability of pneumococci to cause invasive disease or colonization [4]. However, the aetiological role of pneumococ-cal serotypes in relapse episodes of COPD patients remains to be determined.
The aims of this study were to analyse the relationship between serotype and genotype and the ability to cause relapse or reinfection episodes in patients with COPD. In addition, we have explored the influence of previous antimicrobial therapy in this occurrence.

Epidemiological and clinical data
A total of 79 COPD patients were included in the study. Their mean age was 69 (SD 66) years, and 77 (97.5%) of them were males. In terms of COPD severity there was 1 patient with mild (1.3%), 14  Finally, a total of 116 recurrent episodes from the 79 patients were studied; of these, 35 (30.2%) were relapses and 81 (69.8%) reinfections. The mean time between episodes was 166696 days, being shorter among relapses (133689 days) than among reinfections (181696 days; P = 0.020). Table 1 shows the distribution of the episodes based on the mean time between episodes. When the consecutive episodes occurred during a period #120 days, significantly higher frequency of relapse episodes was observed (57.1% vs. 35.8%, P = 0.033). Whereas, when the period of time between episodes was higher than 240 days the frequency of reinfection episodes was higher (14.3% vs. 30.9%, P = 0.061).
In 13 of the 35 (37.1%) relapses and in 31 of the 81 (38.3%) reinfections, pneumococcal strains were isolated together with other potential pathogens.

Discussion
Capsular type is known to play an important role in the invasiveness of pneumococcal strains [4]. Thus, some serotypes have been associated with invasive pneumococcal disease or with acute exacerbations of COPD [5]. Although the isolation of a new pneumococcal strain has been associated with a significantly increased risk of a new acute exacerbation [2], little information is available about the persistence of S. pneumoniae isolates.
Our results agree with other reports in which reinfection through acquisition of a new strain was the most frequent cause of acute exacerbation episodes [3]. However, our study shows that a third of these recurrences were caused by a persistent pneumococcal strain, suggesting that in COPD patients the persistence of the same strain could be underestimated.
Although in 12 of 35 relapses S. pneumoniae was isolated together with another pathogen (P. aeruginosa, H. influenzae, M. catarrhalis or Staphylococcus aureus) the role of S. pneumoniae in causing the acute exacerbation episodes is supported by the high predominance of Gram positive diplococci in the Gram stain of a good quality sputum sample.
Our results show that serotypes 9V and 19F were associated with relapses, suggesting that serotype could play an important role in the persistence of pneumococcal isolates. In addition, differences in genotype distribution were also detected. All isolates that expressed the serotype 9V belonged to the Spain 9V -CC156; hence, this clone was associated with the relapse episodes. In contrast, several genotypes expressed the serotype 19F and none of them was significantly associated with relapses. These results suggest that capsular type, rather than genetic background, may play an important role in the persistence of pneumococci among COPD patients.
Most of the patients included in the study had severe or very severe COPD, suffering frequent episodes of acute exacerbation and they received multiple antibiotic courses [6]. Although there were no differences in the betalactam consumption among groups, the betalactam resistance rates were higher among relapse episodes. This finding could be explained because relapse episodes were caused by few multi-resistant clones (mainly CC156 and CC88). Whereas, strains causing reinfection episodes showed a higher genetic diversity including penicillin-susceptible andresistant clones. However, we found an association between fluorquinolone consumption and development of resistance. In fact, the development of fluorquinolone resistance during or after an antimicrobial course has been largely described in the literature [7][8].
The proportion of serotypes covered by the 23vPPV vaccine was high, especially those causing relapses. Unfortunately, vaccination data of patients included in the present study was not available; however, its protective efficacy in COPD populations is controversial since COPD adults respond differently than the general adult population, due to their impaired antibody response to the vaccine, the colonization of the lower respiratory tract, or the frequent use of inhaled corticosteroids [9]. In the other hand, conjugate vaccines (PV10 and PCV13) vaccine, which have an enhanced immunity potential, could prevent the 40% and a half of the overall relapse episodes, respectively.
The major limitations of our study are the low number of relapse episodes, and also that it is a retrospective study. Nonetheless, our study provides new data about the association of certain serotypes with the persistence of pneumococci and the ability to some clones, especially Spain 9V -CC156, to cause relapse episodes. In addition, our study suggests that new episodes that occurred within the first 3 months after a previous one, had higher probability to be caused by the same pneumococcal strain and this fact could help to give an adequate empirical therapy. Further studies with a high number of recurrent episodes are now needed to investigate not only the role of capsular type in relapses of acute exacerbations, but also whether the pneumococcal conjugate vaccine 13 could be beneficial for COPD patients.

Study design
Pneumococci and other potential pathogens isolated from sputum samples were prospectively collected into our laboratory between 1995 and 2010, and were frozen at 280uC for further analysis. Only pneumococci isolated from good quality sputum were considered (,10 squamous cells and .25 leucocytes per lowpower field), with a predominance of Gram positive diplococci.
All COPD patients (n = 79) with two or more acute exacerbation episodes and seen at the Bellvitge University Hospital during the study period were included, after retrospective review of their computerized medical charts. Only those consecutive acute exacerbations which lasted for between four weeks and one year were included in the study.
The severity of airflow obstruction was categorized according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria [10].
An acute exacerbation of COPD was defined as any sustained increase in respiratory symptomatology compared with the baseline situation requiring an increase in regular medication and hospital treatment. A 'relapse' episode was defined as two or more consecutive acute exacerbations caused by the same pneumococcus (identical serotype and genotype). When the consecutive episodes were caused by pneumococci with different serotype and Pulsed Field Gel Electrophoresis (PFGE) pattern they were defined as 'reinfection'.

Ethical statement
This study and publication of the results were approved by the 'Comité È tic d'Investigació Clínica del Hospital Universitari de Bellvitge' and the written or oral informed consent was considered not necessary, because the source of bacterial isolates was anonymized and the study was retrospective.

Serotyping and genotyping
Serotyping was performed by multiplex PCR, using a previously described methodology [11]. All isolates were genotyped by PFGE. Multi Locus Sequence Typing (MLST) was performed on all relapse isolates in order to confirm the identity of the isolates [12][13].
Antimicrobial susceptibility Antimicrobial susceptibility was tested by microdilution (STRHAE, Sensititre TM ), following the Clinical Laboratory Standards Institute (CLSI) criteria [14]. The ciprofloxacin MIC of resistant strains (MIC $4 mg/L) was confirmed by E-test. S. pneumoniae ATCC49619 was used as the control strain.

Statistical analysis
Statistical analyses were carried out using SPSS 18 for Windows. The odds ratios (OR) and 95% confidence intervals (CI) were calculated, and Fisher's exact test was used when appropriate. Two-sided P values ,0.05 were considered statistically significant.