Multiple colonization with S. pneumoniae before and after introduction of the seven-valent conjugated pneumococcal polysaccharide vaccine

Abstract

Background: Simultaneous carriage of more than one strain of Streptococcus pneumoniae promotes horizontal gene transfer events and may lead to capsule switch and acquisition of antibiotic resistance. We studied the epidemiology of cocolonization with S. pneumoniae before and after introduction of the seven-valent conjugated pneumococcal vaccine (PCV7).

Methodology: Nasopharyngeal swabs (n 1120) were collected from outpatients between 2004 and 2009 within an ongoing nationwide surveillance program. Cocolonization was detected directly from swabs by restriction fragment length polymorphism (RFLP) analysis. Serotypes were identified by agglutination, multiplex PCR and microarray.

Principal findings: Rate of multiple colonization remained stable up to three years after PCV7 introduction. Cocolonization was associated with serotypes of low carriage prevalence in the prevaccine era. Pneumococcal colonization density was higher in cocolonized samples and cocolonizing strains were present in a balanced ratio (median 1.38). Other characteristics of cocolonization were a higher frequency at young age, but no association with recurrent acute otitis media, recent antibiotic exposure, day care usage and PCV7 vaccination status.

Conclusions: Pneumococcal cocolonization is dominated by serotypes of low carriage prevalence in the prevaccine era, which coexist in the nasopharynx. Emergence of such previously rare serotypes under vaccine selection pressure may promote cocolonization in the future.

Figure 1. Ratio of strains present in cocolonized nasopharyngeal swabs.

Strain ratios were determined from the peak heights in the chromatograms obtained by terminal-restriction fragment length polymorphism analysis (T-RFLP, as described in the methods section). The X-axis indicates the fold-higher presence of the more abundant strain. The three samples with three strains were excluded.

Figure 2. Serotype distribution among nasopharyngeal S. pneumoniae isolates.

A. Serotype distribution among nasopharyngeal isolates (n 468) of S. pneumoniae stratified according to the detection of cocolonization or single colonization with S. pneumoniae strains. The group “cocolonizing” includes the 64 isolates from cocolonized samples, for which a serotype could be determined. Serotypes with a relative frequency of less than 2% are summarized as “others”. They included serotypes/-groups 1, 8, 9 (except 9V), 10, 16, 17, 17F, 18 (except 18C), 20, 21, 28, 33, 34, 35, 38 and non-typeable isolates. The rate of rare serotypes (“others”) was significantly higher in cocolonized than in single colonized swabs (23.4% vs. 11.4%, p = 0.01), but difference in the frequency of individual serotypes (for example for the serotypes 11, 15, and 19F) did not reach statistical significance. B. Distribution of rare serotypes (group “other” in Figure 2A) in samples with or without cocolonization. The difference between the proportion of non-typeable strains in cocolonized samples (4.7%) and in those with a single strain (0.5%) was statistically significant (Fisher's exact test p = 0.02).

Figure 3:. Relative distribution of pneumococcal serotypes among 468 pneumococcal strains isolated in prevaccine or vaccine era.

The relative frequency of pneumococcal serotypes isolated from nasopharyngeal swabs before and after the introduction of the seven-valent conjugated polysaccharide vaccine for (A) swabs with one serotype only (prevaccine era n 106, vaccine era n 298), and (B) restricted to isolates from cocolonized samples (prevaccine era n 12, vaccine era n 52). For easier reading of the graph rare serotypes (prevalence <2%) were grouped as “others”. They included serotypes and/or serogroups 1, 8, 9 (except 9V), 10, 16, 17, 17F, 18 (except 18C), 20, 21, 28, 33, 34, 35, 38 and non-typeable isolates.