The effect of carbon dioxide on susceptibility testing of azithromycin, clarithromycin and roxithromycin against clinical isolates of Streptococcus pneumoniae and Streptococcus pyogenes by broth microdilution and the Etest: Artemis Project-first-phase study

Seasonal Azithromycin Use in Paediatric Protracted Bacterial Bronchitis Does Not Promote Antimicrobial Resistance but Does Modulate the Nasopharyngeal Microbiome

Protracted bacterial bronchitis (PBB) causes chronic wet cough for which seasonal azithromycin is increasingly used to reduce exacerbations. We investigated the impact of seasonal azithromycin on antimicrobial resistance and the nasopharyngeal microbiome. In an observational cohort study, 50 children with PBB were enrolled over two consecutive winters; 25/50 at study entry were designated on clinical grounds to take azithromycin over the winter months and 25/50 were not. Serial nasopharyngeal swabs were collected during the study period (12-20 months) and cultured bacterial isolates were assessed for antimicrobial susceptibility. 16S rRNA-based sequencing was performed on a subset of samples. Irrespective of azithromycin usage, high levels of azithromycin resistance were found; 73% of bacteria from swabs in the azithromycin group vs. 69% in the comparison group. Resistance was predominantly driven by azithromycin-resistant S. pneumoniae, yet these isolates were mostly erythromycin susceptible. Analysis of 16S rRNA-based sequencing revealed a reduction in within-sample diversity in response to azithromycin, but only in samples of children actively taking azithromycin at the time of swab collection. Actively taking azithromycin at the time of swab collection significantly contributed to dissimilarity in bacterial community composition. The discrepancy between laboratory detection of azithromycin and erythromycin resistance in the S. pneumoniae isolates requires further investigation. Seasonal azithromycin for PBB did not promote antimicrobial resistance over the study period, but did perturb the microbiome.

Molecular characterisation of Hungarian macrolide-resistant Streptococcus pneumoniae isolates, including three highly resistant strains with the mef gene

The macrolide resistance of 304 Hungarian Streptococcus pneumoniae isolates was investigated. Antibiotic sensitivity testing was performed in air and in 5% CO(2). More erythromycin resistance was noted when growth was in CO(2). A resistance determinant was found in almost all isolates: erm(B) gene (87.4%), mef genes (9.2%) and one strain with the erm(TR) gene. This indicates that screening for carriage of resistance determinants should always be done in the presence of 5% CO(2). We found three isolates with mef(E), which were highly resistant to erythromycin. These contained multiple and some novel, ribosomal mutations. The most prevalent serogroups were 6, 19 and 14. Based on the PFGE pattern, we found identity between the Hungarian isolates and two PMEN clones.

Impact of carbon dioxide on the susceptibility of key respiratory tract pathogens to telithromycin and azithromycin

OBJECTIVES

To determine the quantitative differences in telithromycin and azithromycin MIC values against Streptococcus pneumoniae, Haemophilus influenzae and Streptococcus pyogenes obtained using two recommended and commonly used methodologies: CLSI reference standard broth microdilution in ambient air and Etest((R)) concentration gradient in CO(2).

METHODS

Four hundred clinical isolates (S. pneumoniae, n = 200; H. influenzae, n = 100; S. pyogenes, n = 100) were evaluated in seven independent laboratories. Telithromycin and azithromycin MICs were determined using CLSI broth microdilution panels incubated in ambient air and Etest strips incubated in CO(2). Standard quality control reference strains-S. pneumoniae ATCC 49619 (n = 10) and H. influenzae ATCC 49247 (n = 10)-were also tested.

RESULTS

Telithromycin and azithromycin Etest MICs in CO(2) were elevated for all organisms when compared with values obtained using broth microdilution in ambient air. Telithromycin geometric mean MIC values increased in CO(2) by 2.05, 1.00 and 1.78 log(2) dilutions for S. pneumoniae, H. influenzae and S. pyogenes, respectively. The corresponding values for azithromycin were 2.54, 1.21 and 3.0 log(2) dilutions, respectively.

CONCLUSIONS

Telithromycin MICs measured using Etest in CO(2) are consistently elevated compared with those generated by CLSI broth microdilution measured in ambient air. These findings indicate that Etest should not be routinely used for the determination of telithromycin MICs against S. pneumoniae, H. influenzae and S. pyogenes, unless appropriate corrective factors are applied before reporting MICs or applying interpretive susceptibilities. Based on results from this study, Etest MIC breakpoints and quality control ranges are proposed.

Influence of carbon dioxide on the MIC of telithromycin for Streptococcus pneumoniae: an in vitro-in vivo study

Incubation in CO(2) resulted in higher (> or =3 doubling dilution) MICs of telithromycin than those found in ambient air for 31.2% of 346 Streptococcus pneumoniae ermB-positive strains. An increased telithromycin MIC in CO(2) was not correlated with loss of its activity in the murine sepsis/peritonitis model.

Clinical relevance of laboratory susceptibility data

The breakpoints for macrolides proposed by various organisation differ. In vitro studies may not give a true measure of breakpoints as culture conditions are not necessarily analogous to the in vivo situation. The effects of pH, carbon dioxide and culture media can affect the minimum inhibitory concentrations of macrolides; the extent of these effects varies for the different agents. Microbiological efficacy should ideally be proven under clinical conditions, but this is not always feasible. The breakpoints of macrolides should be reviewed taking into account the site of infection.

Influence of CO(2) incubation on quinolone activity against Streptococcus pneumoniae and Haemophilus influenzae

Quinolone activity can be influenced by the pH change that occurs during CO(2) incubation when testing capnophilic organisms such as Streptococcus pneumoniae and Haemophilus influenzae. This study compares the activity of ciprofloxacin, clinafloxacin, enrofloxacin, fleroxacin, grepafloxacin, levofloxacin, moxifloxacin, norfloxacin, ofloxacin, pefloxacin, sparfloxacin, trovafloxacin, and the glycopeptide vancomycin, under ambient air and supplemental CO(2) incubation conditions. Etest (AB BIODISK, Solna, Sweden) was used to determine the MIC values of 30 S. pneumoniae strains including S. pneumoniae ATCC 49619 and 6030; and 29 H. influenzae strains including H. influenzae ATCC 49247 and 49766, incubated with and without 5% CO(2.) Reference broth microdilution and agar dilution tests were performed under similar conditions. Results determined that MIC values of all quinolones agreed (> or = 95%) within +/- one log(2) dilution for the three test methods tested under identical incubation conditions. Quinolone MICs for S. pneumoniae were minimally influenced by CO(2,) while MIC values for H. influenzae ranged from 0.5 to two log(2) dilutions higher when incubated in the CO(2) environment. The differences in quinolone activity against H. influenzae may be due to a combination of pH effects and enhanced growth in the CO(2) test conditions. Quality control and interpretative criteria, especially for H. influenzae and quinolones published by the National Committee for Clinical Laboratory Standards, may not be fully applicable to results obtained under CO(2) incubation. This may generate potential therapeutic interpretive dilemmas in the susceptibility testing of H. influenzae, where clinical isolates require CO(2) to sustain acceptable growth.

Susceptibility and resistance emergence studies with macrolides

An extended elimination half-life and good tissue penetration enable oral azithromycin to attain high and prolonged concentrations in infected tissues, yielding high antibacterial activity in vivo. It has been suggested, however, that prolonged subinhibitory concentrations of azithromycin from 2 to 4 weeks after therapy may lead to the emergence of azithromycin resistance in vivo, compared with other macrolide antibiotics. Data from two types of in vitro susceptibility studies, an animal tissue infection model, and a clinical pediatric study demonstrate that prolonged tissue concentrations of azithromycin are unlikely to lead to the emergence of resistance in the clinical setting. Further, data from in vitro susceptibility studies indicate that resistance to macrolides, and azithromycin in particular, is significantly over-estimated for bacterial strains incubated in the presence of CO2.