Elsevier

The Lancet

Volume 369, Issue 9560, 10–16 February 2007, Pages 482-490
The Lancet

Articles
Effect of azithromycin and clarithromycin therapy on pharyngeal carriage of macrolide-resistant streptococci in healthy volunteers: a randomised, double-blind, placebo-controlled study

https://doi.org/10.1016/S0140-6736(07)60235-9Get rights and content

Summary

Background

Resistance to antibiotics is a major public-health problem, and studies that link antibiotic use and resistance have shown an association but not a causal effect. We used the macrolides azithromycin and clarithromycin to investigate the direct effect of antibiotic exposure on resistance in the oral streptococcal flora of healthy volunteers.

Methods

Volunteers were treated with azithromycin (n=74), clarithromycin (74), or placebo (76) in a randomised, double-blind trial. Pharyngeal swabs were obtained before and after administration of study treatment through 180 days. The proportion of streptococci that were macrolide resistant was assessed and the molecular basis of any change in resistance investigated. Analyses were done on an intent-to-treat basis. This study is registered with ClinicalTrials.gov, number NCT00354952.

Findings

The number of dropouts (n=20) was much the same in all groups until day 42; dropouts increased substantially at day 180 (105). Both macrolides significantly increased the proportion of macrolide-resistant streptococci compared with the placebo at all points studied, peaking at day 8 in the clarithromycin group (mean increase 50·0%, 95% CI 41·7–58·2; p<0·0001) and at day 4 in the azithromycin group (53·4%, 43·4–63·5; p<0·0001). The proportion of macrolide-resistant streptococci was higher after azithromycin treatment than after clarithromycin use, with the largest difference between the two groups at day 28 (17·4% difference, 9·2–25·6; p<0·0001). Use of clarithromycin, but not of azithromycin, selected for the erm(B) gene, which confers high-level macrolide resistance.

Interpretation

This study shows that, notwithstanding the different outcomes of resistance selection, macrolide use is the single most important driver of the emergence of macrolide resistance in vivo. Physicians prescribing antibiotics should take into account the striking ecological side-effects of such antibiotics.

Introduction

Resistance to antibiotics is a major public-health problem.1 Many ecological studies have shown a clear relation between antimicrobial use and resistance.2 However, these studies are commonly confounded by a number of variables, and they show, at best, an association, not a causal effect. Moreover, they do not link antibiotic exposure in an individual to the outcome for that individual, which creates the so-called ecological fallacy.3 Randomised clinical trials and, to some extent, observational studies that examine antibiotic-exposed versus non-exposed individuals are crucial to study definitively the link between antibiotic use and resistance as well as to provide in-vivo biological samples to study the molecular basis of resistance.4, 5, 6 Previously, such evidence was provided solely by animal experimentation.7, 8

Two macrolides—clarithromycin and azithromycin—are among the drugs of choice for the treatment of respiratory tract infections. Study of the link between antibiotic use and resistance—as well as the molecular mechanisms of resistance—is especially important because resistance to macrolides in common respiratory pathogens (eg, Streptococcus pneumoniae and Streptococcus pyogenes) is increasing,9, 10 and is most likely due to their inappropriate use.2

There is much debate as to which of these macrolides has greater potential for selecting resistant organisms,11 one of the decisive factors for eventual preference for clinical use. Azithromycin has a long half-life,12 and therefore a convenient dose regimen (once daily for 3 days, compared with twice daily for 7 days for clarithromycin). Theoretically, however, shorter drug exposure decreases the chance of the development of resistance, whereas higher tissue persistence and slowly receding azithromycin concentrations increase the chance of development of drug-resistant organisms. Some ecological studies have identified a strong relation between azithromycin use and macrolide resistance,13 whereas others did not find a correlation.14

The few studies that have compared the effect of azithromycin and clarithromycin on the selection of resistance have also shown conflicting results.15, 16, 17 Kastner and Guggenbichler16 showed that a significantly higher proportion of paediatric patients carried resistant organisms for 6 weeks after azithromycin treatment than did those treated with clarithromycin. King and co-workers17 were unable to show any difference between the two macrolides because of a lack of power, whereas Matute and colleagues,15 in a randomised double-blind study on 18 individuals, showed that neither clarithromycin nor azithromycin use selected for any resistant organisms in either the faecal or the oropharyngeal flora.

Resistance to macrolides in streptococci occurs via two main mechanisms. The first is active drug efflux mediated by a pump encoded by the mef (macrolide efflux) gene that confers low to moderate resistance against macrolides, with minimum inhibitory concentrations (MIC) to erythromycin (a prototype macrolide) ranging from 0·5 μg/mL to 32 μg/mL. In the second mechanism, a methylase encoded by the erm(B) gene modifies the macrolide binding site on the bacterial ribosome, generally conferring a high degree of resistance, with erythromycin MIC typically ranging from 32 μg/mL to more than 512 μg/mL.18

Various studies have shown that the oral commensal streptococcal flora endemically harbours the same macrolide resistance genes seen in the genetically related pathogenic streptococci.19, 20 Thus, we used the oral commensal streptococcal flora as model organisms to study the effect of different macrolides in selecting macrolide resistance in a healthy population. We did a randomised double-blind placebo-controlled trial with azithromycin and clarithromycin to investigate the direct effect of antibiotic exposure on resistance in the oral streptococcal flora of healthy volunteers. We also aimed to investigate the molecular basis for any differences in selection for resistance.

Section snippets

Participants

We did a randomised, double-blind, placebo-controlled trial between October, 2002, and May, 2003 (ie, during winter) and between March, 2003, and October, 2003 (ie, during summer) at the University of Antwerp, Belgium. Volunteers were selected on the basis of Belgian identity cards (to exclude those younger than 18 years) and a written questionnaire that included information on their sex, age, smoking status, previous antibiotic use, and employment in hospital with contact with patients. Of the

Results

Figure 1 shows the trial profile. 224 volunteers were randomly assigned a course of clarithromycin, azithromycin, or placebo. Age, sex, mean proportion of macrolide-resistant streptococci, and other characteristics of the volunteers were much the same in all three groups at baseline (table 1).

20 volunteers did not continue with the study; of these individuals, reasons for 15 dropouts were not known, two experienced side-effects due to the antibiotics being administered, one was diagnosed with

Discussion

By use of oral streptococci as model organisms, we have shown that macrolide use is the single most important driver of the emergence of macrolide resistance in human beings. Our results also show important differences in the outcome of selection of resistance by two antibiotics in the same class: azithromycin selected quantitatively more resistant organisms in the early post-therapy phases, whereas clarithromycin qualitatively selected for the higher resistance-conferring erm(B) gene. Finally,

References (39)

  • A Kaukas et al.

    The effect of growth-promoting antibiotics on the faecal enterococci of healthy young chickens

    J Appl Bacteriol

    (1988)
  • SB Levy et al.

    Changes in intestinal flora of farm personnel after introduction of a tetracycline-supplemented feed on a farm

    N Engl J Med

    (1976)
  • RR Reinert et al.

    Antimicrobial susceptibility of Streptococcus pneumoniae in eight European countries from 2001 to 2003

    Antimicrob Agents Chemother

    (2005)
  • S Malhotra-Kumar et al.

    Macrolide- and telithromycin-resistant Streptococcus pyogenes, Belgium, 1999–2003

    Emerg Infect Dis

    (2005)
  • JM Blondeau

    Differential impact of macrolide compounds in the selection of macrolide nonsusceptible Streptococcus pneumoniae

    Therapy

    (2005)
  • G Foulds et al.

    The pharmacokinetics of azithromycin in human serum and tissues

    J Antimicrob Chemother

    (1990)
  • RJ Davidson et al.

    Macrolide-resistant Streptococcus pneumoniae in Canada: correlation with azithromycin use

    Clin Microbiol Infect

    (2001)
  • M Bergman et al.

    Effect of macrolide consumption on erythromycin resistance in Streptococcus pyogenes in Finland in 1997-2001

    Clin Infect Dis

    (2004)
  • AJ Matute et al.

    Double-blind, placebo-controlled study comparing the effect of azithromycin with clarithromycin on oropharyngeal and bowel microflora in volunteers

    Eur J Clin Microbiol Infect Dis

    (2002)
  • Cited by (0)

    View full text