JAC Advance Access originally published online on February 9, 2006
Journal of Antimicrobial Chemotherapy 2006 57(4):741-746; doi:10.1093/jac/dkl014
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Macrolide resistance of Staphylococcus aureus and Haemophilus species associated with long-term azithromycin use in cystic fibrosis
1 Department of Pediatric Pulmonology and Allergology, Erasmus MC–Sophia Children's Hospital, Rotterdam, The Netherlands; 2 Department of Medical Microbiology & Infectious Diseases, Erasmus MC University Medical Center Rotterdam, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands
* Corresponding author. Tel: +31-104633510; Fax: +31-104633875; E-mail: a.ott{at}erasmusmc.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: Azithromycin is used to modulate exuberant inflammatory response in patients with cystic fibrosis (CF). The purpose of this study was to determine the association between long-term use of azithromycin in CF patients and change over time in macrolide susceptibility of Staphylococcus aureus and Haemophilus spp.
Methods: The study was performed at the Erasmus MC–Sophia Children's Hospital. CF patients' sputum cultures were obtained at routine visits and at pulmonary exacerbations. All cultures between January 1999 and March 2004 were included. Antibiotic susceptibility of S. aureus and Haemophilus spp. was tested routinely. Susceptibility was compared with isolates from sputum of non-CF patients. Logistic regression was used to analyse the association between azithromycin use and resistance, adjusting for age, Pseudomonas carriage and time-trends.
Results: In March 2004 one-third of CF patients were on azithromycin maintenance treatment. S. aureus (715 isolates) and/or Haemophilus (537 isolates) were cultured in 141 of the 155 patients on one or more occasions. The study period was divided into octiles. Erythromycin resistance in S. aureus increased from 6.9 to 53.8% and clarithromycin resistance in Haemophilus spp. from 3.7 to 37.5%. Resistance but also isolation rates were strongly related to azithromycin use. Resistance of 3217 S. aureus control isolates remained stable and resistance of 3257 Haemophilus controls increased, although at a slower rate than CF isolates.
Conclusions: Over a 4 year period, azithromycin maintenance therapy in our CF population was associated with an increase in macrolide resistance in S. aureus and Haemophilus spp.
Keywords: clarithromycin , erythromycin , immunomodulation
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Cystic fibrosis (CF) is the most common recessively inherited disease in the Caucasian population. It affects all exocrine glands. Affected organs that are clinically most important are the lung, pancreas, liver and testis. CF lung disease is characterized by exaggerated host inflammatory response and chronic infection of the airways with, among others, Staphylococcus aureus, Haemophilus spp. and Pseudomonas aeruginosa.1 Infection and inflammation results in progressive bronchiectasis and, ultimately, in respiratory failure in most patients.2–4 Treatment consists of sputum mobilization and aggressive control of infection. The latter is mainly achieved by prescribing antibiotics targeting the most relevant bacteria involved, both during pulmonary exacerbations and as maintenance.5 Immunomodulatory treatment that reduces the inflammatory response is thought to be beneficial. However, various anti-inflammatory drugs such as ibuprofen and prednisolone have side effects and uncertain long-term efficacy.6,7 Macrolide antibiotics also exhibit direct immunomodulatory effects.8 Azithromycin, one of the macrolide antibiotics, was recently shown to have a role as maintenance treatment in CF, probably through its anti-inflammatory effects. Three clinical trials have assessed the benefits of long-term azithromycin use. All studies showed positive effects of azithromycin, such as an increase in FEV1 and quality of life, and a decrease in the need for antibiotics.9–11 Though these results are promising, these studies were small and of relatively short duration. In addition, long-term safety has not been established to date. Questions concerning development of antibiotic resistance, and the effect of drug accumulation in specific cells of the body, remain unanswered.
Four years after the introduction of azithromycin maintenance treatment in our CF centre, we noticed an increase in macrolide resistance of S. aureus and Haemophilus spp. The aim of the current study was to investigate the association between maintenance therapy with azithromycin in CF and change over time in macrolide susceptibility of S. aureus and Haemophilus spp.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Study period and population
The study period was from January 1999 through March 2004. The starting date was about when the first patient was put on azithromycin maintenance therapy. For analytical purposes the study period was divided into 8 periods of 240 days (octiles). The study population consisted of all patients with CF treated at the Erasmus MC–Sophia Children's Hospital, Rotterdam, The Netherlands, who had one or more sputum cultures during the study period.
Treatment with azithromycin
Patients were started on maintenance treatment with azithromycin when progression of clinical symptoms and/or worsening of pulmonary function was observed, and when the patient did not respond well to conventional therapy. Patients weighing more than 40 kg received a dose of 500 mg on Monday, Wednesday and Friday; those weighing less than 40 kg received a dose of 250 mg thrice weekly, as described by Saiman et al.10 In our centre maintenance therapy with other antibiotics is restricted to the use of flucloxacillin in the first year of life and to tobramycin for inhalation in almost all patients chronically infected with Pseudomonas.
Microbiology
During the observational period samples for culture were obtained at each routine check-up and at the time of a pulmonary exacerbation. In the present study only cultures from expectorated or aspirated sputum were taken into account. These comprise the majority of respiratory cultures in our CF patients. In our hospital cough swabs are not taken, a throat culture is performed for screening of Pseudomonas, but only in those children not expectorating sputum. Routinely, sputum from CF patients is inoculated on selective media to enhance detection of specific pathogens. Media used for detection of S. aureus and Haemophilus spp. are phenol mannitol salt agar and Haemophilus Chocolate agar® (bioMérieux, France), respectively. Gram stain, catalase test and the Staphaurex Plus® agglutination test (Abbott Murex, Chatillon, France) are used to identify S. aureus. Colonies suspected to be from Haemophilus spp. were characterized by X and V factor requirement and, in case of discrepancy, a confirmation test was performed with a VITEK Neisseria-Haemophilus Identification card (Vitek Systems, bioMérieux).
Disc diffusion (Haemophilus spp.) and VITEK susceptibility panels (S. aureus) in an automated system (bioMérieux) were used to determine antibiotic susceptibility. The three most commonly used macrolides are erythromycin, clarithromycin (both 14-membered ring: M14 macrolides) and azithromycin (M15 macrolide). If resistance develops, there is complete cross-resistance between all macrolides of M14 and M15 classes. Traditionally S. aureus susceptibility is tested against erythromycin. Haemophilus influenzae is not susceptible to erythromycin, but can be so to clarithromycin and azithromycin. We used clarithromycin to test macrolide susceptibility of Haemophilus spp. NCCLS breakpoints were used to determine susceptibility of the isolates.12 According to these criteria S. aureus with erythromycin minimal inhibitory concentration (MIC) of 
8 mg/L and Haemophilus spp. with clarithromycin disc zone 13 mm, corresponding to MIC 32 mg/L, are classified as resistant.
Measurements
The following patient characteristics were recorded: sex, age at time of culture (two decimals), and chronic Pseudomonas infection in March 2004, at the start of the retrospective data collection, and onset and duration of azithromycin use. Pseudomonas carriage was defined as having had at least three sputum cultures positive for Pseudomonas during the last two years, or fewer than three, if these were not followed by negative cultures. For each sputum culture performed during the study period, we recorded presence or absence of S. aureus and/or Haemophilus spp. Additionally, sputum cultures were marked positive for azithromycin use when the patient had started on, or still was on azithromycin treatment in the month prior to the sputum culture.
All S. aureus and Haemophilus spp. cultured in our laboratory during the study period from sputum of non-CF patients (including children and adults) were used as controls. We compared macrolide resistance over time of these isolates with those from CF patients.
Statistical methods
Proportions of S. aureus or Haemophilus spp. positive cultures and resistance rates among cultured isolates were calculated per octile. 
2-statistics were used to compare resistance between groups (per octile), and the 2-test for linear trend was used to test the significance of increasing resistance or decreasing isolation rates within a group (entire study period).
The association between azithromycin use by CF patients and the likelihood of isolating S. aureus and Haemophilus was analysed with logistic regression, adjusting for age, Pseudomonas carriage, and octile (time trend), which all may act as confounders. As increased resistance may only be due to reduced recovery of susceptible isolates, we also analysed the likelihood to isolate macrolide-resistant S. aureus and Haemophilus, adjusting for the same possible confounders.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
CF patients
Of the 156 patients treated at our CF centre, 155 had one or more sputum cultures performed during the study period, and were included in the following analyses. Mean age at 1 March 2004 was 11.7 (range 0.6–25.3) years. Of these patients 77 (50%) were male, and, of those with known CFTR genotype (n = 149), 96 (64%) were dF508 homozygotes, and 43 (29%) were heterozygote with one dF508 gene (Table 1). At the end of follow-up, 63 (41%) were categorized as carriers of P. aeruginosa. In January 1999 a first patient was prescribed azithromycin maintenance therapy. In the following 2 years five other patients were put on azithromycin. After publication of the first azithromycin studies 9–11 the number of patients on azithromycin maintenance therapy rapidly increased to a maximum of 50 (32%) in October 2003. By March 2004, 63 patients (41%) had during some period been on azithromycin maintenance therapy. The median duration of azithromycin treatment was 397 (range 5–1827) days. Starting in the summer of 2003 we discontinued azithromycin in some patients during the summer season, based on the lack of safety data on long-term azithromycin use, and in relation to the observed resistance problems.
|
Sputum cultures
The 155 CF patients had 2522 sputum cultures during the study period. The median number of cultures per patient was 11 (range 1–111).
S. aureus was cultured from 699 (28%) of the sputum samples, including 125 patients (81%) who thus were S. aureus positive on one or more occasions. Of the 699 positive cultures 16 contained two S. aureus phenotypes. Therefore the total number of S. aureus isolates came to 715. Of all patients with S. aureus, five (4%) were carriers of methicillin-resistant S. aureus (including two sib pairs).
A Haemophilus spp. was cultured from 524 (21%) of the sputum samples, and in 122 patients (78%). Of these samples 13 contained two phenotypes of Haemophilus spp., resulting in a total of 537 Haemophilus isolates.
If two S. aureus or Haemophilus phenotypes were detected in one sample, only one was counted in ensuing analyses on culture level, and in case of unequal macrolide susceptibility (one susceptible and one resistant) the culture was classified as being positive with a resistant phenotype.
The proportion of S. aureus positive cultures reduced over time from 33% in the first octile to 25% in the last octile (P for trend <0.001). However, Haemophilus positive cultures were evenly distributed among octiles (P = 0.74) (Table 2).
|
Comparison between isolates from CF patients with and without azithromycin use
From the fifth octile onward, sputum cultures of patients who were using azithromycin in the month prior to culture were significantly (P < 0.01) less often positive with S. aureus or Haemophilus than cultures of patients not on azithromycin maintenance therapy (Table 2).
Macrolide susceptibility of 607 S. aureus from CF patients not using azithromycin could be compared with 86 S. aureus isolates from patients on azithromycin therapy (Table 2). The same was possible with 461 Haemophilus spp. of CF patients not on azithromycin and 35 Haemophilus isolates of patients using azithromycin therapy. Erythromycin and clarithromycin susceptibility could not be determined in 6 of 699 (1%) S. aureus positive cultures, and in 29 of 524 (6%) Haemophilus positive cultures.
The proportion of cultures containing S. aureus resistant to erythromycin increased over time in the azithromycin group, as well as in the non-azithromycin group (Figure 1a). However, in the sixth, seventh, and eighth octile the proportion was significantly higher among those patients using azithromycin (all P < 0.002).
|
The proportion of cultures containing clarithromycin-resistant Haemophilus spp. rose in both groups as well (Figure 1b). The proportion of resistant isolates was significantly higher in the azithromycin group in the fifth and seventh octile (both P = 0.03).
Logistic regression analyses revealed that azithromycin use was significantly associated with reduced isolation of both S. aureus (odds ratio 0.4) and Haemophilus spp. (odds ratio 0.3), independent of Pseudomonas carriage, age (years), and time trend (octile) (Table 3). Also Pseudomonas carriage was associated with reduced isolation of both pathogens (P < 0.001). This analysis shows that there was no trend of reduced S. aureus recovery (P = 0.33), but a slight increase in Haemophilus recovery over time (odds ratio per octile 1.07, P = 0.007).
|
Azithromycin was significantly associated with the presence of macrolide-resistant S. aureus (odds ratio 2.0, P = 0.007), but not Haemophilus spp. (Table 3). This analysis also shows that resistance of isolates was significantly associated with Pseudomonas carriage, and that recovery of resistant S. aureus and Haemophilus spp. strongly increased over time (P < 0.001).
Discontinuation of treatment
Twenty-four of 63 patients who were on azithromycin maintenance treatment were discontinued from therapy within the study period, mostly due to the lack of safety data on long-term use, and because of the observed resistance problems. The proportion of macrolide-resistant S. aureus and Haemophilus spp. cultured from sputum of these patients remained high. Over the entire study period, 67% of S. aureus, and 60% of Haemophilus isolates of patients who had discontinued therapy more than a month before the culture was done, were resistant to macrolides.
Comparison between CF and non-CF isolates
Erythromycin susceptibility of 709 isolates of S. aureus cultured from CF patients' sputum was compared with the susceptibility of 3217 S. aureus control isolates cultured from the sputum of non-CF patients during the same period (Table 4).
|
The proportion of S. aureus isolates resistant to erythromycin in the CF group increased from 6.9% in the first octile to 53.8% in the eighth octile, whereas this proportion in the control sample remained constant (overall 6.4%) (Figure 2a). The proportion of resistant S. aureus was significantly higher in the CF than non-CF samples from the fifth octile onward.
|
Furthermore, we compared the clarithromycin susceptibility of 507 CF patient Haemophilus isolates with that of 3257 Haemophilus controls cultured from sputum of non-CF patients. In CF patients 64% of the isolates were identified as H. influenzae, versus 67% of controls. Clarithromycin resistance in Haemophilus from CF patients increased from 3.7% in the first octile to 37.5% in the last octile (Figure 2b). Also controls gradually became more resistant over time, from 9.4% in the first octile to 26.7% resistance in the last octile. Only in the seventh octile was the proportion of resistant Haemophilus isolates significantly higher in CF patients than in non-CF controls.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
From January 1999 through March 2004 we noticed a considerable increase in the proportion of S. aureus resistant to erythromycin, and Haemophilus spp. resistant to clarithromycin, cultured from sputum of CF patients. Although macrolide resistance increased both in isolates cultured from patients using azithromycin maintenance therapy and those not using this antibiotic, the increase was greater in those on azithromycin. Regression analyses showed that long-term use of azithromycin was associated with lower recovery rates of S. aureus and Haemophilus from sputum. Still, macrolide-resistant S. aureus were significantly more often isolated from patients using azithromycin.
Three mechanisms may be responsible for the increase in macrolide resistance. First, isolates with an intrinsic resistance for macrolides may surface as susceptible ones are eradicated. Second, resistance may be acquired through one- or multi-step mutation. And third, resistant isolates may be acquired through cross-infection from other patients under antibiotic pressure. The fact that lower rates of both bacteria were found in patients using azithromycin suggests that the antibiotic can successfully suppress these bacteria in CF lungs. But our data also show that among those using azithromycin, with time, rates of resistance increase. The increase in resistance among those not using azithromycin could be due to cross-infection, or due to increased prescriptions of macrolides by family practitioners of patients. We cannot prove cross-infection as no molecular typing of recovered isolates was performed. Almost half of the increase however turned out to be due to discontinued azithromycin use: patients who had stopped remained colonized with resistant strains, which, as we noticed, lasted for months. Numbers of patients who discontinued azithromycin were too low to analyse this category separately.
Several differences were noted between S. aureus and Haemophilus. The effect of azithromycin use on resistance appeared to be greater for S. aureus than Haemophilus spp., as suggested by regression analysis (Table 3) and clearly diverging lines in Figure 1a. Also, macrolide resistance in S. aureus controls did not increase over time in our hospital, whereas resistance in Haemophilus controls increased, albeit less so compared with the CF patients' isolates. The latter is suggestive of an overall increase in macrolide-resistant Haemophilus spp., which may be due to increasing macrolide use in specific patient populations.
Figures on antibiotic use and resistance in the Netherlands show a gradual increase of macrolide use in primary healthcare (especially azithromycin and clarithromycin) between 1999 and 2004 from 1.17 to 1.31 DDD/1000 inhabitant-days, and in hospitals from 2.2 to 2.4 DDD/100 patient-days.13,14 In the same period non-selected samples from hospitals cooperating with the Dutch surveillance of antimicrobial resistance showed an increase of erythromycin resistance in S. aureus from 6 to 9%.14
We considered it essential to compare our CF sputum isolates with those from sputum of non-CF patients, because isolates from other sources may differ in susceptibility pattern.14 One explanation for the smaller difference in resistance between CF and non-CF Haemophilus than S. aureus isolates is that the first species is more often than the second cultured from patients with other chronic lung conditions, and, therefore, may have been more exposed to macrolides as well. Resistance in Haemophilus and S. aureus of CF patients may in part have been due to other macrolides than azithromycin. In our patients clarithromycin is prescribed by the family doctor, however, this occurs only occasionally, and can probably not explain resistance in all other patients.
The consistent finding of fewer S. aureus or Haemophilus spp. isolation among Pseudomonas carriers (Table 3) is likely due to repression in the lung of these bacteria by a high Pseudomonas load, or due to overgrowth or growth reduction by Pseudomonas in the microbiology laboratory.
To our knowledge this is the first paper associating the development of macrolide resistance of S. aureus and Haemophilus spp. with long-term use of azithromycin as an immunomodulatory agent. Macrolide resistance of Streptococcus pneumoniae in relation with azithromycin therapy has been described,15 as well as macrolide resistance of S. aureus and Haemophilus spp. after long term use of clarithromycin.16
Previous studies demonstrated benefits of long-term azithromycin use in CF.9–11 Due to the retrospective design, the treatment effect could not be evaluated in this study. We show that long-term azithromycin use to modulate inflammatory responses in CF lung disease has a down side. Long-term azithromycin use in our CF population was associated with a substantial increase in macrolide resistance. How long this effect persists after discontinuing treatment remains to be investigated.
Long-term placebo-controlled studies on the safety of azithromycin maintenance therapy should include measures for development of resistance. Our findings warrant careful consideration of the risk and benefits of starting and continuing CF patients on azithromycin maintenance therapy.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
We did not receive any financial support for the study. We have no potential conflicts of interest.
|
Acknowledgements |
|---|
None.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
1. Govan JR, Nelson JW. Microbiology of lung infection in cystic fibrosis. Br Med Bull 1992; 48: 912–30.
2. Khan TZ, Wagener JS, Bost T et al. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995; 151: 1075–82.[Abstract]
3.
Armstrong DS, Grimwood K, Carlin JB et al. Lower airway inflammation in infants and young children with cystic fibrosis. Am J Respir Crit Care Med 1997; 156: 1197–204.
4.
Ramsey B. Management of pulmonary disease in patients with cystic fibrosis. N Engl J Med 1996; 335: 179–88.
5.
Schultz MJ. Macrolide activities beyond their antimicrobial effects: macrolides in diffuse panbronchiolitis and cystic fibrosis. J Antimicrob Chemother 2004; 54: 21–8.
6. Eigen H, Rosenstein BJ, FitzSimmons S et al. A multicenter study of alternate-day prednisone therapy in patients with cystic fibrosis. Cystic Fibrosis Foundation Prednisone Trial Group. J Pediatr 1995;126: 515–23.[CrossRef][Web of Science][Medline]
7.
Konstan MW, Byard PJ, Hoppel CL et al. Effect of high-dose ibuprofen in patients with cystic fibrosis. N Engl J Med 1995; 332: 848–54.
8. Siddiqui J. Immunomodulatory effects of macrolides: implications for practicing clinicians. Am J Med 2004; 117 Suppl 9A: 26S–29S.
9. Equi A, Balfour-Lynn IM, Bush A et al. Long term azithromycin in children with cystic fibrosis: a randomised, placebo-controlled crossover trial. Lancet 2002; 360: 978–84.[CrossRef][Web of Science][Medline]
10. Saiman L, Marshall BC, Mayer-Hamblett N et al. Azithromycin in patients with cystic fibrosis chronically infected with Pseudomonas aeruginosa: a randomized controlled trial. JAMA 2003; 290: 749–56.
11.
Wolter J, Seeney S, Bell S et al. Effect of long term treatment with azithromycin on disease parameters in cystic fibrosis: a randomised trial. Thorax 2002; 57: 212–6.
12. National Committee for Clinical Laboratory Standards. Performance Standards for Antimicrobial Susceptibility Testing—Ninth Informational Supplement: NCCLS Document M100-S9. NCCLS, Wayne, PA, USA, 1999.
13. SWAB. NethMap 2004—Consumption of antimicrobial agents and antimicrobial resistance among medically important bacteria in the Netherlands. Available at http://www.swab.nl/swab/swabcms.nsf/showfs/foreign (23 December 2005, date last accessed).
14. SWAB. NethMap 2005—Consumption of antimicrobial agents and antimicrobial resistance among medically important bacteria in the Netherlands. http://www.swab.nl/swab/swabcms.nsf/showfs/foreign (23 December 2005, date last accessed).
15.
Pihlajamäki M, Kaijalainen T, Huovinen P et al. Rapid increase in macrolide resistance among penicillin non-susceptible pneumococci in Finland, 1996–2000. J Antimicrob Chemother 2002; 49: 785–92.
16.
Berg HF, Tjhie JH, Scheffer G et al. Emergence and persistence of macrolide resistance in oropharyngeal flora and elimination of nasal carriage of Staphylococcus aureus after therapy with slow-release clarithromycin: a randomized, double-blind, placebo-controlled study. Antimicrob Agents Chemother 2004; 48: 4183–8.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. A. J. Crosbie and M. A. Woodhead Long-term macrolide therapy in chronic inflammatory airway diseases Eur. Respir. J., January 1, 2009; 33(1): 171 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hodge, G. Hodge, H. Jersmann, G. Matthews, J. Ahern, M. Holmes, and P. N. Reynolds Azithromycin Improves Macrophage Phagocytic Function and Expression of Mannose Receptor in Chronic Obstructive Pulmonary Disease Am. J. Respir. Crit. Care Med., July 15, 2008; 178(2): 139 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. McCormack, S. Bell, S. Senini, K. Walmsley, K. Patel, C. Wainwright, D. Serisier, M. Harris, and S. Bowler Daily versus weekly azithromycin in cystic fibrosis patients Eur. Respir. J., September 1, 2007; 30(3): 487 - 495. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Tramper-Stranders, C. K. van der Ent, S. A. M. Gerritsen, A. Fleer, J. L. L. Kimpen, and T. F. W. Wolfs Macrolide-resistant Staphylococcus aureus colonization in cystic fibrosis patients: is there transmission to household contacts? J. Antimicrob. Chemother., September 1, 2007; 60(3): 665 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C Bell and P. J Robinson Exacerbations in cystic fibrosis: 2 {middle dot} Prevention Thorax, August 1, 2007; 62(8): 723 - 732. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||