• cystic fibrosis
• macrolide antibiotics
• azithromycin

There is growing interest in the potential use of macrolide antibiotics as anti-inflammatory agents in cystic fibrosis. This stems from the dramatic success of long term erythromycin in the treatment of diffuse panbronchiolitis (DPB), a condition with a high prevalence in Japan but rare elsewhere.^1–3 Clinically, DPB exhibits some similarities to cystic fibrosis including chronic productive cough, exertional dyspnoea, chronic sinusitis, mucoid Pseudomonas aeruginosa colonisation, and bronchiectasis. The introduction of erythromycin as a treatment for DPB has had a dramatic impact on mortality, increasing 10 year survival from 12.4–21.9% to over 90% in those colonised with P aeruginosa.^3,4 Similar success has been reported with clarithromycin, roxithromycin, and azithromycin.^1,3 While the aetiology of both conditions may be very different, it is the similarities which beg the question “do the macrolides have a role in the treatment of cystic fibrosis?”

The macrolide antibiotics are an intriguing group of drugs with both anti-inflammatory and antibacterial properties.⁴ Their mode of action in DPB is thought to be mediated by mechanisms other than antibacterial as the effect occurs below the minimum inhibitory concentration required for bacteria such as Haemophilus influenzae and P aeruginosa.^1,3

There are several theoretical reasons why the macrolides could modulate the disease process in cystic fibrosis. Firstly, airway inflammation, as in DPB, is recognised as a major factor in the pathogenesis of cystic fibrosis lung disease.^5–7 Anti-inflammatory drugs such as high doses of non-steroidal anti-inflammatory agents and prednisolone have been shown to slow the decline of lung function in patients with cystic fibrosis.^8–10 Several studies suggest that the macrolides also possess important anti-inflammatory activity which appears to be mediated by an inhibition of neutrophil chemotaxis, reduction of neutrophil elastase, and modification of pro-inflammatory cytokines with suppression of interleukin (IL)-1β, IL-6, IL-8, and tumour necrosis factor (TNF)-α production.^1,2,4,11 Secondly, they reduce sputum viscoelasticity and airway adhesion of P aeruginosa.^2,12,13 Certain macrolides have the innate ability to increase the killing of mucoid P aeruginosa, a mechanism that may be mediated by their ability to disrupt the integrity of the protective biofilm and impair the transformation of non-mucoid P aeruginosa to the more virulent mucoid phenotype.^14–16

The clinical evidence to support the use of macrolides in the treatment of cystic fibrosis is poor. Most of the studies have only been published in abstract form and are usually anecdotal with small numbers of patients. Frederiksen et al reported a larger randomised, double blind, placebo controlled, crossover study of the effect of twice daily clarithromycin in cystic fibrosis.¹⁷ Various parameters were measured including pulmonary function but, unexpectedly, 20 of 41 patients were excluded from the study so that no conclusions could be drawn. Importantly, failure to complete the study was not related to the active arm.

“Treatment with azithromycin was associated with significantly fewer courses of intravenous antibiotics, maintenance of lung function, reduction in median CRP levels, and improvement in quality of life scores”

In this issue of Thorax Wolter et al report their findings of the first published prospective, randomised, placebo controlled trial investigating the clinical effect of macrolides in the treatment of cystic fibrosis.¹⁸ A total of 49 adults with cystic fibrosis completed the 3 month trial of 250 mg azithromycin versus placebo. Treatment with azithromycin was associated with significantly fewer courses of intravenous antibiotics, maintenance of lung function, reduction in median C reactive protein (CRP) levels, and improvement in quality of life scores. While there was no difference in baseline microbiology, Staphylococcus aureus was isolated from 41.3% of patients at the start of the study. This suggests that some of the clinical response seen in the azithromycin group may have been mediated through the antibacterial activity of the drug. Similar results have been reported in children. In a non-randomised open labelled study Pirzada et al compared the effect of 250 mg azithromycin in 18 children with cystic fibrosis and 18 age and sex matched controls over a mean of 0.78 years.¹⁸ The azithromycin treated group showed significant improvement in lung function and weight gain. The drug was well tolerated and no significant side effects were observed.

In the only other study to be formally published, Jaffe et al reported their findings from an open study of seven children with cystic fibrosis given 250 mg azithromycin for more than 3 months.²⁰ Although azithromycin was associated with a significant increase in lung function, the results are difficult to interpret.

While the study by Wolter et al supports the potential role of macrolides in the treatment of cystic fibrosis, larger double blind, placebo controlled trials are needed which can differentiate between the anti-inflammatory and antibacterial properties of these agents. With the potential ability of the macrolides to alter the complex bacteria/epithelial/biofilm interaction, it is possible that they may have a role both in reducing the incidence of new P aeruginosa colonisation and improving conventional early eradication treatment.²¹