Azithromycin for non-CF Bronchiectasis
Bronchiectasis — the permanently dilated, tortuous bronchi that can result after previous lung infections — is a frustrating problem for pulmonologists to treat, but not nearly as frustrating as it can be for patients to live with. People with bronchiectasis are plagued by chronic coughing, and many experience a steady decline in lung function and frequent exacerbations of their illness requiring hospital admission.
Bronchiectasis can be hard to see on chest X-rays unless it’s severe, and has probably been historically under-recognized. Owing to the increased use (or overuse) of CT scanning, bronchiectasis is being diagnosed more and more often, perhaps properly re-categorizing many people who were misdiagnosed with chronic bronchitis or some other etiology of chronic cough.
The treatments for bronchiectasis not due to cystic fibrosis fall squarely under the rubric of medical lore, with few-to-none supported by evidence from randomized trials (not that that means they don’t work). These include cycles of oral antibiotics, inhaled tobramycin or gentamicin, chest physiotherapy, bronchodilators and inhaled corticosteroids. Lacking credible evidence, though, pulmonologists and their patients with bronchiectasis have been stuck with a trial-and-error, faith-based approach to managing this sometimes debilitating condition.
With the publication of the EMBRACE trial in the August 18 2012 Lancet, Conroy Wong, Lata Jayaram, Toni Ashton et al have moved the dial a bit on the treatment of bronchiectasis. They found that oral azithromycin taken three times weekly reduced the rate of exacerbations in people with non-CF bronchiectasis by about two-thirds. In their small trial, though, they don’t seriously address the major specters of theoretical risk with chronic azithromycin therapy: macrolide resistance, hearing loss and sudden cardiovascular death.
What They Did
Authors randomized 141 patients in New Zealand with bronchiectasis on high-resolution CT scan and at least one respiratory exacerbation (requiring antibiotics) in the past year to receive either placebo or azithromycin 500 mg three times weekly for six months, followed by observation for six months off study drug. Patients were excluded if they had cystic fibrosis, allergic bronchopulmonary aspergillosis (APBA), colonization or disease with non-tuberculous mycobacteria (NTBM), or hypogammoglobulinemia (as these etiologies all have other indicated treatments), as well as “unstable arrhythmia”.
The main endpoint was respiratory exacerbations requiring treatment with antibiotics (as judged by the treating physician, who was blinded to which group patients were in). Spirometry and quality-of-life (SGRQ) were co-primary endpoints, with numerous secondary endpoints.
They did not perform audiology at baseline or at any time throughout the study; hearing loss was only recorded if patients complained of it spontaneously. They collected sputum samples for culture at baseline and 6 months, but did not routinely check for macrolide-resistant organisms.
What They Found
Patients taking azithromycin had fewer bronchiectasis exacerbations requiring antibiotic treatment during the 6-month study period (0.59 exacerbations per patient with azithromycin vs. 1.57 per patient taking placebo, rate ratio 0.38, 95% CI 0.26-0.54, p <0.0001).
Over the subsequent six months after stopping azithromycin, that protection against bronchiectasis exacerbations continued. The one-year exacerbation rate among those assigned to azithromycin prophylaxis was 1.58, vs. 2.73 in the placebo group — a rate ratio of 0.58 or a 42% relative risk reduction.
In absolute terms, over 12 months, azithromycin prevented 69 exacerbations among 70 patients (109 vs. 178 total exacerbations over 12 months).
The other endpoints were not statistically different between groups:
- No change in prebronchodilator FEV1.
- Non-statistically significant improvement in quality-of-life with azithromycin by 3 points on the SGRQ.
- No change the number of in patient-reported exacerbations (ones that a doctor did not prescribe antibiotics for).
Two patients in the azithromycin group (4%) were found to be carrying a newly-caught strain of macrolide-resistant Streptococcus pneumoniae at their 6-month checkups.
There were no adverse events specifically attributable to azithromycin, besides gastrointestinal complaints.
What It Means
This study is great news for people who suffer from frequent exacerbations or disabling symptoms of bronchiectasis, who finally have a treatment option with proven efficacy. Although many treating physicians were probably already trying chronic azithromycin at some point in their treatment algorithm to try to help these patients, it’s very helpful to have convincing evidence that azithromycin really works to prevent bronchiectasis exacerbations.
What about the cardiovascular risks of azithromycin? The authors point out that their tiny study and the previous small study showing that azithromycin prevents COPD exacerbations “did not show an increased risk of death or cardiovascular adverse events.” But of course, they wouldn’t be expected to. The estimated excess risk for sudden cardiac death from a single course of azithromycin in the NEJM study was on the order of 1 in 20,000 (maybe as high as 1 in 4,000 in people with cardiovascular disease), and a huge observational database was necessary to detect the signal. The implication that randomized trials this small have the power to demonstrate safety is surprising to hear from such respected scholars, although they acknowledge that “further studies are needed to define the cardiovascular risks of azithromycin.”
In a recent NEJM article advocating azithromycin to prevent COPD exacerbations in some patients, those authors recommended stringent criteria be applied, including baseline ECG, repeated audiology monitoring, exclusion of anyone with cardiovascular disease, and other restrictions.
Wong C et al. Azithromycin for prevention of exacerbations in non-cystic fibrosis bronchiectasis (EMBRACE): a randomised, double-blind, placebo-controlled trial. Lancet 2012;380:660-667.