Azithromycin to Prevent Asthma Exacerbations: What AMAZES Us and What Doesn’t

Written By

Reviewed By


Gibson PG, Yang IA, Upham JW, et al. Effect of azithromycin on asthma exacerbations and quality of life in adults with persistent uncontrolled asthma (AMAZES): a randomised, double-blind, placebo-controlled trial. Lancet. 2017;390:659-668.

Asthma costs the US health care system roughly 50 billion dollars annually.1 And patients with uncontrolled severe asthma (> 2 exacerbations per year) incur far more cost than patients with persistent asthma without exacerbations ($5174 vs $1775 in 2013).2 Despite good adherence with high-dose inhaled corticosteroids (ICS) and concomitant long-acting beta agonists (LABA), millions of people continue to experience exacerbations.3  What more can patients and clinicians do to reduce the risk of exacerbations?  Does the routine use of antibiotics reduce the frequency of exacerbations?  This is the question the recently published AMAZES study attempted to answer.4


Asthma exacerbations are often triggered by allergens or viral infections, and far less frequently by bacteria. Macrolide antibiotics have been shown to improve asthma symptoms, but previous systematic reviews have been inconclusive or did not examine exacerbation frequency as the primary endpoint.5 Given their theoretical anti-inflammatory, antiviral, and antibacterial properties, macrolide antibiotics may be useful in both eosinophilic and non-eosinophilic asthma phenotypes; the latter is generally insensitive to inhaled corticosteroids and adjunctive treatments like omalizumab. The Global Initiative for Asthma Guideline recommends against the use of antibiotics for the acute treatment of asthma exacerbations unless there is strong evidence of a bacterial infection.  Moreover, these guidelines do not mention the prophylactic use of antibiotics in patients with uncontrolled asthma.6


AMAZES aimed to determine the efficacy and safety of oral azithromycin 500 mg three times a week as add-on therapy in adult patients with asthma. The study was a 48-week, multi-site, randomized, double-blind, placebo-controlled study conducted in Australia. Patients were included if they had objective evidence of asthma with uncontrolled symptoms (defined as asthma control score [ACQ6] > 0.75) despite ICS and LABA maintenance therapy. Patients with recent exacerbations, infections, or changes in asthma medication dose within the previous 4 weeks were excluded, along with current smokers and those with diagnosed parenchymal lung disease, prolonged QTc (>480 ms), or hearing impairment.  Sputum, throat and nose cultures were performed at randomization and end of treatment (week 48). Patients were assessed in clinic at weeks 6, 12, 24, 36, 48, and 52.


The primary efficacy outcome was the total number of moderate and severe asthma exacerbations (See Table 1) and asthma quality of life (QoL).  Asthma QoL was assessed using the asthma quality of life questionnaire (AQLQ).  Notable secondary outcomes were the number of antibiotic courses for a respiratory infection, microbial assessments, and adverse events. Statistics were done on a modified intent-to-treat basis with last observation carried forward.


Table 1: Moderate and severe exacerbations defined

Moderate Exacerbation

Severe exacerbation

Any increase in inhaled corticosteroids

At least 3 days of oral corticosteroids (dose >10 mg/d)

Addition of antibiotics with deterioration in asthma symptoms

At least 3 days of increased oral corticosteroid dose from maintenance dose

Increase in Beta-2 agonist for at least 2 days

Asthma-specific hospitalization

Emergency room visit NOT requiring oral corticosteroids

Emergency room visit requiring oral corticosteroids


A total of 420 patients were randomized (N=213 treatment; N=207 placebo).   The majority of patients were female with an average age of 60 years old. Roughly 38% were former smokers with an average of 7.5 pack-year history. At baseline, 98% of patients were on a medium or high dose ICS with a LABA. There were no significant differences between the intervention and control groups.


Azithromycin add-on therapy led to a statistically significant reduction in the mean number of combined moderate and severe asthma exacerbations compared to placebo (1.86/person-year versus 1.07/person-year; p<0.0001). Severe asthma exacerbations were also reduced with azithromycin use (1.07/person year vs. 0.61/person year, p<0.002).  Furthermore, the total number of patients who experienced an exacerbation was significantly lower in the azithromycin group (127 versus 94; p<0.0001, NNT=6). The reduction in exacerbations with azithromycin was significant in both eosinophilic and non-eosinophilic sub-types. Asthma-related QoL improved in both the azithromycin and control groups across all domains of the AQLQ; however, only the symptom domain in the azithromycin group met the threshold for clinical significance. A sensitivity analysis did not change the QoL results.


In terms of adverse events, there was an increase in resistant bacteria in sputum cultures of the azithromycin group compared to placebo at the end of the study (49% versus 29%; p=0.062).  While this did not quite reach statistical significance, it might be clinically important. Matched sputum samples from baseline to end of treatment showed four resistant bacteria before and after treatment with placebo, and three resistant bacteria at baseline compared to six at end of treatment with azithromycin (p=0.38).  Other adverse events are shown in Table 2.


Table 2: Notable Adverse Events


Azithromycin (N=213)

Placebo (N=207)

P Value

Serious Adverse Events




Withdrawal due to ADR








Prolonged QTc




Hearing loss/tinnitus




Infections (self-reported)




Participants requiring  ≥ 1 antibiotic course




NS = not significant


Strengths of this study included the use of clinical outcome measures (exacerbations) supplemented by the patient-perceived quality of life.  Additional strengths were the size, length, and funding source (Australian government) of the study. The AMAZES trial is the largest and longest study exploring macrolide use for the prevention of asthma exacerbations.  Previous studies used lower weekly doses of azithromycin, thus this trial establishes a target dose.7,8 AMAZES also reported adherence rates.


Despite its strengths, there are several limitations that should be considered.  First, the results cannot be generalized to children, younger adults (<50 years of age), or current tobacco users.  These findings might not be generalizable to the United States given the possible difference in microbial resistance patterns and genetics when compared to Australians.  A subset of patients in the study may have Asthma-COPD Overlap Syndrome (ACOS). Given the older patient population, average pre-bronchodilator FEV1/FVC (68%), the number of former smokers, and high percentage of patients using long-acting muscarinic antagonists (LAMAs) at baseline, it seems plausible that a significant proportion of these patients had ACOS.9 This may skew the results in favor of the intervention as macrolide therapy is known to prevent COPD exacerbations and is frequently used in that population. The study was not powered to determine changes in antimicrobial resistance and the total number of sputum samples was small; thus, the impact of chronic azithromycin treatment on bacterial resistance requires further investigation.


The sub-group analysis needs to be viewed with caution due to sample size and repeated analysis (meaning the trial was not powered to determine which subsets of patients benefited from the intervention).  Lastly, while the treatment duration (48 weeks) is reasonably long, we don’t know with certainty what the long-term consequences of chronic macrolide use, potentially years and years, might be.


The clear reduction in asthma exacerbations is a compelling reason to consider azithromycin as an option in the arsenal of add-on therapy in uncontrolled asthma. Beyond lowering costs to the health system by reducing exacerbations, azithromycin is less expensive when compared to other add-on options.  For example, the estimated drug cost of omalizumab is around $10,000-15,000 annually, compared to approximately $500 for azithromycin.10


Treatment with oral azithromycin 500 mg three times a week significantly decreased asthma exacerbation rates in patients with uncontrolled asthma.  Moreover, in carefully selected patients who were screened for hearing decrements and prolonged QTc, there were a limited number of adverse events. However, the long-term impact of chronic azithromycin therapy on antimicrobial resistance is unknown. Thus, we cannot recommend azithromycin in all patients with uncontrolled asthma - additional studies are needed to confirm these results, as well as help define the subset(s) of asthma patients who most benefit.  Would you consider azithromycin as add-on therapy in patients with severe, uncontrolled asthma patients who are already using high-dose ICS and LABA therapy? Tell us what you think.


Acknowledgements:  The authors would like to thank Christine Sorkness, PharmD and Susanne Barnett, PharmD for their guidance.