Macrolides for Exacerbation Prevention in Bronchiectasis


This is Andy Shorr from Washington, DC, with a pulmonary and critical care literature update. I would like to bring to your attention the EMBRACE trial, which was published in August in the Lancet. This study focused on preventing exacerbations from non-cystic fibrosis bronchiectasis. Bronchiectasis is recognized as an increasingly common condition, particularly in the era of the CT scanner.

Many patients without cystic fibrosis have some degree of bronchiectasis, whether it is related to interstitial lung disease, chronic obstructive pulmonary disease (COPD), or is idiopathic. Bronchiectasis is an inflammatory condition of the airways. Its hallmark is lung loss in terms of lung function decline, recurrent exacerbations, and airway colonization, just as we see in cystic fibrosis. Very few studies in non-cystic fibrosis bronchiectasis have been conducted, and a paucity of randomized controlled trials have considered management of this syndrome.

These investigators focused on the role of a macrolide -- azithromycin -- in preventing exacerbations of bronchiectasis. Azithromycin has been studied for the prevention of exacerbations in cystic fibrosis, and recently it has been studied for the prevention of exacerbations in COPD. Wong and colleagues studied a group of patients who had non-cystic fibrosis bronchiectasis and randomly assigned them to either azithromycin 500 mg 3 times weekly or placebo. The study was conducted at multiple centers and it was double-blind. This was a very high-quality study. The investigators studied 141 patients, 71 in the azithromycin group and 70 in the placebo group.

The primary endpoint was time to exacerbation and rate of event-related exacerbations, which they defined using the criteria for COPD exacerbation (at least 2 cardinal manifestations -- sputum production and purulence, cough, or dyspnea). The exacerbation also had to be treated with antibiotics by the patient's primary care provider, but it did not have to require hospitalization. The decision to give antibiotics was left to the clinicians, who were blinded to patient group (placebo or azithromycin).

The population was as expected. Their mean FEV₁ (forced expiratory volume in 1 second) was in the 67% range. Very few patients were on continuous oral steroids, and most of them were nonsmokers or, at most, distant smokers. The study met its primary endpoint. There were significantly fewer exacerbations in the group taking the macrolide. The event rate was reduced by about two thirds, and not only was the proportion of exacerbations per patient per month in the study reduced, but the total burden of exacerbations was reduced in terms of frequency and time to first exacerbation.

The study continued out to 12 months, and at that time they saw a reduction in exacerbations as well. When they looked at other markers, azithromycin (even though it was preventing exacerbations) did not improve quality of life and/or lung function. There were changes directly favoring the macrolide, but they were not statistically significant.

Azithromycin was generally very well tolerated. There were no signals of adverse events or any cardiovascular issues, and surprisingly, no concern was raised about hearing loss or effects that have been reported with long-term macrolide use in the past.

The primary endpoint for this study was an exacerbation requiring a course of antibiotic therapy. So, the question you have to ask yourself is: What is the value of preventing an exacerbation that might require antibiotic therapy if it doesn't necessarily lead to hospitalization or lost work days?

A large number of patients were taking a macrolide, so the question isn't about the internal validity of the data. The trial was well executed, it met its primary endpoint, and it was internally consistent. It's the external validity and the generalizability to the non-cystic fibrosis bronchiectasis population that must be questioned. They studied a very mildly affected population. What is the minimal clinically important difference that we need to see to change our management?

If you think about it, in this study they put more than 70 patients on a macrolide 3 times weekly for a year -- more than 5000 doses of macrolide -- to prevent, in this case, approximately 80 exacerbation events, which are associated with a course of therapy for 7-10 days. So, they used a ton of antibiotics to prevent a modicum of antibiotic use. Is that a reasonable tradeoff, given what we know about antibiotic resistance?

Wong and colleagues clearly showed that focusing on a secondary endpoint, the macrolide, achieved what it was supposed to in terms of having systemic anti-inflammatory properties, so, biologically, the mechanism is relatively clear. The question is whether it is a valuable tradeoff. This is the same issue that we have with the data on COPD exacerbations and prevention with macrolides. Clearly, the study in the New England Journal of Medicine also met its primary endpoint, but it didn't reduce hospitalizations. It just reduced a little bit of antibiotic use at the cost of a lot of antibiotic use.

That study was in a population with very severe COPD, and you wonder whether the real mechanism was its effect on the bronchiectasis in that cohort in addition to the mechanisms that were seen in this non-cystic fibrosis population.

Regardless, we have made a major advance in our understanding of the anti-inflammatory properties of macrolides. The question we have to answer now is whether it is worthwhile in terms of the risk for antibiotic resistance. Neither of the studies (the non-cystic fibrosis bronchiectasis study or the COPD study) provided us with a lot of long-term safety data, which is concerning in light of all the risks that have been described in the studies about cardiovascular toxicity with macrolides.

So, you need to weigh the risks and benefits. I urge you to look at this article in the August 18 issue of the Lancet. It's a very well-done study and an important study because we have very few studies in bronchiectasis. It would be harder to do a larger study that is very thought-provoking. But as with many studies, this one actually raises more questions than it answers.

This is Andy Shorr from Washington, DC.