Conference Coverage

Azithromycin for COPD exacerbations may reduce treatment failure



– In patients with a chronic obstructive pulmonary disease (COPD) exacerbation, initiating azithromycin at the time of hospitalization provided improvement in a variety of outcomes at 90 days, including risk of death, according to a placebo-controlled trial presented as a late-breaker at the annual congress of the European Respiratory Society.

Dr. Wim Janssens, division of respiratory medicine, University Hospital, Leuven, Belgium Ted Bosworth/MDedge News

Dr. Wim Janssens

In patients with COPD, “azithromycin initiated in the acute setting and continued for 3 months appears to be safe and potentially effective,” reported Wim Janssens, MD, PhD, division of respiratory medicine, University Hospital, Leuven, Belgium.

The phrase “potentially effective” was used because the primary endpoint, which was time to treatment failure, fell just short of statistical significance (P = .053), but the rate of treatment failures, which was a coprimary endpoint (P = .04), and all of the secondary endpoints, including mortality at 90 days (P = .027), need for treatment intensification (P = .02) and need for an intensive care unit (ICU) admission (P = .003), were significantly lower in the group receiving azithromycin rather than placebo.

In a previous trial, chronic azithromycin therapy on top of usual care in patients frequently hospitalized for COPD was associated with a reduction in the risk of exacerbations and an improvement in quality of life (N Engl J Med. 2011;365:689-98). However, Dr. Janssens explained that this strategy is not commonly used because it was associated with a variety of adverse events, not least of which was QTc prolongation.

The study at the meeting, called the BACE trial, was designed to test whether azithromycin could be employed in a more targeted approach to control exacerbations. In the study, 301 COPD patients hospitalized with an acute exacerbation were randomized within 48 hours of admission to azithromycin or placebo. For the first 3 days, azithromycin was administered in a 500-mg dose. Thereafter, the dose was 250 mg every second day. Treatment was stopped at 90 days.

The primary outcome was time to treatment failure, a novel composite endpoint of any of three events: the need for treatment intensification, the need for step-up hospital care (either ICU admission or hospital readmission), or death by any cause. The two treatment arms were also compared for safety, including QTc prolongation.

The treatment failure rates were 49% in the azithromycin arm and 60% in the placebo arm, producing a hazard ratio (HR) of 0.73. Although this outcome fell short of significance, Dr. Janssens suggested that benefits over the 90 days of treatment are supported by the secondary outcomes. However, he also cautioned that most relative advantages for azithromycin over placebo were found to dissipate over time.

“The maximum separation between the azithromycin and placebo arms [for the primary outcome] occurred at 120 days or 30 days after the medication was stopped,” Dr. Janssens reported. After this point, the two arms converged and eventually overlapped.

However, the acute benefits appeared to be substantial. For example, average hospital stay over the 90-day treatment period was reduced from 40 to 10 days (P = .0061), and the ICU days fell from 11 days to 3 days in the azithromycin relative to the placebo group. According to Dr. Janssens, the difference in hospital stay carries “important health economic potential that deserves further attention.”

Of the three QTc events that occurred during the course of the study, one was observed in the placebo group. There was no significant difference in this or other adverse events, according to Dr. Janssens.

It is notable that the design for the BACE trial called for 500 patients. When enrollment was slow, the design was changed on the basis of power calculations indicating that 300 patients would be sufficient to demonstrate a difference. It is unclear whether a larger study would have permitted the difference in the primary endpoint to advance from a trend.

Dr. Janssens reports no conflicts of interest relevant to this study.

Next Article:

   Comments ()