Impact of erythromycin discontinuation owing to supply disruption on exacerbations in non-cystic fibrosis bronchiectasis: A retrospective cohort study.

BACKGROUND

Long-term macrolide therapy reduces exacerbations in bronchiectasis; however, evidence regarding the outcomes after treatment discontinuation is limited. This study aimed to evaluate the effect of erythromycin discontinuation caused by nationwide supply disruption on exacerbations of bronchiectasis.

METHODS

This retrospective observational cohort study enrolled adults with non-cystic fibrosis bronchiectasis who had received long-term erythromycin therapy (≥3 months) and discontinued treatment between August 2024 and September 2025 because of supply disruption. Exacerbations during the 12 months before and after discontinuation were compared. Patients were classified according to post-discontinuation management: without replacement (No-macrolide group) or switching to alternative macrolide therapy (Alternative macrolide group). The primary outcome was the time to first exacerbation, and secondary outcomes included the annual exacerbation frequency and factors associated with post-discontinuation exacerbations.

RESULTS

Among 135 patients, erythromycin discontinuation was associated with a significantly shorter time to first exacerbation than the pre-discontinuation period (log-rank P = 0.001). Discontinuation was independently associated with earlier exacerbation (hazard ratio 1.90, 95% confidence interval 1.12-3.23). This association was evident in the No-macrolide group but not in the Alternative macrolide group. The exacerbation frequency increased after discontinuation in the No-macrolide group. Higher baseline C-reactive protein levels, using biologics, and a prior history of exacerbations were independently associated with subsequent exacerbations.

CONCLUSIONS

Discontinuation of long-term erythromycin therapy was associated with more frequent and earlier exacerbations over the subsequent year, particularly without alternative macrolide therapy. These findings highlight the clinical consequences of macrolide interruption and the importance of appropriate substitution and monitoring in high-risk patients.