Long-term azithromycin was not associated with new NTM isolation in bronchiectasis or COPD

Design

Nationwide nested case-control study

Population

538 NTM cases and 10,709 matched controls with bronchiectasis or COPD in Denmark

Primary outcome

Incident respiratory NTM isolation

Effect

Bronchiectasis OR 1.31 (0.79-2.18); COPD OR 1.29 (0.78-2.13)

A nationwide Danish nested case-control study in Chest, published on 28 October 2026, examined whether azithromycin in the preceding year was associated with a first positive respiratory culture for non-tuberculous mycobacteria. It included 173 cases and 3,409 matched controls with bronchiectasis, and 365 cases and 7,300 controls with COPD.

Under the prespecified definition — at least 3,000 mg over two or more prescriptions, with a three-month lag — azithromycin was not associated with NTM isolation in bronchiectasis (OR 1.31, 95% CI 0.79 to 2.18) or COPD (OR 1.29, 0.78 to 2.13). In COPD, any single prescription was associated (OR 1.61), but this signal clustered within 90 days of the culture and disappeared when only definite isolates were counted. Other, non-macrolide antibiotics showed stronger associations than azithromycin.

The pattern suggests that unwell patients are both given antibiotics and sampled, rather than that azithromycin causes NTM. It is reassuring for long-term macrolide prophylaxis, though it does not remove the need to exclude NTM before starting.

• Continue long-term azithromycin for appropriately selected patients with bronchiectasis or frequent COPD exacerbations.
• Send sputum for mycobacterial culture before starting long-term azithromycin.
• Repeat mycobacterial cultures if symptoms or imaging worsen on treatment.
• Do not stop effective azithromycin prophylaxis on fear of NTM alone.

Why it matters

Fear of inducing NTM often holds clinicians back from macrolide prophylaxis that reduces exacerbations.

Don't overread it

Observational registry data; a weaker association with any COPD exposure was not fully explained.

The statistics, in plain English

Odds ratios of about 1.3 with intervals crossing 1 mean no clear association. The stronger signal close to the culture date points to testing bias: people being treated for a flare are more likely to have sputum sent.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.