Azithromycin didn't ease preschool wheezing, even when bacteria were present

Azithromycin, a broad antibiotic often reached for when a wheezing toddler lands in the emergency department, did not ease those children's symptoms any more than a dummy pill — and that held true even in the children whose noses tested positive for the bacteria the drug was meant to hit [s1]. The trial that showed this, called AZ-SWED, was stopped early for futility once an interim analysis made clear the antibiotic was not working [s1]. It is a clean negative result in a setting where antibiotics are prescribed constantly, often on the hope that a hidden bacterial infection is driving the attack.

Wheezing illnesses are a leading cause of hospitalisation in preschool-age children, and clinicians frequently treat them with antibiotics [s1]. The rationale is not baseless: observational studies have found three pathogenic bacteria — Streptococcus pneumoniae, Moraxella catarrhalis, and Haemophilus influenzae — more often in the nasopharynx of children who wheeze repeatedly than in those who do not [s1]. AZ-SWED was designed to turn that association into a test of cause and effect: if these bacteria contribute to the flare, then clearing them should shorten it.

What the trial did

Researchers with the Pediatric Emergency Care Applied Research Network ran the multicentre trial and randomly assigned children aged 18 to 59 months who arrived at an emergency department with a moderate-to-severe wheezing episode to receive either azithromycin at 12 mg per kilogram once daily or a matching placebo for five days [s1]. Crucially, the team measured each child's nasopharyngeal bacteria and analysed the two groups separately: a "positive cohort" carrying one or more of the target pathogens, and a "negative cohort" that did not [s1]. This split let them ask not just whether the antibiotic worked on average, but whether it worked in exactly the children it should help most.

The primary outcome was the sum of scores on the Asthma Flare-up Diary for Young Children over five days, a parent-completed measure that runs from 5 to 35, with higher numbers meaning more severe wheezing symptoms [s1]. Secondary measures included time spent in the emergency department, length of any hospital stay, and return visits within 72 hours [s1]. The trial was funded by the National Heart, Lung, and Blood Institute — a public agency, not a drug company [s1].

What it found

Among the 840 children randomised, 521 tested positive for the target bacteria [s1]. The data and safety monitoring board stopped the trial for futility after a planned interim analysis [s1]. In the positive cohort, the median symptom score was 9.59 (interquartile range 7.29 to 12.60) with azithromycin versus 9.72 (7.66 to 12.17) with placebo — a difference that was not significant (P=0.70) [s1]. In the negative cohort the picture was the same: 9.30 (6.97 to 11.62) versus 9.10 (7.19 to 11.45), P=0.69 [s1]. Secondary outcomes, including emergency-department and hospital length of stay and return visits, appeared similar between groups in both cohorts [s1].

The one thing the antibiotic clearly did was kill bacteria. In the positive cohort, the target pathogens cleared in 58.7% of children given azithromycin versus 11.4% given placebo [s1]. That gap is the crux of the story: the drug did its microbiological job, yet the children were no better for it. Rates of adverse events and of newly detected bacterial resistance were similar between the groups [s1].

How to read it

The value of AZ-SWED is that it breaks a plausible chain of reasoning. It was reasonable to think that if certain bacteria are enriched in wheezy children, and an antibiotic clears those bacteria, then the antibiotic should shorten the flare. The first two links held; the last one did not. Clearing the bacteria changed nothing about the symptoms, which points to wheezing episodes being driven by something else — most often a respiratory virus and the airway inflammation it triggers — with the bacteria along for the ride rather than steering [s1].

For parents, the practical translation is that a course of azithromycin is unlikely to make a wheezing toddler recover faster, even when a swab finds bacteria. Antibiotics are not free of cost: they carry side effects and feed antimicrobial resistance, the slow erosion of the drugs' usefulness across the whole population. A well-run negative trial like this one is exactly the kind of evidence that lets clinicians hold back a prescription with confidence. Related coverage has examined why antibiotics do not help ordinary colds and how overuse breeds resistance, a bacterial-lysate product tested for preventing preschool wheezing, and what the evidence says about reliever inhalers in asthma.

What to watch

The finding does not settle every question about antibiotics and childhood wheeze — different drugs, longer courses, or specific subgroups could behave differently — but the burden of proof now sits with anyone claiming a benefit. The larger task is untangling which wheezing children are heading toward asthma and which are simply reacting to a passing virus, a distinction that better biomarkers, not antibiotics, will have to draw [s1].

This article describes research and is not medical advice. Decisions about antibiotics for a wheezing child are for treating clinicians.

Sources

• Azithromycin for Preschoolers with Wheezing in the Emergency Department — New England Journal of Medicine, 18 May 2026