Intrapartum azithromycin did not improve neurodevelopment after birth asphyxia

Design

neurodevelopmental follow-up of a randomised, placebo-controlled trial, masked assessors

Population

403 children born at 34 weeks or later with concern for asphyxia, six sites in five countries including two in India

Primary outcome

Bayley-III cognitive composite score at 24 ± 1 months corrected age

Effect

90.9 ± 11.2 with azithromycin against 90.9 ± 11.7 with placebo; mean difference 0.29 (95% CI -1.77 to 2.34)

A-PLUS randomised labouring women to a single 2 g oral dose of azithromycin or placebo, across sites in India, Pakistan, Zambia, the Democratic Republic of Congo and Guatemala. This follow-up took the children born at 34 weeks or later who had concern for asphyxia - a five-minute Apgar under 7, or the need for bag-and-mask ventilation - and assessed them at a corrected age of 24 months with the Bayley Scales of Infant and Toddler Development, third edition.

Of 529 eligible mother-child pairs screened, 403 completed follow-up: 197 azithromycin, 206 placebo. The cognitive composite score was identical in the two arms (90.9 ± 11.2 against 90.9 ± 11.7; mean difference 0.29, 95% CI -1.77 to 2.34). Language and motor composite scores showed no difference, nor did any of the five domains of the Ages and Stages Questionnaire. A subgroup analysis splitting sub-Saharan Africa from South Asia found nothing either. Examiners were masked.

The result is a clean negative, and useful because the hypothesis was a reasonable one: intrapartum azithromycin reduces maternal sepsis, and it was plausible that reducing infection-related inflammation around the time of asphyxia would protect the brain. It does not. Two of the six sites were Indian, so this is directly relevant evidence rather than evidence to be extrapolated - a point worth making, because most neurodevelopmental follow-up data in this field are not.

• Do not offer intrapartum azithromycin as neuroprotection, and do not let it displace anything that is
• Keep the case for intrapartum azithromycin where the trial actually made it - maternal infection, not neonatal brain injury
• Continue to prioritise the measures with evidence after perinatal asphyxia: resuscitation quality, temperature control, and therapeutic hypothermia where available and appropriate
• Note that the mean cognitive score of about 91 in both arms is below the test norm of 100, which is the more useful observation for these populations

Why it matters

It closes off a plausible and cheap neuroprotective intervention, so the effort goes back to resuscitation and cooling.

Don't overread it

This addresses neurodevelopment after asphyxia only - it says nothing about azithromycin's established effect on maternal infection in labour.

The statistics, in plain English

A mean difference of 0.29 points with a confidence interval of -1.77 to 2.34 excludes any effect of clinical size in either direction, which makes this an informative negative rather than an underpowered one. A negative result on a primary outcome is also more trustworthy than a positive subgroup finding would have been, and the authors report the regional subgroup as null rather than mining it.

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.