JOURNAL CLUB · ARTICLE OF THE WEEK
Delivery-room CPAP in late-preterm and term infants: a review
Chin Y, Hill M, Sett A, Thomas N, Razak A. J Paediatr Child Health. 2026. doi:10.1111/jpc.70516. A systematic review and meta-analysis (open access).
TRIALS SAY HELP -
THE EVIDENCE IS WEAK
Observational data flagged harm
DR-CPAP was linked to more pneumothorax (adjusted OR ~24; moderate certainty), more surfactant use (OR 5.7) and more NICU admission (OR ~15).
But the RCTs disagreed
In the 2 RCTs there were no pneumothoraces in either group, and NICU admission was lower with DR-CPAP (OR 0.24).
Why - confounding by indication
In observational studies the sicker baby is the one who gets CPAP, so it looks harmful. Randomisation removes that bias.
Certainty is low
The RCT evidence is very low certainty and the trials are small; well-designed trials are urgently needed.
WHY THIS MATTERS
- DR-CPAP for well-ish late-preterm and term babies is being used more (see last week's PLaNT trial)
- If it prevents NICU admission that is a real win; if it causes pneumothorax that is real harm
- Observational data can badly mislead when the sicker baby is the one treated
- It shows why you cannot judge DR-CPAP from observational data alone
- It tempers enthusiasm for routine DR-CPAP until better trials arrive
STRENGTHS
- Comprehensive, pre-registered systematic review with GRADE
- Large pooled numbers (135,472 infants)
- Kept observational and randomised evidence separate rather than lumping them
- A real, current practice question
- Australian and open access
LIMITATIONS
- Dominated by observational studies with confounding by indication
- Only 2 small RCTs, at very low certainty
- Pneumothorax is rare, so even large studies struggle to estimate it
- Substantial heterogeneity between studies
- Cannot give a definitive answer - the trials are not there yet
Practice implications
Delivery-room CPAP for late-preterm and term babies is being used more, but the evidence is genuinely unsettled. Observational studies make it look harmful (more pneumothorax, more NICU admission) - but that is almost certainly confounding by indication, because the sicker baby is the one who gets CPAP. The few randomised trials point the other way (fewer NICU admissions, no pneumothoraces), but they are small and very low certainty. The honest position: do not adopt routine prophylactic DR-CPAP for well term / late-preterm babies on current evidence; use it for a clinical reason and per your local guideline, and if you do, watch for air leak. Read it alongside last week's PLaNT trial - together they show a promising signal that still needs a proper trial.
Why do the observational studies and the RCTs disagree, and which do you trust?
What is "confounding by indication", and where else does it fool us in neonatology?
Given this review and the PLaNT trial, would you offer DR-CPAP to a well late-preterm baby?
What would a definitive trial need to measure?
J Paediatr Child Health 2026 PLaNT TRIAL
Last week's DR-CPAP RCT ATTENDING A BIRTH
Delivery-room support JOURNAL CLUB
Article of the week & archive
Take-home message: This review shows why delivery-room CPAP for late-preterm and term babies is still an open question. Observational data suggest harm (more pneumothorax and NICU admission), but that is likely confounding by indication - CPAP goes to the sicker baby. The small randomised trials suggest the opposite (fewer NICU admissions, no air leaks) but are very low certainty. Bottom line: not enough evidence for routine prophylactic DR-CPAP in well term / late-preterm babies - use it for a reason, watch for air leak, and await a proper trial.
For educational purposes only. Journal club appraisal - figures paraphrased from the published review; read the full article for complete data. Always align practice to your local guidelines.