There is robust evidence to support the provision of positive end expiratory pressure (PEEP)/continuous positive airway pressure (CPAP) in the delivery room for preterm babies 1. The 2025 ILCOR update issued 51 recommendations (vs. 2020), of which 48 were not strong 2. In this context, ILCOR concluded there is insufficient evidence to recommend for or against routine CPAP for spontaneously breathing late preterm and term infants with respiratory distress 1. In contrast, the UK Newborn Life Support (NLS) and European Resuscitation Council 2025 guidelines recommend CPAP for spontaneously breathing infants ≥ 32 weeks' gestation with significant respiratory distress requiring supplemental oxygen 3. This recommendation leaves several practical questions unanswered: the optimal PEEP/CPAP level; how long CPAP should be continued; and what objective criteria should trigger escalation or discontinuation. In practice, PEEP/CPAP is sometimes continued beyond initial stabilisation (heart rate > 100/min with spontaneous breathing) when respiratory distress persists. This approach is often used to support transition and potentially avoid mother–infant separation. A practical question remains: how should infants with heart rate > 100/min and spontaneous breathing, but persistent respiratory distress (± desaturation), be managed? We undertook an audit to describe delivery-room CPAP/PEEP use in infants ≥ 34 weeks' gestation requiring any resuscitation, quantify the frequency of ‘extended’ CPAP/PEEP (> 10 min after birth), and examine its association with neonatal unit admission directly from the delivery room. We conducted a retrospective single-centre audit in a UK level 3 neonatal unit between June and December 2023. We identified all liveborn infants ≥ 34 weeks' gestation who received any resuscitation at birth (positive pressure breaths, CPAP, chest compression and drugs). In our unit, infants ≥ 34 weeks with persistent respiratory distress (grunting/chest retractions and others), with or without low oxygen saturation, may receive delivery-room CPAP/PEEP delivered via a T-piece resuscitator. Local practice is to use PEEP 5–6 cmH2O (aligned with NLS guidance) and to titrate inspired oxygen to maintain oxygen saturation > 90%. Extended PEEP was defined as PEEP continued beyond 10 min of life. The primary outcome was neonatal unit admission directly from the delivery room. We summarised data using descriptive statistics and the audit registered with hospital authority. During the study period, there were 2667 deliveries at ≥ 34 weeks' gestation. Of these, 280 infants required some form of resuscitation (10.5%). Baseline characteristics are shown in Table 1. Our study is limited by its retrospective design, single-centre practice and modest sample size. Extended PEEP is likely confounded by indication and should be interpreted as a marker of underlying illness severity rather than a causal factor in neonatal unit admission. The multicentre DRIVE Network registry data also demonstrate wide inter-hospital variation in delivery-room CPAP use among infants ≥ 34 weeks, underscoring uncertainty in practice 4. A substantial proportion of infants ≥ 34 weeks required delivery-room respiratory support. In our cohort, one in four infants who received any resuscitation were admitted to the neonatal unit, and around one third received extended delivery-room PEEP for > 10 min. Given the limited evidence base, further prospective research is needed in this topic. Samantha Davies: data curation, writing – review and editing, investigation. Sian Oldham: investigation, writing – review and editing, data curation. Prakash Kannan Loganathan: conceptualization, investigation, writing – original draft, writing – review and editing, methodology, formal analysis, data curation, supervision. AI Use Disclosure: During the preparation of the manuscript, the authors used OpenAI's ChatGPT (GPT-4, April 2024 version) to assist with improving the clarity, grammar and structure of the abstract and sections of the main text. No part of the manuscript's content, interpretation or conclusions was generated by the AI. The authors reviewed and approved all AI-assisted outputs and take full responsibility for the integrity and accuracy of the work. The authors have nothing to report. The authors declare no conflicts of interest. Deidentified individual participant data will be made available, in addition to study protocols, the statistical analysis plan, and the consent form. The data will be made available upon publication to researchers who provide a methodologically sound proposal and research ethics board approval. Proposals should be submitted to email protected.
Davies et al. (Sat,) studied this question.