Premature babies often need help immediately after birth to open their lungs to air, start breathing and keep their hearts beating. Opening their lungs can be difficult, and once open the under-developed lungs of premature babies will often collapse again between each breath. To prevent this nearly all premature babies receive some form of mechanical respiratory support to aid breathing. Common to all types of respiratory support is the delivery of a treatment called positive end-expiratory pressure, or PEEP. PEEP gives air, or a mixture of air and oxygen, to the lung between each breath to keep the lungs open and stop them collapsing.

Currently, clinicians do not have enough evidence on the right amount, or level, of PEEP to give at birth. As a result, doctors around the world give different amounts (or levels) of PEEP to premature babies at birth.

In this study, the Investigators will look at 2 different approaches to PEEP to help premature babies during their first breaths at birth. At the moment, the Investigators do not know if one is better than the other. One is to give the same PEEP level to the lungs. The others is to give a high PEEP level at birth when the lungs are hardest to open and then decrease the PEEP later once the lungs are opened and the baby is breathing.

Very premature babies have a risk of long-term lung disease (chronic lung disease). The more breathing support a premature baby needs, the more likely the risk of developing chronic lung disease. The Investigators want to find out whether one method of opening the baby's lungs at birth results in them needing less breathing support.

This research has been initiated by a group of doctors from Australia, the Netherlands and the USA, all who look after premature babies.

All infants born <29 weeks' postmenstrual age (PMA) require positive end-expiratory pressure (PEEP) at birth. PEEP is a simple, feasible and cost-effective therapy to support extremely preterm infants that is used globally. The effective and safe level of PEEP to use after preterm birth remains the most important unanswered question in neonatal respiratory medicine.

The Investigators will undertake an international multi-centre randomised controlled trial to address in extremely preterm infants, whether the use of a high, dynamic PEEP level strategy to support the lung during stabilisation ('resuscitation') at birth, compared to the current practice of a static PEEP level, will reduce the rate of death or bronchopulmonary dysplasia (BPD).

This trial will address the following four key knowledge gaps:

1. Assessing whether individualising (dynamic) PEEP is superior to static PEEP
2. The uncertainty regarding applied pressure strategies to support the lung during stabilisation at birth arising from the lack of a properly powered, well-designed randomised trial specifically addressing important outcomes for respiratory support in the Delivery Room
3. The optimal PEEP strategy to use
4. Determining the differential effects of PEEP at different gestational ages.

For this study, the term PEEP refers to the delivery of positive pressure (via a bias flow of gas) to the lungs during expiration by any method of assisted respiratory support, this includes:

1. Continuous Positive Applied Pressure (CPAP; a method of non-invasive respiratory support). During CPAP no other type of positive pressure is delivered as the infant supports tidal ventilation using her/his own spontaneous breathing effort. PEEP during CPAP has also been called 'continuous distending pressure.
2. Positive Pressure Ventilation (PPV). During PPV PEEP is delivered between periods of an applied inflating pressure (PIP) delivered at a clinician-determined rate. PPV can be delivered via a mask or other non-invasive interface (also termed non-invasive positive pressure ventilation; NIPPV), or via an endotracheal tube (often termed continuous mechanical ventilation; CMV).
3. High-frequency oscillatory ventilation (HFOV) or high-frequency jet ventilation. These are modes of invasive PPV in which PIP is delivered at very fast rates (>120 inflations per minute) and at very small tidal volumes. During HFOV a mean airway pressure is determined by the clinician which is equivalent to the PEEP during other modes. During high-frequency jet ventilation the clinician sets a PEEP similar to CMV.

As all of these modes of ventilation have a similar goal of applying a pressure to the lung during expiration (usually to prevent lung collapse) the term PEEP has the same physiological result despite different methods of application.

The specific aim of the trial is to establish whether the use of a high, dynamic 8-12 cmH2O PEEP level ('dynamic') strategy to support the lung during stabilisation at birth, compared with a static 5-6 cmH2O PEEP level ('static') strategy, increases the rate of survival without bronchopulmonary dysplasia (BPD) in extremely preterm infants born <29 weeks' PMA, and reduces rates of common neonatal morbidities.

The Investigators hypothesise that in preterm infants born <29 weeks PMA who receive respiratory support during stabilisation at birth, a high, dynamic PEEP strategy (i.e. PEEP 8-12 cmH2O individualised to clinical need) as compared to a standard, static PEEP of 5-6 cmH2O, will:

1. Increase survival without BPD (primary outcome); and
2. Reduce rates of common neonatal morbidities such as failure of non-invasive respiratory support in the first 72 hours of life (secondary outcome).

This trial is a phase III/IV, two parallel group, non-blinded, 1:1 randomised controlled, multi-national, multi-centre study comparing dynamic PEEP (dynamic group) with standard PEEP strategy (static group).

The intervention will take place in the Delivery Room. The intervention period will be from the time of birth until 20 minutes of life or transfer from Delivery Room to NICU (whatever comes first). The follow-up period will extend to 36 weeks PMA (primary endpoint), and 24 months corrected GA to determine important long-term neurodevelopmental and respiratory outcomes.

The clinical team within the Delivery Room managing enrolled and randomised infants will not be masked/blinded to the intervention. Clinicians need to be able to see the PEEP delivery device to assess efficacy of pressure delivery. The Research Coordinator/Study team at site will also not be masked/blinded to the intervention, as they will be entering trial data into the data management system.

Research staff based at the central Trial Coordinating Centre (TCC), the Data Coordinating Centre (DCCe) and the trial statistician will be blinded to assigned treatment.

There will be a total of 906 infants recruited (453 in the Dynamic group, 453 in the Static group), over 25 recruitment centres across Australia, Europe, the United Kingdom, the Middle East, Canada and North America.

The study will have Regional Coordinating Centres (RCCs) established in the following jurisdictions:

1. Australia - The Murdoch Children's Research Institute/Royal Women's Hospital, Melbourne, AUS
2. The Netherlands - Amsterdam University Medical Centre, Netherlands, EU
3. The United Kingdom - The University of Oxford / National Perinatal Epidemiology Unit (NPEU), Oxford, UK, and
4. North America - the Hospital of the University of Pennsylvania, Pennsylvania, USA.

Inclusion Criteria:

• Infants born between 23 weeks 0 days and 28 weeks 6 days PMA (by best obstetric estimate).
• Receives respiratory intervention (resuscitation) at birth with CPAP and/or positive pressure ventilation in the Delivery Room, to support transition and/or respiratory failure related to prematurity.
• Has a parent or other legally acceptable representative capable of understanding the informed consent document and providing consent on the participant's behalf either prospectively or after birth and randomisation if prenatal consent was not possible (at sites where the Ethics Committee permits waiver of prospective consent).

Exclusion Criteria:

• Not for active care based on assessment of the attending clinician or family decision
• Anticipated severe pulmonary hypoplasia due to rupture of membranes <22 weeks with anhydramnios or fetal hydrops
• Major congenital anomaly or anticipated alternative cause for respiratory failure
• Refusal of informed consent by their legally acceptable representative
• Does not have a guardian who can provide informed consent.

The de-identified data set collected for this analysis of the POLAR trial will be available six months after publication of the primary outcome.

The study protocol, statistical analysis plan and consent forms will also be available. The data may be obtained from the Murdoch Children's Research Institute by emailing david.tingay@rch.org.au and mctc@mcri.edu.au.

6 months after publication of primary outcome.

Prior to releasing any data the following are required:

1. A Data Transfer Agreement must be signed between relevant parties.
2. The MCRI Sponsorship Committee must review and approve your protocol and statistical analysis plan which must include and describe how the data will be used and analysed.
3. An Authorship Agreement to be agreed to and signed between relevant parties. The Agreement must include details regarding appropriate recognition. Authorship may not be justifiable but some form of acknowledgement is requested.
4. Agreement to cover any additional costs relating to the provision of the data.
5. Evidence of ethics approval or waiver of approval, to be compliant with the data transfer agreement and ethics requirements at our end.

Data will only be shared with a recognised research institution where the MCRI Sponsorship Committee has approved the proposed analysis plan.

The clinical team within the Delivery Room managing enrolled and randomised infants will not be masked/blinded to the intervention.

Members of the Research Team at participating sites will also not be masked/blinded to the intervention.

Research staff based at the central Trial Coordinating Centre (TCC), the Data Coordinating Centre (DCCe) and the Trial Statistician will be blinded to assigned treatment.