International guidelines recommend giving positive pressure ventilation (PPV) by face mask to newborns who do not breathe or have a slow heart rate at birth. Preterm infants are at high risk of developing respiratory distress syndrome (RDS) and many are treated with continuous positive airway pressure (CPAP) in the neonatal intensive care unit (NICU). Though the majority of preterm infants breathe spontaneously at birth, many clinicians routinely apply a face mask to preterm infants shortly after birth in the delivery room (DR) to give them CPAP. However, applying a face mask may inhibit spontaneous breathing in newborns. In this study, premature babies will be randomly assigned to have a face mask routinely applied for CPAP shortly after birth; or to have a face mask selectively applied only for PPV if they are not breathing or have a slow heart beat in the first 5 minutes of life, or for CPAP if they have signs of respiratory distress after 5 minutes. The investigators will determine whether fewer participants who have the mask selectively applied receive PPV in the DR.

Newly born babies have fluid-filled lungs that they must quickly aerate after birth. Compared to infants born at term, preterm babies have greater difficulty in establishing and maintaining aeration of their lungs; this leaves them at increased risk of developing respiratory distress syndrome (RDS). Infants who develop RDS are treated with nasal continuous positive airway pressure (CPAP), and may progress to treatment with surfactant and mechanical ventilation, in the Neonatal Intensive Care Unit (NICU).

The Neonatal Task Force of the International Liaison Committee on Resuscitation (ILCOR) makes recommendations on the treatment of infants at birth. ILCOR recommends assessing the breathing and heart rate (HR) of all newborns, and:

1. Giving positive pressure ventilation (PPV) to babies who have a HR < 100bpm, gasping or apnoea
2. Considering giving CPAP to babies who have laboured breathing or persistent cyanosis

Most preterm infants breathe spontaneously at birth. Despite this, the majority of preterm infants have a facemask applied for respiratory support immediately after birth, usually before the HR has been determined. Clinicians presumably do this to give early support to infants they believe are at high risk of developing RDS in an attempt to prevent or lessen the severity of the disease.

There is little evidence that giving preterm infants prophylactic nasal CPAP may be superior to supportive care with oxygen. A study that compared nasopharyngeal CPAP to supportive care with oxygen performed before antenatal steroids were routinely given found no difference in the rate of development of RDS with the application of CPAP. Two more recent studies did not show that early application of nasal CPAP reduced the rate of intubation or treatment with surfactant. In these studies CPAP was given by nasal prongs and was started at 15 - 30 minutes of life, not immediately with a facemask. There is no evidence that facemask CPAP immediately after birth prevents or reduces the severity of RDS.

Application of a face mask has been demonstrated to inhibit spontaneous breathing in many term infants. Cold gas flow, such as that provided by a T-piece, can inhibit spontaneous breathing in term infants. Application of a face mask for breathing support appears to inhibit breathing in a greater proportion of preterm infants. Considerable force is applied to the head when face mask PPV is given to a mannequin in the supine position.

Routinely applying a face mask for respiratory support may be unnecessary in many premature babies. It may inhibit their spontaneous breathing and result in them receiving facemask PPV more frequently in the DR. It may also affect how well they breathe in the first day of life and increase the rate of treatment with nasal CPAP in the NICU.

METHODS

RANDOMISATION AND GROUP ASSIGNMENT Participants will be randomly assigned to "SELECTIVE" or "ROUTINE" groups in a 1:1 ratio. The group assignment schedule will be generated in blocks of 4 using a random number table, and will be stratified by gestational age (23 - 27, 28 - 31+6). It will be kept on a concealed from investigators and treating clinicians. Group assignment will be written on cards and placed in sequentially numbered, sealed, opaque envelopes. Infants of multiple gestations will be randomised as individuals.

The envelopes will be contained in 2 boxes for GA strata and kept in the NICU. The next envelope in the sequence will be taken from the appropriate box and brought to the DR, where it will be opened just before delivery.

MASKING It is not possible to mask caregivers to group assignment.

INTERVENTIONS

ALL INFANTS Infants in both groups will have their umbilical cord clamped at or after 1 minute of age; they will not have a mask applied for respiratory support during this time. Infants will then be transferred to resuscitaire, placed supine under radiant heat and have a hat applied. All infants will be placed in polyethylene bag. A pulse oximeter will be placed on the right wrist.

A T-piece device (Neopuff, Fisher & Paykel Healthcare, Auckland NZ; or Dräger Resuscitaire (Drager healthcare, Lübeck, Germany) and round face mask (Fisher & Paykel Healthcare, Auckland NZ) of appropriate size for the baby will be used to give respiratory support to infants in both groups. The T-piece will have gas flow set at 8 - 10 L/min, with settings PEEP 6cmH2O, and PIP 25cmH2O. There will be an air/oxygen blender in the circuit and the FiO2 will be initially set at 30%. The FiO2 will not be adjusted before 5 minutes of age.

SELECTIVE GROUP Infants assigned to the "SELECTIVE" group will be placed supine to breathe spontaneously. If SpO2 < 70% at 5 minutes and their respiratory effort is good, they will be given give free flow oxygen (i.e. the mask will be placed in front of, but not directly in contact with, the infant's face). The FiO2 may be increased at 1 minute intervals thereafter, aiming for SpO2 ≥ 90% at 10 minutes of life.

Infants in the "SELECTIVE" group will have mask PPV if they are apnoeic or have HR < 100bpm at any time in the DR. Clinicians may consider applying facemask to give CPAP if the infants breathing is laboured (i.e. there are signs of respiratory distress - grunting, intercostal/subcostal/sternal recessions) after 5 minutes of age.

ROUTINE GROUP Infants assigned to the "ROUTINE" group will be placed supine on the resuscitaire and have facemask CPAP applied as soon as possible after they arrive. If SpO2 < 70% at 5 minutes, the FiO2 may be increased at 1 minute intervals, aiming for SpO2 ≥ 90% at 10 minutes of life.

Infants in the "ROUTINE" group will have facemask PPV if they are apnoeic or have HR < 100bpm at any time in the DR. Mask respiratory support may be withdrawn as and when desired by clinical staff.

RESUCE TREATMENT All other treatments - i.e. endotracheal intubation for PPV, chest compressions, adrenaline, volume - w ill be given at the discretion of the treating clinicians and in accordance with ILCOR recommendations.

SAMPLE SIZE ESTIMATION To show a reduction in the proportion of babies receiving PPV with routine application from 60% to 40% with selective application with 80% power and α of 0.05, the investigators need to recruit 200 infants.

Inclusion Criteria:

• Born before 32 weeks gestational age by best obstetric estimate

Exclusion Criteria:

• Infants with major congenital anomalies

Anonymised data will be shared for studies approved by the chief and principal investigators

Data will be available after the primary report of the study has been published and will be available for 5 years

Requests for data that are accompanied by a protocol that describes the nature and purpose of the study, the data that is required and how it will be analysed will be considered by the chief and principal investigators. Studies that are determined not to be contrary to partcipants' interests; and to be consistent with the purpose for which consent was given for the original study and methodologically sound may be approved

Masking of caregivers not feasible due to the nature of the intervention