This randomised controlled trial aims to investigate the effectiveness of closed-loop automated oxygen control (CLAC) in preterm infants receiving non-invasive respiratory support and determine if it reduces the duration of supplementary oxygen treatment and improves achievement of oxygen saturation targets, reduces the incidences of hypoxia and hyperoxia, the number of manual adjustments to the inspired oxygen concentration (FiO2), and adverse outcomes including bronchopulmonary dysplasia (BPD).

The study will take place at King's College Hospital neonatal intensive care unit (NICU). Parents of preterm infants born at less than 34 weeks of gestation and receiving non-invasive respiratory support will be approached, informed and if appropriate consented to join the trial.

Participants will be randomised to receiving either automated or manual oxygen control.

The study will measure outcomes including the duration of supplementary oxygen treatment, the percentage of time spent within oxygen saturation targets, the incidences of hypoxia and hyperoxia, the number of manual FiO2 adjustments required, the overall duration of non-invasive respiratory support and length of neonatal unit stay, and the incidence of BPD at 36 weeks postmenstrual age (PMA). Results will be compared between the two groups.

Preterm infants frequently require supplemental oxygen to achieve adequate oxygen delivery to the tissues and allow for normal cell metabolism. Oxygen treatment, although life-saving, can further increase the risk of complications. Hyperoxia leads to the development of reactive oxide species (ROS) and increases oxidative stress for the neonates due to their immature antioxidant defence systems. Oxidative stress increases the risk of complications such as bronchopulmonary dysplasia and retinopathy of prematurity (ROP). On the other hand, hypoxia increases morbidity and mortality.

Studies have demonstrated that automated compared to manual oxygen control systems in preterm ventilated infants result in an improvement in the achievement of oxygen saturation targets, reduced time spent in hypoxia and hyperoxia and fewer manual adjustments to the inspired oxygen concentration. In addition, CLAC has been associated with earlier weaning of the inspired oxygen. In a RCT in preterm ventilated infants, CLAC was associated with shorter durations of mechanical ventilation (MV) and supplemental oxygen and reductions in the incidence of BPD and the proportion of infants discharged on home oxygen.

Preterm infants are increasingly managed on non-invasive respiratory support with an aim to minimise lung injury and reduce respiratory morbidity. Optimisation of supplemental oxygen treatment could further improve respiratory outcomes in this population. Previous studies on CLAC that included preterm infants on non invasive respiratory support showed promising results but they were of short duration. In addition, they did not report on the effect of CLAC on longer term outcomes. These limitations highlighted the need for an adequately powered randomised controlled trial (RCT) of CLAC versus manual oxygen control in preterm non-ventilated infants and for the whole duration of non-invasive respiratory support.

This study's hypothesis is that the use of CLAC in preterm infants < 34 weeks of gestation receiving non-invasive respiratory support will reduce the duration of non-invasive ventilation (NIV). The investigators will also evaluate the time spent within oxygen saturation targets, the incidences of hypoxemia and hyperoxemia, the number of manual adjustments required by clinical staff, the total duration of supplemental oxygen treatment, the length of neonatal unit stay and the incidence of BPD at 36 weeks postmenstrual age (PMA).

This is a single centre, non-blinded, randomised controlled trial in preterm infants born at less than 34 weeks of gestation requiring non-invasive respiratory support at any postnatal age.

The investigators aim to recruit a minimum of 76 preterm infants (39 in each group) and over 12 months.

Informed written consent will be requested from the parents or legal guardians of the infants.

Eligible infants whose parents' consent to the study will be enrolled within 48 hours of initiation of non-invasive respiratory support. Infants who have not been eligible for recruitment within 48 hours of initiation of non-invasive support (for example outborn infants transferred to our unit at a later date) but who remain on it on day seven of life and beyond, they will be enrolled to the study immediately after obtaining parental consent and if less than fourteen days old.

Randomisation will be performed using an online randomisation generator. Patients will be receiving non-invasive support using SLE6000 ventilators or SLE6000N non-invasive respiratory device. Non-invasive support will include nasal continuous positive airway pressure (CPAP), non-invasive positive pressure ventilation (NIPPV) or humidified high flow nasal cannula oxygen (HHFNC). Settings will be manually adjusted by the clinical team as per unit's protocol. All participants will be connected to the standard bedside monitor (Philips IntelliVue MX750) that uses the Nellcor Neonatal SpO2 sensor. The intervention group, in addition to standard care will be also connected to the OxyGenie closed-loop oxygen saturation monitoring software (Inspiration Healthcare, Croydon, UK). Manual adjustments including the percentage of FiO2 will be allowed at any point during the study if deemed appropriate by the clinical team.

The nurse-to-patient ratio will be according to the unit's protocol that is determined on the patient's acuity.

Patients will be studied from enrolment until weaning of non-invasive ventilation or 36 weeks PMA or discharge. If an infant is weaned to low flow nasal cannula oxygen or room air and subsequently requires resuming non-invasive ventilation, they will be studied in their initial arm if less than 36 weeks PMA. Therefore, for those infants randomised at the intervention group CLAC will resume. Preterm infants that remain on non-invasive respiratory support beyond 36 weeks PMA will continue at their study arm (CLAC or manual oxygen control) till their first weaning attempt.

Inclusion Criteria:

Preterm infants <34 weeks of gestation and at any postnatal age on non-invasive respiratory support including:

• non-invasive positive pressure ventilation (NIPPV)
• nasal CPAP
• HHFNC oxygen, either as primary or post extubation respiratory support.

Exclusion Criteria:

• Infants with congenital cyanotic heart disease.
• Infants with other know major congenital abnormalities.