Face mask versus nasal mask device use for initial resuscitation in extremely and very preterm infants (FONDUE): an open-label, single-centre, randomised, controlled trial.

Background: More than 85% of very preterm infants (born <32 weeks' gestation) breathe spontaneously within 1 min of birth, however, more than 60% of infants receive positive pressure ventilation. Face mask application soon after birth might suppress breathing through the trigeminal nerve reflex, causing vocal cord closure and hypoxia. We aimed to investigate whether nasal mask continuous positive airway pressure (nCPAP) would improve CPAP success, reducing the need for positive pressure ventilation and intubation at birth, compared with face mask CPAP (fCPAP).

Methods: This open-label, randomised controlled trial was done at Monash Medical Centre (Melbourne, VIC, Australia). Eligible infants were very preterm (born at 23 weeks 0 days [23w0d]-31w6d of gestation) without known relevant congenital anomalies. Infants were randomly assigned (1:1) immediately before birth to receive initial respiratory support with nCPAP or fCPAP. Randomisation was done using a pre-generated randomisation schedule stratified by gestational age (23w0d-27w6d weeks vs 28w0d-31w6d gestation). Due to the nature of the intervention, investigators or clinicians were aware of treatment allocation, but the trial statistician was masked to allocations. The primary outcome was CPAP success, defined as adequate respiratory support with CPAP only, without escalation to positive pressure ventilation or intubation before neonatal unit admission. The primary outcome was assessed in the intention-to-treat population, which included all infants who were randomly assigned as per their allocated treatment. Safety was assessed in all randomly assigned infants. This study was prospectively registered with the Australian New Zealand Clinical Trials Registry, ACTRN12620001086954.

Findings: Between Dec 2, 2020, and March 17, 2023, we enrolled 151 infants: 74 were assigned to nCPAP and 77 to fCPAP. Mean gestation was 28 weeks and 6 days (SD 2 weeks and 6 days), mean birthweight was 1155 g (381), and 82 (54%) of 151 infants were female. 51 (34%) of 151 infants were born before 28 weeks' gestation. More infants in the nCPAP group were successfully managed without escalation to positive pressure ventilation than infants in the fCPAP group (43 [58%] of 74 infants vs 30 [39%] of 77 infants; risk ratio 1·49 [95% CI 1·06-2·10]). Two infants in each group died. Pneumothorax occurred in three (4%) of 77 infants in the fCPAP group (none in the nCPAP group). Intraventricular haemorrhage occurred in 26 (34%) of 77 infants in the fCPAP group and 19 (26%) of 74 infants in the nCPAP group. Three infants had periventricular leukomalacia (one in the fCPAP group and two in the nCPAP group). Two infants in each group underwent surgery for necrotising enterocolitis and one (1%) of 77 infants in the fCPAP group had surgery for intestinal perforation (none in the nCPAP group). 13 (17%) of 77 infants in the fCPAP group and 14 (19%) of 74 infants in the nCPAP group were treated for sepsis. Two (3%) of 77 infants in the fCPAP group and seven (9%) of 74 infants in the nCPAP group were treated for retinopathy of prematurity.

Interpretation: In very preterm infants, compared with fCPAP, the initial use of nCPAP at birth was more likely to result in CPAP success, avoiding the need for positive pressure ventilation and reducing exposure to higher level respiratory support.

Funding: Australian National Health and Medical Research Council and the Jack Brockhoff Foundation.