Three trials (360 participants) were included. All were considered to have low risk of bias for all items in the risk of bias assessment.
NIPPV significantly reduced the need for invasive ventilation within the first 72 hours of life compared with NCPAP (RR 0.60, 95% CI 0.43 to 0.83; three trials). No heterogeneity was found (Ι²=1%) . Results for the two trials where infants received surfactant by the INSURE (intubate-surfactant-extubate) approach were combined and there was no significant difference between the ventilation methods.
No significant differences between ventilation methods were found for bronchopulmonary dysplasia, the composite outcome of death and bronchopulmonary disease, pneumothorax, intraventricular haemorrhage, necrotising enterocolitis, time to full feeds and duration of hospital stay; moderate heterogeneity was observed for bronchopulmonary dysplasia and duration of hospital stay. No gastric perforation was reported amongst the 180 infants in the NIPPV groups from all three trials.