Non-invasive ventilation (NIV) is now considered the first-line treatment for respiratory distress syndrome in preterm infants. We aimed to evaluate the rates of non-invasive ventilation failure rate in very preterm infants, as well as to identify its predictors and associated outcomes. We designed a single-center retrospective cohort study including infants ≤32 weeks gestational age and ≤1500 g. The NIV failure was defined as the need for intubation at <72 h of life. After applying inclusion and exclusion criteria, 154 patients were included in the study, with a mean GA of 29.7 ± two weeks. The NIV failure rate was 16.2% (n = 25) and it was associated with lower bronchopulmonary dysplasia (BPD)-free survival (OR 0.08; 95% CI 0.02−0.32) and higher incidence of intraventricular hemorrhage > II (OR 6.22; 95% CI 1.36−28.3). These infants were significantly smaller in GA and weight. Higher FiO2 during resuscitation (OR 1.14; 95% CI 1.06−1.22) and after surfactant administration (OR 1.17; 95% CI 1.05−1.31) represented independent risk factors for NIV failure. In conclusion, NIV failure is frequent and it could be predicted by a higher oxygen requirement during resuscitation and a modest response to surfactant therapy. Importantly, this NIV failure is associated with worse clinical outcomes.
Keywords: continuous positive airway pressure; hyaline membrane disease; noninvasive ventilation; prematurity and respiratory distress syndrome.