Objective: Non-invasive respiratory support has reduced the need for mechanical ventilation and surfactant administration in very premature neonates. We sought to determine how the increased use of non-invasive ventilation and less surfactant instillation has impacted the development of bronchopulmonary dysplasia (BPD) and compared BPD outcome applying four currently used definitions. Study Design: This is a retrospective, single center cohort study of neonates born at less than 28 weeks gestation between 2010 and 2018. A respiratory practice change (less surfactant and more non-invasive ventilation) occurred in 2014 following participation in the SUPPORT trial. Therefore, patients were divided into 2 epochs to compare postnatal respiratory and clinical course and BPD outcomes across four currently relevant definitions (VON, NICHD, Canadian, NRN). Results: Clinical and demographic variables were similar between epochs. Despite significant differences in maternal and infant characteristics and clinical course, the incidence of BPD was not significantly different between the 2 epochs regardless of the BPD definition utilized. There was a wide range in the incidence of BPD depending on the definition used. Conclusions: Despite decreased use of invasive mechanical ventilation and surfactant administration between the two epochs, the incidence of BPD did not change and there was wide variation depending on the definition used. A better understanding of the risk factors associated with BPD and a consensus definition is urgently needed in order to facilitate the conduct of clinical trials and the development of novel therapeutic interventions to improve outcome.