OBJECTIVE: To investigate the epidemiology, clinical features, treatment, and prognostic factors of neonatal acute respiratory distress syndrome (NARDS) through a retrospective study of NARDS based on the Montreux definition. METHODS: A retrospective analysis was performed on the medical records of neonates who were hospitalized from January 2017 and July 2018, among whom 314 neonates who met the Montreux definition were enrolled as subjects. According to oxygen index, they were divided into a mild NARDS group with 130 neonates, a moderate NARDS group with 117 neonates, and a severe NARDS group with 67 neonates. The clinical features were compared among the three groups to investigate the influencing factors for the severities of NARDS and the length of hospital stay. RESULTS: The neonates with NARDS accounted for 2.46% (314/12 789) of the neonates admitted to the neonatal ward during the same period of time and had a mortality rate of 9.6% (30/314). The multivariate ordinal logistic regression analysis showed that the neonates who used pulmonary surfactant (PS) or had a long duration of assisted ventilation tended to have a higher risk of severe NARDS (P < 0.05). The Cox regression analysis showed that the neonates with low birth weight/macrosomia, preterm birth, invasive ventilation, PS therapy, or positive pathogenic detection had a higher risk of prolonged hospital stay (P < 0.05). CONCLUSIONS: Preterm birth, low birth weight/macrosomia, and perinatal infection may be associated with an increased risk of severe NARDS. The neonates requiring invasive ventilation, prolonged assisted ventilation, or PS therapy tend to have a poor prognosis.