BACKGROUND Pulmonary hemorrhage (PH) is a life-threatening respiratory complication of extremely low-birth-weight infants (ELBWIs). However, the risk factors for PH are controversial. Therefore, the purpose of this study was to analyze the perinatal risk factors and short-term outcomes of PH in ELBWIs. METHODS This was a retrospective cohort study of live born infants who had birth weights that were less than 1000 g, lived for at least 12 hours, and did not have major congenital anomalies. A logistic regression model was established to analyze the risk factors associated with PH. RESULTS There were 168 ELBWIs born during this period. A total of 160 infants were included, and 30 infants were diagnosed with PH. Risk factors including gestational age, small for gestational age, intubation in the delivery room, surfactant in the delivery room, repeated use of surfactant, higher FiO2 during the first day, invasive ventilation during the first day and early onset sepsis (EOS) were associated with the occurrence of PH by univariate analysis. In the logistic regression model, EOS was found to be an independent risk factor for PH. The mortality and intraventricular hemorrhage rate of the group of ELBWIs with PH were significantly higher than those of the group of ELBWIs without PH. The rates of periventricular leukomalacia, moderate-to-severe bronchopulmonary dysplasia and severe retinopathy of prematurity, and the duration of the hospital stay were not significantly different between the PH and no-PH groups. CONCLUSIONS Although PH did not extend hospital stay or increase the risk of bronchopulmonary dysplasia, it increased the mortality and intraventricular hemorrhage rate in ELBWIs. EOS was the independent risk factor for PH in ELBWIs.