PURPOSE:The aim of this study was to determine whether postnatal sepsis is a risk factor for developing bronchopulmonary dysplasia (BPD) in premature infants. METHODS:Premature infants of less than 32 weeks of gestation or birth weight under 1,500 g who admitted at Kangnam Sacred Heart Hospital of Hallym University during 1997-2005 were investigated. Among them, infants born at other hospitals or died within 4 weeks of life and infants who got sepsis after neonatal period were all excluded. Total 268 cases were included in this study. They were divided into those with respiratory distress syndrome (RDS) (n=77) and without RDS (n=191) and each was subdivided into groups with sepsis and without sepsis. Based on the medical records, the incidence of BPD and its correlation with sepsis were analyzed. RESULTS:Incidence of typical BPD was 63.0% in group with sepsis and 26% in group without sepsis. In case of atypical BPD, incidence was 20.5% in group with sepsis and 1.3% in group without sepsis. When logistic regression analysis was performed for correcting the confounding factors, sepsis was found to be statistically significant for BPD de development (OR 3.159, 95% CI=1.241-8.039, P<0.05). RDS, birth weight were also independently significant. Other perinatal factors such as gestational age, 1 min & 5 min Apgar score, total duration of assisted ventilation, total duration of O2 administration, PDA, ROP, and IVH were found to be significant risk factors for developing BPD. We could not find any statistically significant maternal risk factors for BPD occurrence. CONCLUSION:For premature infants under 1,500 g of birth weight or 32 week of gestational age and less, sepsis during the first 4 weeks of age was a significant risk factor for developing both typical and atypical BPD.