Abstract: Objective　To explore and analyze the clinical characteristics of extremely premature infants at low risk for early-onset sepsis (EOS), so as to avoid overuse of antibiotics. Method　The clinical data of extremely premature infants hospitalized from January 1, 2010 to December 31, 2017 were collected. Extremely premature infants born from mothers without premature rupture of membranes and without maternal clinical manifestations of chorioamnionitis during pregnancy were classified assigned into the low-risk group, and those who did not meet the low-risk conditions were regarded assigned intoas the control group. EOS was diagnosed according to the results of blood culture within 72 hours after birth. The clinical characteristics, treatment and outcome of extremely premature infants between the two groups were retrospectively analyzed. Results　A total of 245 extremely preterm infants were enrolled, including 153 (62.4%) in low-risk group. Compared with the control group, mothers in low-risk group had higher rates of gestational diabetes and hypertension, higher rates of antenatal hormone use and lower rates of antenatal antibiotics use; furthermore, neonates in low-risk group had lower rates of Apgar score < 5, higher rates of pulmonary surfactant use, respiratory support and mechanical ventilation, and lower risk of death and incidence of early-onset sepsis. The differences were statistically significant (P<0.05). Among In extremely premature infants whose having survival time > 24 hours, compared with control group, infants in low-risk group had higher incidences of respiratory distress syndrome, patent ductus arteriosus, intracranial hemorrhage and bronchopulmonary dysplasia, and lower incidence of pulmonary hemorrhage than control group, and the differences were statistically significant (all P<0.05). In low-risk group, the risks of death, distress syndrome, pulmonary hemorrhage and bronchopulmonary hemorrhage in long-term antibiotic group were higher than the short-term antibiotic group. Conclusion　Early identification of extremely preterm infants at low risk of early-onset sepsis in extremely preterm infants is of clinical significance in reducing early empirical use of antibiotics therapy. respiratory

Key words: early-onset sepsis;　risk factor;　clinical characteristic;　antibiotic;　extremely premature infant