RESUMO
OBJECTIVE: To determine whether prolonged exposure to antibiotics (>5 days) increases the risk of late-onset sepsis (LOS) in newborns of 1,000-1,500 g. METHODS: A cohort study in newborns with suspected perinatal infection, with a survival greater than seven days. The exposed cohort was composed of newborns with antibiotic therapy initiated at first postnatal day, lasting >5 days, with negative blood cultures before the fifth day of life, and without clinical evidence of sepsis. The non-exposed cohort was identical but with antibiotics stopped before the fifth day of life. Patients were followed daily for clinical and laboratory evidence of LOS. Others risk factors for LOS were analyzed. STATISTICAL ANALYSIS: We analyzed the incidence, the relative risk (RR) with 95% CI. To measure the time to occurrence of an event of LOS after exposure, Kaplan-Meier survival curve and log-rank test were used. RESULTS: We followed up 49 patients in each group. The incidence was 33.6%. The time of follow-up was 839 vs. 1,291 person-days. Prolonged exposure to antibiotics was associated with a higher risk of LOS (RR: 21.1; 95% CI: 6.5-68.9; p=0.000). The late-onset sepsis-free time was 17.1±1.1 vs. 26.3±0.8 days. CONCLUSIONS: The risk of LOS was higher in newborns with prolonged exposure to antibiotics and increased with the days of exposure.