Influences of Initial Empiric Antibiotics with Ampicillin plus Cefotaxime on the Outcomes of Neonates with Respiratory Failure: A Propensity Score Matched Analysis.

Background: Empiric antibiotics are often prescribed in critically ill and preterm neonates at birth until sepsis can be ruled out. Although the current guideline suggests narrow-spectrum antibiotics, an upgrade in antibiotics is common in the neonatal intensive care unit. The impacts of initial broad-spectrum antibiotics on the outcomes of critically ill neonates with respiratory failure requiring mechanical intubation have not been well studied. Methods: A total of 1162 neonates from a tertiary level neonatal intensive care unit (NICU) in Taiwan who were on mechanical ventilation for respiratory distress/failure at birth were enrolled, and neonates receiving ampicillin plus cefotaxime were compared with those receiving ampicillin plus gentamicin. Propensity score-matched analysis was used to investigate the effects of ampicillin plus cefotaxime on the outcomes of critically ill neonates. Results: Ampicillin plus cefotaxime was more frequently prescribed for intubated neonates with lower birth weight, higher severity of illness, and those with a high risk of early-onset sepsis. Only 11.1% of these neonates had blood culture-confirmed early-onset sepsis and/or congenital pneumonia. The use of ampicillin plus cefotaxime did not significantly contribute to improved outcomes among neonates with early-onset sepsis. After propensity score-matched analyses, the critically ill neonates receiving ampicillin plus cefotaxime had significantly worse outcomes than those receiving ampicillin plus gentamicin, including a higher risk of late-onset sepsis caused by multidrug-resistant pathogens (11.2% versus 7.1%, p = 0.027), longer duration of hospitalization (median [IQR], 86.5 [47-118.8] days versus 78 [45.0-106.0] days, p = 0.002), and a significantly higher risk of in-hospital mortality (14.2% versus 9.6%, p = 0.023). Conclusions: Ampicillin plus cefotaxime should not be routinely prescribed as the empiric antibiotics for critically ill neonates at birth because they were associated with a higher risk of infections caused by multidrug-resistant pathogens and final worse outcomes.