Is Tradition Trumping Evidence in the Treatment of Young, Febrile Infants?

Abstract

Ask a pediatric trainee which organisms should be targeted in empiric therapy for febrile infants <1 month of age, and we will bet you a carton of raw milk that the answer will include Listeria . Traditional teaching is that the 3 most common pathogens in neonatal sepsis are group B Streptococcus (GBS), Escherichia coli, and Listeria , and current textbooks continue to promote these organisms as the leading pathogens.1,2 However, recent evidence suggests that the modern epidemiology for bacteremia has shifted.3,4 E. coli has now surpassed GBS, and in multiple reports Listeria has gone from rare to exceedingly rare or nonexistent (a decrease that may be explained by enhanced regulation around food safety or a “collateral benefit” of GBS screening and prophylaxis).5 Nonetheless, ampicillin is still included in most current empiric regimens, presumably to cover for Listeria or Enterococcus . Between 2011 and 2013, the regimen for 71% of infants <28 days old hospitalized at 37 children’s hospitals for fever was ampicillin with a third-generation cephalosporin.6 Because bacteremia and bacterial meningitis in young infants are potentially life-threatening infections, high numbers needed to treat (NNTs) generally have been considered acceptable. However, in considering the addition of ampicillin to cefotaxime, just how high an NNT are we talking about? Given that ampicillin does not confer additional benefit to cefotaxime against E. coli or GBS and other concerns about known and unknown risks of …