Intravenous immune globulin prophylaxis of late-onset sepsis in premature neonates.

To determine whether a single dose of intravenously administered immune globulin (IVIG) decreases late-onset sepsis in premature infants, we prospectively entered 753 neonates with birth weight 500 to 2000 gm, gestation < or = 34 weeks, and age < or = 12 hours into a multicenter, double-blind, controlled trial. Infants were randomly selected to receive a single intravenous infusion, 10 ml/kg, of either IVIG (500 mg/kg) or albumin (5 mg/kg) and were observed for 8 weeks for infection. Maternal and neonatal risk factors for infection did not differ between groups. Although serum IgG values before infusion were related to gestation (R = 0.62), the change in serum IgG or half-life of IgG after IVIG infusion was not (R < or = 0.09). The serum IgG concentration was increased (p < 0.05) in IVIG-treated patients for 8 weeks. There were 88 episodes of late-onset sepsis in 79 neonates (10.5%). Causative organisms included the following: Staphylococcus epidermidis (37 episodes), Enterococcus (9), Staphylococcus aureus (7), Candida (6), Escherichia coli (6), and multiple organisms (11). Sepsis, death, and death as a result of infection were unaffected by treatment. We conclude that a single infusion of IVIG, 500 mg/kg, shortly after birth was not effective prophylaxis for late-onset infection in premature neonates. Future studies of late-onset sepsis prophylaxis should consider IVIG with known pathogen-specific antibody concentrations against organisms causing these infections, in particular S. epidermidis.