Abstract

Group B streptococci are common causative agents of early-onset neonatal sepsis (EOS). Pharmacokinetic (PK) data for penicillin G have been described for extremely preterm neonates but poorly for late-preterm and term neonates. Thus, evidence-based dosing recommendations are lacking. We described PK of penicillin G in neonates with gestational age (GA) ≥32 weeks and postnatal age <72 h. Penicillin G was administered intravenously at a dose of 25,000 or 50,000 IU/kg/q12h. At steady state, PK blood samples were collected prior to and at 5 min, 1 h, 3 h, 8 h, 12 h after injection. Non-compartmental PK analysis was performed with WinNonlin. In combination with data from neonates with GA ≤28 weeks we developed a population PK model using NONMEM software and performed probability of target attainment (PTA) simulations. In total, 16 neonates with GA ≥32 weeks were included in non-compartmental analysis. The median (interquartile range) volume of distribution (VD) was 0.50 (0.42-0.57) L/kg, clearance (CL) 0.21 (0.16-0.29) L/h and half-life 3.6 (3.2-4.3) h. In population PK analysis that included 35 neonates, a two-compartment model best described the data. The final parameter estimates were 10.3 L/70kg and 29.8 L/70kg for VD of the central and peripheral compartment, respectively, and 13.2 L/h/70kg for CL. Considering fraction of unbound penicillin G of 40%, PTA of time when the unbound drug exceeds MIC of 40% was >90% for MICs ≤2 mg/L with doses of 25,000 IU/kg/q12h. In neonates, regardless of GA, PK parameters of penicillin G are similar. The dose of 25,000 IU/kg/q12h is suggested for treatment of group B streptococcal EOS diagnosed within the first 72 hours of life.