Vancomycin population pharmacokinetics in neonates.

BACKGROUND: Recently the value of vancomycin therapeutic drug monitoring, as well as the required therapeutic range, has been subject of debate, resulting in new recommendations. This study was performed to incorporate these new insights in an up-to-date dosing scheme for neonates of various gestational ages.
METHODS: In this retrospective study with prospective validation, 108 newborns with suspected central line-related septicemia during the first month of life received 30 mg/kg/day vancomycin divided into two doses regardless of gestational or postconceptional age. Trough and peak vancomycin serum concentrations were determined before and after the third dose. Vancomycin data were analyzed according to a one-compartment open model with use of NONMEM population pharmacokinetic software. Model parameters were evaluated and then used to simulate vancomycin dosing for different dose and dose interval combinations. Targets were a trough concentration between 5 and 15 mg/L and a peak below 40 mg/L. In the prospective study, the optimal scheme was tested in 22 patients.
RESULTS: Of the 108 patients, 34.3% of measured trough concentrations and 17.6% of peak concentrations were outside the desired therapeutic range. The model that best fitted the data included clearance and volume per kilogram and was independent of gestational age. Simulation of various dosing schemes showed that a dosing schedule of 30 mg/kg/day, irrespective of gestational age, in three doses was optimal, and this scheme was prospectively tested. Mean trough concentrations before the second dose were 8.2 +/- 2.2 mg/L versus a predicted trough of 8.9 +/- 2.5 mg/L. No peak levels higher than 40 mg/L were found.
CONCLUSIONS: The use of the proposed schedule leads to adequate vancomycin trough serum concentrations, and there is no need for routine monitoring of peak serum concentrations.