Background and aims An essential step in drug dosing optimalisation is prospective validation of newly proposed dosing regimens. Based on a recently published population pharmacokinetic (PK) model, a neonatal amikacin dosing regimen was developed. The aim of the current study was to prospectively validate this model-derived dosing regimen.
Methods Routine amikacin therapeutic drug monitoring (TDM) concentrations were prospectively collected. To test efficacy of the dosing regimen, early observed TDM results (i.e. prior to and 1 h after the second intravenous amikacin dose) reaching target concentrations (trough <3 mg/L, peak >24 mg/L) were defined. To test stability and accuracy of the model, all observed concentrations were compared with the predicted concentrations and a normalised prediction distribution error (NPDE) was performed. Monte Carlo simulations were used to evaluate amikacin exposure.
Results In total, 1195 TDM results of 579 neonates [median gestational age 34 (range 24–41) weeks, postnatal age 2 (range 1–30) days] were included. Sixty percent of the early trough levels was below 3 mg/L, 90.4% of the peak levels reached 24 mg/L. Comparable parameter estimates were obtained between the final PK model and the prospective dataset. No trend was seen in the NPDE versus time and the NPDE versus predicted concentrations. Based on the Monte Carlo simulations, peak concentrations above 24 mg/L were reached in almost all patient subgroups.
Conclusions After 14 years experience of amikacin dosing optimalisation in (pre)term neonates, a model-based dosing algorithm was prospectively validated confirming its efficacy, stability and accuracy over the entire neonatal population.