Amikacin dosing in neonates: evaluation of target attainment using a simplified and complex pharmacokinetic model-derived dosing regimen in clinical practice.

Abstract

Amikacin is frequently used for the treatment of neonatal sepsis. The Dutch Pediatric Formulary recommends a complex pharmacokinetic (PK) model-derived dosing regimen, which consists of dosing categories based on postnatal age and weight that results in adequate PK/pharmacodynamic (PK/PD) target attainment. However, a simplified dosing regimen may be easier to apply in clinical practice. We evaluated PK/PD target attainment of amikacin in neonates using this simplified or complex dosing regimen. This retrospective cohort study included neonates with routinely measured amikacin concentrations at the neonatal intensive care units of the Amsterdam University Medical Center (simplified dosing regimen) or University Hospitals Leuven (complex dosing regimen). Peak (C_max) and trough (C_min) concentrations and the area under the concentration-time curve (AUC) for the first dosing interval were calculated by Bayesian estimation for both populations. Targets of C_max (≥15, ≥25, and ≥35 mg/L), C_min (≤3 and ≤5 mg/L), and AUC/minimal inhibitory concentration (MIC: 2, 4, and 8 mg/L for Enterobacterales species) for bacteriostasis and 1-log reduction were evaluated. A target attainment of ≥90% was considered adequate. In total, 366 neonates (768 concentrations) and 579 neonates (1,195 concentrations) received the simplified and complex dosing regimen, respectively. Both regimens achieved target attainment of 100% for C_max ≥ 15 mg/L, C_min ≤ 5 mg/L, AUC/MIC for bacteriostasis, and AUC/MIC for 1-log reduction up to a MIC of 2 mg/L. Target attainment was achieved for less stringent targets (C_max ≥ 15 mg/L, C_min ≤ 5 mg/L, and AUC/MIC for bacteriostasis) with the simplified and complex amikacin dosing regimen. Clinicians can choose one of both dosing regimens, depending on their local circumstances, and the availability of integrated (electronic) prescription tools.