Background Aminoglycosides are administered to treat (suspected) neonatal sepsis. The pharmacokinetics (PK) of this antibiotic class are expected to be different in neonates with perinatal asphyxia (PA) treated with therapeutic hypothermia (TH). Effective exposure of the aminoglycoside amikacin in neonates is achieved using a prospectively validated population PK mod-el-derived dosing regimen.1 However, dosing adjust-ments in case of PA with TH are lacking. The aim of the current (AMICOOL) study was to further explore amikacin disposition in neonates by quantifying the impact of PA treated with TH on amikacin clearance and to provide dosing recommendations for this specific patient population.

Methods Amikacin therapeutic drug monitoring data were retrospectively collected from term neonates with PA treated with TH and admitted to the neonatal inten-sive care units of VUmc Amsterdam and the University Hospitals Leuven between 2010–2015. Data were added to the original published amikacin population PK dataset.2 A data-driven covariate analysis was performed to assess the impact of PA treated with TH on amikacin clearance. Monte Carlo simulations facilitated the comparison of simulated amikacin exposures using the current dosing guidelines.1 and proposed dosing adaptations for PA treated with TH. We hereby aimed to achieve optimal amikacin trough (<5 mg/L) and peak (>24 mg/L) levels. Stochastic simulations were used to investigate the differ-ences in exposure among typical neonates with PA and TH with varying birth weights (1965–4220 g).

Results Data of 55 neonates with PA treated with TH were added to the original amikacin population PK dataset of 930 neonates.2 A 40.6% (RSE 9%) decrease in amikacin clearance for neonates with PA with TH was documented. Based on Monte Carlo simulations, the current dosing guidelines resulted in 40%–57% of neonates with PA and TH displaying amikacin trough concentrations above the toxic trough level (>5 mg/L), while an additional increase of the dosing interval with 12 hours decreased this percentage to 14%. Stochastic simulations showed that among typical neonates the percentage of patients with trough concentrations>5 mg/L ranges 14% to 25%.

Conclusion In neonates with perinatal asphyxia treated with therapeutic hypothermia, amikacin clearance is reduced with 40.6%. Based on simulations, an additional prolongation of the dosing interval with 12 hours results in optimised amikacin exposure and reduces toxicity in this specific population. As a future perspective, the model-based dosing proposal needs prospective validation. Since amikacin can be used as a surrogate for glomerular filtration, clearance of other drugs using the same elimination route could also be reduced in case of perinatal asphyxia treated with therapeutic hypothermia and may require further dosing adaptations.