Population pharmacokinetic and pharmacodynamic analysis of amikacin and gentamicin in hospitalised neonates: a comparative simulation study from two tertiary hospitals in Ghana

Abstract Background Neonatal sepsis accounts for 17–29% of neonatal deaths in sub-Saharan Africa, where the neonatal mortality rate remains approximately 27 deaths per 1,000 live births. In Ghana, infection causes approximately 31% of neonatal deaths. Aminoglycosides are central to empirical management, yet gentamicin resistance among the predominant neonatal pathogen Klebsiella pneumoniae now exceeds 90% in Ghanaian NICUs, while therapeutic drug monitoring is unavailable in most neonatal units. No comparative PK/PD analysis of gentamicin and amikacin has previously been conducted in this setting. Methods A retrospective cross-sectional study was conducted at two tertiary hospitals in Ghana (37 Military Hospital and University Hospital, KNUST). Population pharmacokinetic modelling and Monte Carlo simulations (n = 500 per regimen) were performed using Pumas.jl (v2.6.0) in Julia. Efficacy was quantified as the probability of target attainment (PTA; Cmax/MIC ≥ 8) and safety was assessed using trough concentration thresholds for each drug. Results Amikacin clearance was 0.0492 mL/min (95% CI 0.0430–0.0577) and volume of distribution was 0.865 L (95% CI 0.808–0.918). Gentamicin clearance was 0.0439 mL/min (95% CI 0.0398–0.0483) with volume of distribution 0.837 L (95% CI 0.797–0.877). Both parameters were strongly influenced by body weight and maturational age. Amikacin 7.5 mg/kg q12h achieved PTA ≈ 64% at MIC 2 mg/L with 2% trough toxicity risk; gentamicin 4 mg/kg q24h achieved 100% PTA at MIC ≤ 1 mg/L with 6% trough toxicity risk. Amikacin demonstrated superior pharmacodynamic coverage at higher MIC thresholds. Conclusions Aminoglycoside pharmacokinetics in Ghanaian neonates are substantially shaped by body size and renal maturation. Simulation-derived dosing benchmarks support a risk-stratified approach: gentamicin 4 mg/kg q24h for susceptible organisms and amikacin 7.5 mg/kg q12h where resistance is suspected. Integration of these findings into neonatal antimicrobial stewardship programmes offers a practical pathway to improving sepsis outcomes in resource-limited settings.