Combination antibiotics consisting of beta-lactam and aminoglycoside are commonly utilized in the treatment of neonatal septicaemia. The aims of this study were to (i) develop an externally validated population pharmacokinetic (PK) model, (ii) evaluate the attainment of target peak and trough serum concentrations by the current hospital amikacin dosing protocol and side effects, and (iii) compare intravenous (IV) versus intramuscular (IM) route of amikacin administration, in terms of attaining peak and trough serum concentration targets. Retrospective chart review was carried out over a 5-year period. All neonates who received amikacin with therapeutic drug monitoring performed were included in this study. A one-compartment population PK model was built, and external validation was performed. A total of 181 neonates (534 serum concentrations) were included in the population PK modeling and external validation. There was no apparent systematic bias in the predictions of the model. The external validation performed in the current study found the model to be generally unbiased. Sixty-one percent of the peak and 99% of the trough levels were within the targeted therapeutic ranges of 15-25 and <5 mg/L, respectively. There was no statistical difference in the proportion of trough concentrations that were within therapeutic range for IV as compared to IM, while IM resulted in a higher proportion of trough concentrations within therapeutic range, as well as higher peak concentrations. The current population PK model and external validation study have proven that the PK model built in the current study can be used to conduct reliable population simulations. IM injection can be an alternative route of administration for amikacin in neonates.