Optimization of Vancomycin Therapy in Neonatal Intensive Care Unit (NICU)

Abstract

Vancomycin is the first-line treatment for staphylococcal infections in neonates, yet safe and effective dosing remains difficult because of limited clinical data and major developmental differences in pharmacokinetics compared with adults. These challenges highlight the need for reliable therapeutic drug monitoring (TDM) in neonatal intensive care units. This thesis focuses on improving vancomycin dosing and TDM practices for neonates through four complementary studies. A systematic review assessed current dosing strategies and showed that all existing algorithms achieved less than 80% target attainment, revealing the limitations of conventional approaches. A retrospective study at Westmead Hospital examined real-world practice in 69 neonates. When TDM was performed appropriately, 75% reached the target trough after the first dose, rising to 84% after TDM-guided adjustments, although many courses still required further dose optimization. To support more precise dosing, a model-informed precision dosing (MIPD) evaluation was conducted. Among 25 identified population pharmacokinetic models, nine were suitable for deeper analysis, and the De Cock et al. (2014) model showed the most reliable predictive performance across apriori, aposteriori, and Bayesian forecasting. These findings support the value of MIPD for individualized and accurate dosing in neonates. The thesis also highlights the important role of clinical pharmacists in leading TDM. Their expertise in interpreting drug exposure and applying MIPD tools is essential for achieving safe and effective therapy, yet often underutilized. Overall, the results demonstrate substantial variability in current neonatal vancomycin practice and the need for more consistent, model-informed, and pharmacist-supported TDM. Bayesian and MIPDguided approaches offer a promising path toward safer and more individualized dosing in this vulnerable population.