Development and evaluation of machine learning-based prediction-modelling for initial vancomycin serum concentrations in septic ICU patients using clinical health record data.

BACKGROUND: Sepsis remains a life-threatening condition with highly heterogeneous and dynamic pathophysiology, limiting the effectiveness of uniform therapeutic strategies. Beyond timely source control, antimicrobial therapy represents the only causal treatment option. Vancomycin is widely used for treatment of Gram-positive infections; however, optimal dosing in septic patients is challenging due to pronounced pharmacokinetic variability and substantial interindividual heterogeneity. Underdosing may promote antimicrobial resistance, whereas overdosing increases the risk of toxicity. This study aimed to develop and validate a machine learning-based prediction model to support individualized vancomycin dosing using routinely available clinical data. METHODS: This single-center retrospective study included adult sepsis patients admitted to the intensive care unit, using routinely collected data from the hospital's electronic medical records. Patients were eligible if they received a vancomycin loading dose followed by continuous infusion and had at least one measured serum concentration. Three machine learning models-elastic net regression, random forest, and XGBoost-were developed to predict the initial vancomycin serum concentration. To minimize bias and enhance generalizability, model training, hyperparameter tuning, and performance evaluation were conducted using a stratified nested cross-validation approach. Model performance was compared with seven commonly used population pharmacokinetic models. RESULTS: The developed best performing elastic net model achieved a notable improvement with an average RMSE of 6.19, compared to 7.83 for the best pharmacokinetic model highlighting the potential of early and individualized dosing supported by a machine learning model. Final model analysis revealed that noradrenaline administration, together with classical pharmacokinetic parameters including body weight, serum creatinine, and the presence of chronic kidney disease, significantly influenced predictive performance. CONCLUSIONS: This machine learning-based approach for predicting initial vancomycin serum concentrations outperforms conventional PK models and enables more precise, individualized dosing prior to the availability of therapeutic drug monitoring results. By integrating key clinical variables, the model facilitates data-driven decision-making in sepsis care and underscores the potential of machine learning to advance personalized antimicrobial therapy.