Vancomycin: Predictive Performance of a Population Pharmacokinetic Model and Optimal Dose in Neonates and Young Infants

Introduction: Model evaluation is an important issue in population pharmacokinetic analyses. The objectives were to evaluate the predictive performance of previously published pediatric population pharmacokinetic models for vancomycin in a new data set and to propose an optimal dose to obtain a vancomycin concentration target. Methods: External evaluation was conducted for all the published models of vancomycin in neonates and young infants with a new data set of 70 patients. Bias and accuracy were calculated. Advanced analyses were performed to evaluate the predictive performance of the best model. This population pharmacokinetic analysis was performed to simulate doses of vancomycin according to the appropriate target concentration. Results: All models gave almost the same results, except 2 that were not acceptable. Nevertheless, the model described by Oudin et al presented the best results with a bias and accuracy of 4.0% and 27.8%, respectively. Simulations showed that the maintenance dose should be adjusted more precisely to each neonate based on his or her weight and serum creatinine value. Conclusion: Simulations have allowed the authors to describe new dosage schedules, and a chart was created to help clinicians to adapt dosage of vancomycin. Because of pharmacokinetic variability, vancomycin still requires therapeutic drug monitoring.