Use of a common spreadsheet program to demonstrate the ability of Bayesian forecasting to estimate the pharmacokinetic parameters of antibiotics.

This article focuses on demonstrating the use of Bayesian forecasting to estimate pharmacokinetic parameters (PKP) of the antibiotics gentamicin and vancomycin using a common spreadsheet program. By simulating patient data with population-based pharmacokinetic models and incorporating assay variability, the study showed that Bayesian methods can accurately estimate key PKP such as clearance (CL) and volume of distribution (Vd) from sparse and randomly timed blood samples. For gentamicin, a one-compartment model was used, yielding strong correlations between true and Bayesian-estimated PKP, with better assay precision improving estimates. For vancomycin, a two-compartment model was applied, and while most PKP correlations were weak, clearance estimates remained strongly correlated, especially when at least one sample was taken during the drug's distribution phase. The study supports Bayesian forecasting as a valuable tool in therapeutic drug monitoring (TDM) for these antibiotics, highlighting its suitability for clinical application despite challenges in teaching and sampling timing.