Predicting Antibiotic Effect of Vancomycin Using Pharmacokinetic/Pharmacodynamic Modeling and Simulation: Dense...

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Drug Dosing Optimization
• Antibiotic Therapy

Background:

• Accurate pharmacokinetic (PK) modeling is crucial for optimizing vancomycin dosing and predicting treatment success.
• Sparse sampling strategies can simplify data collection but may impact model accuracy.

Purpose of the Study:

• To evaluate the impact of reduced-compartment PK models, developed from sparse sampling, on vancomycin PK parameter estimation.
• To assess the effect of these simplified models on the prediction of vancomycin's probability of target attainment (PTA).

Main Methods:

• Virtual simulation of vancomycin concentration-time profiles using two- and three-compartment PK models.
• Generation of reduced datasets simulating sparse blood sampling times.
• Monte Carlo simulations to evaluate PTA and PK parameter accuracy (RBias, RRMSE).

Main Results:

• Reduced one-compartment models showed over 90% relative bias and RRMSE for vancomycin clearance (CL) in two- and three-compartment profiles.
• Reduced two-compartment models exhibited significant bias in steady-state volume of distribution.
• Fewer compartments led to lower predicted area under the concentration-time curve (AUC).

Conclusions:

• Simplified PK models derived from inadequate sparse sampling designs yield inaccurate and imprecise vancomycin PK parameters.
• Misprediction of PTA and suboptimal vancomycin dosage regimens can result from using oversimplified models.
• Careful consideration of sampling strategy and model complexity is essential for reliable vancomycin dosing.