In silico evaluation of a beta-lactam dosing guideline among adults with serious infections

Abstract

Study Objective: The aim of this study was to compare the achievement of therapeutic pharmacokinetic–pharmacodynamic (PK-PD) exposure targets for beta-lactam antibiotics using product information dosing or guideline-based dosing for the treatment of serious infections. Design: In silico study. Data Source: ID-ODSTM (Individually Designed Optimum Dosing Strategies). Patients and Intervention: None. Measurements and Main Results: In silico product information and guideline-based dosing simulations for cefepime, ceftazidime, flucloxacillin, meropenem, and piperacillin/tazobactam were performed using pharmacokinetic models from seriously ill patient populations. The median simulated concentration at 48 and 96 h was used to measure the probability of target attainment (PTA) to achieve predefined therapeutic and toxicity PK-PD targets. A multiple linear regression model was constructed to identify the effect of guideline-based dosing covariates on achieving pre-defined therapeutic targets. In total, 480 dosing simulations were performed. The PTA percentage with guideline-based dosing at 48 and 96 h was 80% and 68%, respectively, yielding significantly higher results when compared to product information dosing (48.45% and 49%, respectively), p < 0.001 at both time points. At 48 h, predefined toxicity thresholds were exceeded in 4.7% and 0% of simulations for guideline-based and product information-based dosing, respectively (p = 0.002). eGFR was significantly associated with the % PTA by guideline-based dosing, with eGFR values of 20 and 50 ml/min both statistically significant in leading to an increase in PTA. Conclusions: Our study demonstrated that achievement of PK-PD exposures associated with an increased likelihood of effectiveness was more likely to occur with guideline-based dosing; especially at 48 h.