Population pharmacokinetic model-guided optimization of intravenous voriconazole dosing regimens in critically ill patients with liver dysfunction

Study Objectives This study aimed to establish a population pharmacokinetic (PPK) model of intravenous voriconazole (VRC) in critically ill patients with liver dysfunction and to explore the optimal dosing strategies in specific clinical scenarios for invasive fungal infections (IFIs) caused by common Aspergillus and Candida species. Design Prospective pharmacokinetics study. Setting The intensive care unit in a tertiary-care medical center. Patients A total of 297 plasma VRC concentrations from 26 critically ill patients with liver dysfunction were included in the PPK analysis. Methods Model-based simulations with therapeutic range of 2-6 mg/L as the plasma trough concentration (C-min) target and the free area under the concentration-time curve from 0 to 24 h (integral AUC(24)) divided by the minimum inhibitory concentration (MIC) (ie, integral AUC(24)/MIC) >= 25 as the effective target were performed to optimize VRC dosing regimens for Child-Pugh class A and B (CP-A/B) and Child-Pugh class C (CP-C) patients. Results A two-compartment model with first-order elimination adequately described the data. Significant covariates in the final model were body weight on both central and peripheral distribution volume and Child-Pugh class on clearance. Intravenous VRC loading dose of 5 mg/kg every 12 h (q12h) for the first day was adequate for CP-A/B and CP-C patients to attain the C-min target at 24 h. The maintenance dose regimens of 100 mg q12h or 200 mg q24h for CP-A/B patients and 50 mg q12h or 100 mg q24h for CP-C patients could obtain the probability of effective target attainment of >90% at an MIC <= 0.5 mg/L and achieve the cumulative fraction of response of >90% against C. albicans, C. parapsilosis, C. glabrata, C. krusei, A. fumigatus, and A. flavus. Additionally, the daily VRC doses could be increased by 50 mg for CP-A/B and CP-C patients at an MIC of 1 mg/L, with plasma C-min monitored closely to avoid serious adverse events. It is recommended that an appropriate alternative antifungal agent or a combination therapy could be adopted when an MIC >= 2 mg/L is reported, or when the infection is caused by C. tropicalis but the MIC value is not available. Conclusions For critically ill patients with liver dysfunction, the loading dose of intravenous VRC should be reduced to 5 mg/kg q12h. Additionally, based on the types of fungal pathogens and their susceptibility to VRC, the adjusted maintenance dose regimens with lower doses or longer dosing intervals should be considered for CP-A/B and CP-C patients.