Electronic-nomogram-guided micafungin dosing regimen in critically ill patients: A population pharmacokinetic study.

OBJECTIVES

To develop a population pharmacokinetic model for micafungin and an electronic nomogram to support the selection of the optimal initial dosing regimen.

METHODS

A retrospective, monocentric pharmacokinetic study was conducted in a digestive intensive care unit. Population pharmacokinetic modelling and covariate analysis were performed using Monolix.2024R1. Monte Carlo simulations were subsequently used to build the e-nomogram.

RESULTS

Overall, 404 micafungin plasma concentrations obtained from 100 patients were analysed, including 40% with cirrhosis and/or a history of liver transplantation. The concentration-time data were adequately described by a linear one-compartment model. The estimated PK parameters (between-subject variability) were 17.3 L (15%) for volume of distribution and 0.85 L/h (25%) for clearance (Cl). Total body weight and total bilirubin were identified to influence volume of distribution and/or Cl. External model evaluation showed acceptable predictive performance for both population and individual predictions. Monte Carlo simulations suggested that the currently recommended regimen of 100 mg/d may be insufficient to achieve the pharmacokinetic/pharmacodynamic target for all Candida species when clinical breakpoint MIC values are considered.

CONCLUSION

The resulting RShiny application provides clinicians with a practical decision support tool for selecting the best loading and maintenance doses according to total body weight, BILI and the target pathogen, thereby maximising the probability of pharmacokinetic/pharmacodynamic target attainment.