RESUMEN
Colistin remains one of the few available options for the treatment of infections caused by resistant bacteria. Pharmacokinetic (PK) studies have been successful in estimating the appropriate colistin methanesulfonate (CMS) dose to achieve a target colistin concentration. Currently, there is a consensus that the dose of CMS should vary according to the patient renal function since CMS is mainly eliminated by renal route. For this same reason, the loading dose should vary according to the patient's renal capacity; however, this is not the current clinical practice. In this study we develop a framework to determine two key parameters for the loading dose regimen: (1) the optimal dose according to the characteristics (renal function and weight) of the patient; (2) the waiting time before the maintenance dose. Based on a previous PK model, our framework allows a fast parameter sweep so as to select optimal loading dose and waiting time minimizing the deviation between the plasma concentration and a target value. The results showed that patients presenting low creatinine clearance (CrCL) should receive a lower CMS loading dose with longer interval to start maintenance treatment to avoid nephrotoxic colistin concentrations. In cases of high CrCL, the dose should be higher and the interval to the next dose shorter to avoid subtherapeutic concentrations. Optimization of the loading dose should considerably improve colistin therapy, as the target concentration is reached more quickly, without reaching toxic values.
Asunto(s)
Antibacterianos , Colistina , Humanos , Colistina/farmacocinética , Colistina/uso terapéutico , Antibacterianos/farmacocinética , Enfermedad CríticaRESUMEN
OBJECTIVES: An optimized dosing regimen of the prodrug of colistin, colistin methanesulphonate (CMS), against resistant Pseudomonas aeruginosa is needed to ensure effective bacterial killing. The objectives of this study were to develop a pharmacokinetic (PK)/pharmacodynamic (PD) model that characterizes the time course of the antibacterial activity of colistin against P. aeruginosa in a static in vitro system and to perform simulations of different dosing regimens and dosing algorithms to evaluate the effect of interindividual variability and interoccasion variability in PK on bacterial killing. METHODS: Static in vitro time-kill curve experiments were conducted on two different strains of P. aeruginosa (MIC 1 and 1.5 mg/L). Mechanism-based PK/PD models were fitted in NONMEM7 and the final model was combined with a previously developed population PK model of CMS and colistin to perform simulations of variability based on different dosing algorithms. RESULTS: A model with compartments for growing and resting bacteria, with a function allowing the maximal bacterial killing of colistin to reduce upon increasing colistin exposure, characterized both the fast bactericidal effect and the adaptive resistance. The variability in PK was shown to translate into pronounced interoccasion variability in bacterial killing. A flat fixed loading dose was demonstrated to result in less variability than an algorithm based on weight. CONCLUSIONS: The developed PK/PD model described the growth, death and resistance development of P. aeruginosa in response to colistin for two different strains. Based on simulations, a flat fixed loading dose followed by an 8 or 12 hourly maintenance dose with an infusion duration of up to 2 h appeared adequate.