Patients with acute-on-chronic liver failure (ACLF) are highly vulnerable to invasive fungal infections often requiring renal replacement therapy (RRT). The pharmacokinetics (PK) of isavuconazole, an antifungal against Aspergillus spp. and Mucorales , may be altered in ACLF due to effects of altered distribution and clearance. We investigated the impact of ACLF on isavuconazole PK in critically ill patients undergoing continuous RRT due to a paucity of data in this patient population and aimed to inform decisions about dosing. This prospective observational cohort study was conducted at the University Medical Center Hamburg-Eppendorf across twelve intensive care units with 140 beds. We included critically ill adults receiving isavuconazole and continuous RRT. Patients <18 years or with extracorporeal circuits other than RRT were excluded. Plasma, ultrafiltrate, and pre-/postfilter blood samples were analyzed using high-performance liquid chromatography. Key areas of interest were the effect of ACLF and continuous RRT on isavuconazole pharmacokinetics, particularly on clearance. Population PK modeling was performed using, inter alia, the integrated dialysis pharmacometric model in NONMEM®. Probability of target attainment (PTA) for established PK/PD thresholds developed in mice (50% or 90% survival) was assessed using Monte Carlo simulations. Thirteen patients were enrolled, including 4 with ACLF. A two-compartment model best described the data. Median clearance was similar between ACLF (4.2 L/h) and non-liver failure (4.8 L/h) groups (p = 0.92). Dialysis clearance was negligible (<1% of dose). The 50% efficacy target (AUC 0−24h /MIC = 24.7) was achieved in 12/13 patients (92%), and the 90% target (AUC 0−24h /MIC = 33.4) in 8/13 patients (62%). Simulations indicated suboptimal PTA at the Aspergillus fumigatus breakpoint of 1 mg/L. Considerable interindividual variability on clearance, particularly in ACLF, led to both sub- and supratherapeutic exposures. ACLF did not significantly alter isavuconazole clearance during RRT. However, high PK variability and frequent subtherapeutic exposures support routine therapeutic drug monitoring in critically ill patients.
Behrens et al. (Fri,) studied this question.