Model-informed individualized administration of ceftazidime–avibactam in critically ill patients: population pharmacokinetics studies and a parametric time-to-event analysis

Objectives Ceftazidime–avibactam is effective against resistant Gram-negative bacteria but associated with a risk of acute kidney injury (AKI), with incidence rates ranging from 13.3% to 33.0%. This study developed a population pharmacokinetic (PopPK) model for ceftazidime–avibactam in critically ill Chinese adults, identified AKI risk factors using a time-to-event (TTE) model, and optimized dosing. Methods The PopPK model was developed using non-linear mixed-effects modelling (31 patients; 104 samples) and validated with 59 samples (32 patients). A TTE model identified AKI risk factors, and Monte Carlo simulations optimized dosing strategies based on these risks. Results Both ceftazidime and avibactam were best described by a one-compartment model. Creatinine clearance (CrCL) affected clearance of both, while avibactam clearance decreased with increasing C-reactive protein. AKI patients had higher exposure. The TTE analysis identified ceftazidime AUCss,24 h and mechanical ventilation (MV) as risk factors for AKI. Simulations supported a dual-objective optimization (efficacy, AUCss,24 h ≥ 784 μg·h/mL; safety, MV ≤ 1200 μg·h/mL, non-MV ≤ 2000 μg·h/mL) and renal function-stratified dosing recommendations [CrCL > 80 mL/min, 2.5 g q8h for MV/non-MV; CrCL 50–80, 1.25 g q8h for MV/non-MV; CrCL 30–50, 0.94 g q12h (MV) versus 1.25 g q8h (non-MV); CrCL 15–30, 0.94 g q24h (MV) versus 0.94 g q12h/q24h (non-MV); CrCL 6–15, 0.94 g q48h (MV) versus 0.94 g q24h/q48h (non-MV)]. Conclusions The PopPK model for ceftazidime–avibactam was developed for critically ill Chinese adults. TTE analysis identified ceftazidime AUCss,24 h and MV as AKI risk factors. Renal function-guided dosing optimizes efficacy and safety.