Optimization and Evaluation of Piperacillin-Tobramycin Combination Dosage Regimens against Pseudomonas aeruginosa for Patients with Altered Pharmacokinetics via the Hollow-Fiber Infection Model and Mechanism-Based Modeling

摘要

Augmented renal clearance (ARC) in critically ill patients can result in suboptimal drug exposures and treatment failure. Combination dosage regimens accounting for ARC have never been optimized and evaluated against Pseudomonas aeruginosa by use of the hollow-fiber infection model (HFIM). Using a P. aeruginosa isolate from a critically ill patient and static-concentration time-kill experiments (SCTKs), we studied clinically relevant piperacillin and tobramycin concentrations, alone and in combinations, against two inocula (10(5.8) and 10(7.6) CFU/ml) over 72 h. We subsequently evaluated the effects of optimized piperacillin (4 g every 4 h [q4h], given as 0.5-h infusions) plus tobramycin (5 mg/kg of body weight q24h, 7 mg/kg q24h, or 10 mg/kg q48h, given as 0.5-h infusions) regimens on killing and regrowth in the HFIM, simulating a creatinine clearance of 250 ml/min. Mechanism-based modeling was performed in S-ADAPT. In SCTKs, piperacillin plus tobramycin (except combinations with 8 mg/liter tobramycin and against the low inoculum) achieved synergistic killing (= 2 log(10) versus the most active monotherapy at 48 h and 72 h) and prevented regrowth. Piperacillin monotherapy (4 g q4h) in the HFIM provided 2.4-log(10) initial killing followed by regrowth at 24 h and resistance emergence. Tobramycin monotherapies displayed rapid initial killing (= 5-log(10) at 13 h) followed by extensive regrowth. As predicted by mechanism-based modeling, the piperacillin plus tobramycin dosage regimens were synergistic and provided = 5-log(10) killing with resistance suppression over 8 days in the HFIM. Optimized piperacillin-tobramycin regimens provided significant bacterial killing and suppressed resistance emergence. These regimens appear to be highly promising for effective and early treatment, even in the near-worst-case scenario of ARC.