Résumé
Background Sustained low-efficiency dialysis (SLED) is a hybrid form of dialysis that is increasingly used in critically ill patients with kidney injury and hemodynamic instability. Antimicrobial dosing for patients receiving SLED is informed by pharmacokinetic studies that describe the drug clearance. Studies available to assist in the dosing of vancomycin in the context of SLED are lacking. Objective The objective of this prospective observational study was to describe the population pharmacokinetics of vancomycin in critically ill patients receiving SLED, and use simulation studies to propose dosing strategies. Methods Serial serum samples were obtained from 31 critically ill patients prescribed vancomycin while receiving SLED. Vancomycin concentrations were quantified in plasma using a validated liquid chromatography mass spectrometry/mass spectrometry method. A population pharmacokinetic model was developed, and Monte Carlo simulation was used to determine the probability of target attainment at different doses. Results From a total of 335 serum samples from 31 patients receiving 52 sessions of SLED therapy, a two-compartment linear model with zero-order input was developed. The mean (standard deviation) clearance of vancomycin on and off SLED was 5.97 (4.04) and 2.40 (1.46) L/h, respectively. Using pharmacodynamic targets for efficacy (area under the concentration-time curve from time zero to 24 h [AUC(24)]/minimum inhibitory concentration [MIC] >= 400) and safety (AUC(24) >= 700), a loading dose of 2400 mg followed by daily doses of 1600 mg is recommended. Subsequent dosing should be informed by therapeutic drug monitoring of vancomycin levels. Conclusions In critically ill patients receiving SLED, vancomycin clearance is highly variable with a narrow therapeutic window. Empiric dosing is proposed but subsequent dosing should be guided by drug levels.