Optimizing cloxacillin prophylaxis in hip and knee arthroplasty based on population pharmacokinetics of unbound plasma concentrations

Objectives To characterize the population pharmacokinetics of unbound cloxacillin in patients undergoing total arthroplasty of the hip (THA) or knee (TKA), and to explore alternative dosing regimens for cloxacillin prophylaxis. Methods Plasma concentrations of total and unbound cloxacillin from 200 patients undergoing primary elective THA (n & 95) or TKA (n & 105) were analysed. Intravenous cloxacillin doses of 2 g were administered pre-surgery, and repeated after 2 and 6 hours. Samples (n & 496) were analysed using HPLC-MS/MS. Non-linear mixed-effects modelling was performed to develop a population pharmacokinetic model describing unbound cloxacillin exposure. Using this model, alternative prophylaxis regimens were explored with Monte Carlo simulations. Results A two-compartment model with non-linear protein binding adequately described the data. Estimated glomerular filtration rate (eGFR) and body weight (kg) were significant covariates on unbound cloxacillin clearance. In 13% of patients sampled at the end of surgery (n & 25/187), unbound cloxacillin <2 mg/L was observed. A model-predicted 18–22% (n & 36–43/200) of patients failed to sustain plasma levels ≥2 mg/L throughout the two-hour dosing interval with the current regimen. In contrast, a continuous 1 g/h infusion after a 1 g loading dose would ensure target attainment in >99% of patients, according to the model predictions. Conclusions For many THA and TKA patients, the current cloxacillin prophylaxis regimen may fail to provide adequate target site concentrations during the entire surgical procedure. Transitioning to a prolonged infusion protocol could increase attainment of PK/PD targets without exceeding the currently recommended total perioperative dose amounts.