Population pharmacokinetics and dosing optimisation of polymyxin B in patients with severe burns

Data on polymyxin B (PMB) pharmacokinetics (PK) in patients with severe burns are scarce. We aimed to develop a population PK (PopPK) model for PMB in this population, identify factors influencing PK parameters, and optimise dosing regimens.

Plasma concentrations were measured using ultra-performance liquid chromatography-tandem mass spectrometry. PopPK modelling was performed with nonlinear mixed-effects modelling (NONMEM). Monte Carlo simulations were used to design PMB regimens targeting an optimal therapeutic window (AUC_ss,24 h, 50-100 mg·h·L⁻¹) and a PK/pharmacodynamic threshold of AUC_ss,24 h/MIC > 50, with a ≥ 90 % probability of target attainment (PTA).

A total of 118 PMB concentrations from 53 patients were analysed. A one-compartment model with first-order elimination was used to describe the data. Blood urea nitrogen (BUN) was a significant covariate, showing a negative correlation with clearance. Loading doses had a minimal effect on PTA or the proportion of patients achieving an AUC_ss,24 h within the therapeutic window. At minimum inhibitory concentration (MIC) ≤ 0.5 mg·L⁻¹ , a 75 mg q12h regimen achieved > 90 % PTA with low risk of supratherapeutic exposure (AUC_ss,24 h > 100 mg·h·L⁻¹). At MIC 1 mg·L⁻¹ , only the 100 mg q12h regimen in patients with moderate-to-high BUN achieved ≥ 90 % PTA.

Overall, this study provides PK insights into PMB dosing in patients with severe burns, supporting 75 mg q12h (for MIC ≤ 0.5 mg·L⁻¹) as an effective regimen. Regular BUN monitoring and therapeutic drug monitoring are recommended for personalised dosing.