International Journal of Infectious Diseases (Mar 2025)
Pharmacokinetics of colistin in critically-ill patients on sustained low-efficiency dialysis: a population pharmacokinetic study
Abstract
Introduction: Colistin, administered intravenously as its predominantly renally eliminated inactive prodrug colistin methanesulphonate (CMS), is used to treat infections caused by multidrug-resistant Gram-negative bacteria. Sustained low-efficiency dialysis (SLED) is a popular form of renal replacement therapy in patients requiring renal support. However, there is limited knowledge on the impact of SLED on the pharmacokinetics (PK) of colistin formed in vivo from CMS and on dosing requirements for this last-line antibiotic. We hypothesised that the PK of formed colistin is substantially impacted by SLED. Methods: A prospective population PK study included 13 critically-ill patients (six females) treated with CMS and receiving SLED (duration 6-8h) for renal support. Median (range) age was 68yrs (33-76), body weight (BW) 75.6kg (55.6-109) and eGFR 18.8mL/min/1.73m² (9.22-37.3). For each subject, the PK of formed colistin was studied on both a non-SLED day and a SLED day (n=8 studied during SLED day first). A single intravenous daily dose of CMS [150mg colistin base activity (CBA)] was administered on a non-SLED day. On a SLED day, subjects received 150mg CBA 12-hourly based on the paucity of prior data. Serial blood samples were collected over 24h on both days. Colistin plasma concentrations were measured by high-performance liquid chromatography and analysed using non-compartmental and population PK approaches. Results: The average plasma colistin concentration across 0-24h was >2mg/L (recommended target concentration) in all subjects on both the non-SLED and SLED days, and was not different (P>0.05) between the two days. A linear one-compartment PK disposition model best described the data. The population mean apparent body clearance (CLbody) of colistin, excluding SLED clearance, was 1.62L/h (27% interindividual variability). The colistin clearance by SLED was 3.66L/h (33%), equivalent to 69% of total aggregated colistin clearance on a SLED day. The apparent volume of distribution was estimated to be 52.9L (24%). No covariate relationships were found between the evaluated patient characteristics [BW and creatinine clearance (CrCL)] and the PK parameter estimates for colistin. Discussion: To our knowledge, this study is the first to define the population PK of formed colistin exclusively in patients undergoing SLED, and can assist in optimisation of dosing regimens in this patient group. We found that the clearance of colistin was substantially higher during SLED, as has been reported for other types of renal replacement therapies. Although CrCL was identified as a covariate for the apparent clearance of colistin in a previous report, that study included a considerably larger number of patients, the majority of whom were not receiving renal replacement therapy, and they had a much wider range of CrCL. Conclusion: Our findings suggest that the CMS dose should be higher on days when SLED is undertaken; and, they enable ongoing analyses to optimise dosing regimens.