Objectives To describe polymyxin B pharmacokinetics in patients receiving veno-venous extracorporeal membrane oxygenation (ECMO) in comparison with critically ill patients without ECMO support and to explore potential covariates that could affect the pharmacokinetics in this group of patients. Patients and methods In 13 critically ill patients on ECMO and in 21 critically ill patients without ECMO support, 6–8 blood samples were collected during 12 h intervals after reaching steady state. Polymyxin B concentration in serum was determined using a previously developed ELISA. Protein binding was assessed by rapid equilibrium dialysis. Results In 13 critically ill patients on ECMO who received polymyxin B, the median area under the concentration–time curve over 12 h (AUC0–12h) was 48.38 mg/h/L for the total drug and 14.08 mg/h/L for the free drug. The unbound fraction was 0.35. Total body clearance was 1.16 L/h. In non-ECMO patients, the median AUC0–12h was 34.7 mg/h/L and the median CL was 1.76 L/h. The volume of distribution was significantly lower in ECMO patients (19.7 versus 30.4 L, respectively). We found a moderate negative correlation between the ECMO blood flow rate and AUC0–12h, a strong negative correlation between SOFA score and polymyxin B clearance and a moderate correlation between polymyxin B clearance and renal function in ECMO patients. Conclusions Currently recommended polymyxin B dosage regimens are sufficient for patients receiving ECMO and no dosage increase is required. In our study, polymyxin B exposure was higher in ECMO patients compared with the control group.

No abstract