The objectives of this study were to describe meropenem pharmacokinetics (PK) in plasma and/or subcutaneous adipose tissue (SCT) in critically ill patients receiving extracorporeal membrane oxygenation (ECMO) treatment and to develop a population PK model to simulate alternative dosing regimens and modes of administration. We conducted a prospective observational study. Ten patients on ECMO treatment received meropenem (1 or 2 g) intravenously over 5 min every 8 h. Serial SCT concentrations were determined using microdialysis and compared with plasma concentrations. A population PK model of SCT and plasma data was developed using NONMEM. Time above clinical breakpoint MIC for (8 mg/liter) was predicted for each patient. The following targets were evaluated: time for which the free (unbound) concentration is maintained above the MIC of at least 40% (40%T>MIC), 100% T>MIC, and 100% T>4×MIC. For all dosing regimens simulated in both plasma and SCT, 40% T>MIC was attained. However, prolonged meropenem infusion would be needed for 100% T>MIC and 100% T>4×MIC to be obtained. Meropenem plasma and SCT concentrations were associated with estimated creatinine clearance (eCL). Simulations showed that in patients with increased eCL, dose increment or continuous infusion may be needed to obtain therapeutic meropenem concentrations. In conclusion, our results show that using traditional targets of 40% T>MIC for standard meropenem dosing of 1 g intravenously every 8 h is likely to provide sufficient meropenem concentration to treat the problematic pathogen for patients receiving ECMO treatment. However, for patients with an increased eCL, or if more aggressive targets, like 100% T>MIC or 100% T>4×MIC, are adopted, incremental dosing or continuous infusion may be needed.

Background and purposeThe incidence of orthopedic methicillin-resistant Staphylococcus aureus (MRSA) infections is increasing. Vancomycin may therefore play an increasingly important role in orthopedic perioperative antimicrobial prophylaxis. Studies investigating perioperative bone and soft tissue concentrations of vancomycin are sparse and challenged by a lack of appropriate methods. We assessed single-dose plasma, subcutaneous adipose tissue (SCT) and bone concentrations of vancomycin using microdialysis in male patients undergoing total knee replacement.Methods1,000 mg of vancomycin was administered postoperatively intravenously over 100 minutes to 10 male patients undergoing primary total knee replacement. Vancomycin concentrations in plasma, SCT, cancellous, and cortical bone were measured over the following 8 hours. Microdialysis was applied for sampling in solid tissues.ResultsFor all solid tissues, tissue penetration of vancomycin was significantly impaired. The time to a mean clinically relevant minimal inhibitory concentration (MIC) of 2 mg/L was 3, 36, 27, and 110 min for plasma, SCT, cancellous, and cortical bone, respectively. As opposed to the other compartments, a mean MIC of 4 mg/L could not be reached in cortical bone. The area under the concentration-time curve from 0 to the last measured value and peak drug concentrations (Cmax) for SCT, cancellous, and cortical bone was lower than that of free plasma. The time to Cmax was higher for all tissues compared with free plasma.InterpretationPostoperative penetration of vancomycin to bone and SCT was impaired and delayed in male patients undergoing total knee replacement surgery. Adequate perioperative vancomycin concentrations may not be reached using standard prophylactic dosage.