Search

Objectives: Antibiotics are the cornerstone in the treatment of sepsis. Microdialysis (MD) data in adults suggest an impaired antibiotic tissue penetration in the case of sepsis. Tissue pharmacokinetics (PK) remain largely understudied in children. Juvenile pig models have proven to provide an accurate prediction of PK behavior in pediatric patients. This study aimed to investigate the influence of sepsis on the tissue penetration of piperacillin (PIP) - tazobactam (TAZ) in a piglet model. Methods: In 16 piglets, PIP-TAZ was administered over 4 days (75 mg/kg IV over 30 minutes, 6h dosing interval). Blood and MD samples (muscle) were collected in first-dose and steady-state (>24h) conditions. On day 3 and 4, in 10 piglets a continuous lipopolysaccharide (LPS) infusion was administered to induce a septic state. In the 6 control animals (no LPS) time effects during the study period were evaluated. Non-compartmental PK analysis was used to quantify the tissue penetration (Area Under the concentration-time Curve (AUC) ratio tissue/plasma). The AUC ratios were pairwise compared between the healthy and septic states in each piglet, data are reported as mean ± SD. Results: For PIP, the AUC ratio in first-dose conditions was significantly lower in the septic state (0.84 ± 0.22) compared to the healthy baseline measurement (1.06 ± 0.46) (P = 0.042). In steady-state conditions, comparable results were found with an AUC ratio of 0.80 ± 0.22 in the septic state and 1.09 ± 0.27 during baseline (p=0.009). Comparable results for TAZ were observed. There were no time effects observed in the control group. Conclusion: In this juvenile pig model, experimental endotoxemia impaired the PIP-TAZ tissue penetration. The results of this study warrant further research into the tissue PK of septic children to optimize antibiotic dosing in this population. Correlating the results of this study with future clinical data could endorse the applicability of the piglet model for drug research in septic children.

Objectives: Calibration is an essential aspect of microdialysis research. In drug trials, retrodialysis by drug (RD) and retrodialysis by internal calibrator (RC) are the most frequently used calibration techniques. In this study, we aimed (1) to evaluate these calibration techniques for piperacillin (PIP) - tazobactam (TAZ) by comparing the recovery rates (RRs) during RD and RC, both in vitro and in vivo, and (2) to assess the reliability of in vitro to in vivo extrapolation of microdialysis methodology and RR for PIP-TAZ. Methods: Microdialysis catheters with a membrane length of 10 mm were used. The flow rate was set to 2 µL/min. During RD both PIP and TAZ were added to the perfusion solution (NaCl 0,9%). Benzylpenicillin was chosen as internal calibrator for PIP and TAZ. This methodology was first tested in vitro and was subsequently applied in a preclinical study in sixteen piglets and a clinical study in five pediatric intensive care (PICU) patients. Both calibration techniques were performed in each animal and study patient. Microdialysis catheters were placed in the paraspinal and thigh muscle, respectively. Data are reported as mean ± standard deviation. Results: Following the in vitro experiments, both RD and RC were deemed successful for PIP and TAZ, with mean RRs of 47 ± 6% (PIP) and 59 ± 5% (TAZ) for RD and 52 ± 5% for RC. In the piglet study, in general, RC recovery (29 ± 10%) rendered higher RRs and resulted in more consistent and reliable tissue concentration-time curves than RD (22 ± 10% PIP; 27 ± 11% TAZ). In the first two PICU study patients both calibration techniques performed poorly with very low RRs: 0.2 ± 0.4% (PIP) and 1.5 ± 0.7% (TAZ) for RD and 7 ± 6% for RC. As these discrepancies between the preclinical and clinical in vivo RRs were unexpected, methodological changes were made in the subsequent study patients. Reducing the flow rate from 2 to 1 µL/min led to an important increase of the RR for RC (24 ± 9%), but satisfactory RR for RD could not be reached. Additionally omitting TAZ from the perfusion solution during RD significantly improved the performance of RD for PIP (23 ± 8 %). Conclusion: We documented that an in vitro optimization of experimental design does not always translate into satisfactory RRs in vivo and thus a successful measurement of tissue concentrations. Also, extrapolation of RRs between different in vivo settings should be done with caution, as was shown by the marked differences in RRs between the preclinical piglet study and initial PICU patients. The finding in the clinical study that simultaneous RD for PIP and TAZ seemed to reduce the loss rate of both compounds, requires further study.