Objectives: Knowledge regarding antimicrobial therapy strategies in deep sternal wound infections (DSWI) following cardiac surgery is limited. Therefore, we aim to determine the steady-state plasma and mediastinal concentrations of oxacillin administered by continuous infusion in critically ill patients with DSWI and to compare these concentrations with the susceptibility of Staphylococci recovered.
Methods: A continuous infusion of oxacillin (150-200 mg/kg/24 h) was administered after a loading dose (50 mg/kg). Plasma and mediastinal concentrations of total and unbound oxacillin were determined 4 h after the loading dose (H4) and then at day 1 (H24) and day 2 (H48).
Results: Twelve patients were included. Nine patients exhibited bacteraemia, 5 were in septic shock, 8 were positive for Staphylococcus aureus and 4 for coagulase-negative staphylococci. The median MIC was 0.25 (0.24-0.41) mg/L. Median plasma concentrations of total and unbound oxacillin at H4, H24 and H48 were, respectively, 64.4 (41.4-78.5) and 20.4 (12.4-30.4), 56.9 (31.4-80.6) and 21.7 (6.5-27.3), and 57.5 (14.3-35.7) and 20 (14.3-35.7) mg/L. The median mediastinal concentrations of total and unbound oxacillin at H4, H24 and H48 were, respectively, 2.3 (0.7-25.9) and 0.9 (<0.5-15), 29.1 (19.7-38.2) and 12.6 (5.9-19.8), and 31.6 (14.9-42.9) and 17.1 (6.7-26.7) mg/L.
Conclusions: High-dose oxacillin delivered by continuous infusion is a valuable strategy to achieve our pharmacokinetic target (4×MIC) at the site of action at H24. But concerns remain in cases of higher MIC emphasising the need of clinicians to obtain the MIC of the bacteria and to monitor oxacillin concentrations, especially the unbound forms, at the target site.