(534) Analysis of Vancomycin Dosage and Plasma Levels in Critically Ill Adult Patients Requiring ECMO

Introduction: Critically ill patients, especially those on ECMO, exhibit unique pharmacokinetic and pharmacodynamic conditions that affect the dosage adjustment and therapeutic objectives of drugs such as vancomycin. This study compared the achievement of therapeutic targets in these patients, considering the ECMO connection configuration, initial creatinine level and clearance, requirement for hemodialysis, and use of vancomycin loading dose.

Methods: This was a non-concurrent, unicentric cohort of critically ill patients requiring ECMO support between 2010 and 2018 in a private, high-complexity center. The therapeutic objective of vancomycin was assessed and analyzed according to the ECMO modality (arteriovenous, VA versus venovenous, VV), baseline creatinine levels, clearance, requirement for hemodialysis, and the application of a vancomycin loading dose.

Results: Demographics 69% male, median age 50 years. The median duration of vancomycin therapy was 5 days, and 48.5% of patients received a loading dose. Despite dosage adjustments, only 30% of vancomycin plasma level measurements were within the therapeutic range. No significant differences were observed between the ECMO modality, initial creatinine level, glomerular filtration rate, requirement for hemodialysis, or the use of a loading dose with respect to the achievement of therapeutic targets. A strong correlation (R2 0.98) was found between the baseline vancomycin levels and AUC/MIC, with most patients achieving an AUC/MIC of 400-600 with plasma levels of 10-20.

Conclusions: The data indicated substantial variability in achieving therapeutic targets with vancomycin in critically ill patients on ECMO. There were no significant differences in the achievement of therapeutic goals despite the analyses performed. Despite these patients' severity, plasma levels of 10-20 had a good correlation with the predicted AUC/MIC, suggesting a potentially wider therapeutic range in these patients.

The limitations of this study include its unicentric and retrospective design. Also changes in clinical practice during the study period from the baseline plasma level to the AUC/CIM target. Finally, although several variables were analyzed, there may be confounding factors that do not affect the achievement of the target.