Aminoglycosides in Critically Ill Septic Patients With Acute Kidney Injury Receiving Continuous Renal Replacement Therapy: A Multicenter, Observational Study

Abstract

Purpose: Data on aminoglycoside stewardship in critically ill septic patients with acute kidney injury (AKI) needing continuous renal replacement therapy (CRRT) are scarce. The objectives of the study were to determine, during CRRT, the time window with low likelihood for safe reinjection and the proportion of inappropriate reinjection.

Methods: A post hoc observational analysis of a multicenter randomized trial comparing the risk of hemodialysis catheter infection with ethanol lock vs placebo in critically ill patients with AKI was conducted. Eligible patients were adults in intensive care units from 6 French hospitals. Any patient with AKI needing CRRT and receiving an antimicrobial therapy for a septic episode occurring before (≤24 hours) or during CRRT was included. The aminoglycoside orders were left to the physicians' discretion, but high dose once daily was the schedule of aminoglycoside administration.

Findings: A total of 145 septic episodes treated by aminoglycosides were analyzed in patients receiving CRRT. A mean (SD) of 1.6 (0.8) amikacin and 1.8 (1.2) gentamicin administrations per patient were observed. During CRRT, C_max was 17.3 mg/L (interquartile range, 13.2-22.5 mg/L) for gentamicin and 50 mg/L (interquartile range, 43.7-76.6 mg/L) for amikacin. The plasma drug concentration at 24 hours (C_H24) was 2.3 mg/L (interquartile range, 1.6-3.2 mg/L) for gentamicin and 9.3 (interquartile range, 6.6-12.0 mg/L) for amikacin. Sixty-five C_min dosages remained above the reinjection threshold. Inappropriate reinjection was observed in 11 of 65 episodes (17%). Inappropriate reinjection (defined by, at the reinjection time, C_min dosages above the threshold; ie, C_min >2 mg/L for gentamicin and >5 mg/L for amikacin) was observed in 17% of analyzed episodes. Most patients did not need reinjection until approximately ≥30 hours after their initial administration.

Implications: During CRRT, as indicated by the C_H24 value, which was higher than the recommended threshold, the interval to obtain a C_min low enough to allow for redosing aminoglycosides is significantly longer than 24 hours. This interval is not always respected and leads to an of inappropriate reinjection rate of 17%. ClinicalTrials.gov identifier: ISRCTNCT00875069.

Keywords: Acute kidney injury; aminoglycosides; continuous renal replacement therapy; critical care; sepsis.