Evaluating the optimal time for amikacin administration with respect to haemodialysis using an in vitro pharmacodynamic simulation against epidemic nosocomial OXA-48 producing Klebsiella pneumoniae ST405 strains

Abstract

Objectives: Bacterial viability and enrichment of resistance resulting from three different amikacin administration schedules with respect to haemodialysis (HD) were assessed against three OXA-48-producing Klebsiella pneumoniae isolated during an outbreak in a Spanish hospital. Methods: A previously described two-compartment dynamic system was used. Three possible amikacin administration schedules were simulated: (i) haemodialysis immediately after amikacin infusion (pre-HD); (ii) infusion immediately after haemodialysis (post-HD); and (iii) infusion at 50% interdialytic period. Amikacin standard dose (SD) and double dose (DD) were simulated for each schedule. Values of Cmax/MIC, Cmax/MPC (mutant prevention concentration), AUC0-48h/MIC, AUC0-48h/MPC and %TMSW (percentage of time that the concentration was inside the mutant selection window) were determined with experimental data and were correlated with the area under the bacterial killing curve of the total population and the resistant subpopulation. Results: Both with SD and DD, the pre-HD schedule resulted in increases at 48h in bacterial counts of the total population at the expense of enrichment of pre-existing resistant subpopulations from 12h onwards for all strains. The estimated %TMSW that prevented enrichment of resistant mutants was <61.5%. The AUC0-48h/MPC (with values of ≈40 being associated with countering of increases in resistant subpopulations) was better than the %TMSW as a predictive parameter. Conclusion: This study showed that the longest times concentrations were above the MPC (i.e. highest AUC0-48h/MPC, lowest %TMSW), the lowest enrichment of resistant subpopulations. This implies use of the highest possible amikacin dose (limited by toxicity) and post-HD as the best administration schedule.