Abstract
OBJECTIVE: Carbapenem-resistant Acinetobacter baumannii (CRAB) is a global concern as effective treatments are very limited. We previously used a modified susceptibility testing approach to predict growth suppression in carbapenem-resistant Enterobacterales, but there are uncertainties about the generalizability of the model. The objective of this study is to verify if a similar approach can be extended to CRAB. METHOD: A clinical isolate of CRAB resistant to ceftazidime/avibactam (CAZ/AVI, MIC = 32/4 mg/L) was examined. CAZ susceptibility was determined using increasing concentrations of AVI (0-64 mg/L), and MIC reduction was characterized with a sigmoid inhibitory maximum effect (Emax) model. The effectiveness of CAZ/AVI was validated in a hollow fibre infection model (HFIM) over 72 hours, using simulated unbound serum / epithelial lining fluid (ELF) exposures of 2.5 g over 2 hours every 8 hours. Baseline inocula of approximately 5.5 log CFU/mL were examined. RESULTS: An AVI concentration-dependent reduction in CAZ MIC was observed (r(2) = 0.99). CAZ MIC was dramatically reduced from 512 mg/L (no AVI) to 32 mg/L (AVI = 4 mg/L), and further to 8 mg/L (AVI = 16 mg/L). Pharmacokinetic simulations were satisfactory in the HFIM (r(2) > 0.96). Bacterial suppression was observed >24 hours with the serum exposure, but not that from the ELF. CONCLUSION: Using multiple AVI concentrations within the clinically relevant range, our susceptibility testing approach could have better insights of treatment outcome for infections caused by CRAB. This could potentially lead to effective intervention(s) overlooked by conventional susceptibility testing method. This case highlights the importance of site-specific drug exposures on determining treatment outcome.