(172b) Deciphering Longitudinal Optical-Density Measurements to Guide Antibiotic Use: A Model Based Approach | AIChE

(172b) Deciphering Longitudinal Optical-Density Measurements to Guide Antibiotic Use: A Model Based Approach


Nikolaou, M., University of Houston
Tam, V. H., University of Houston
Background: individualized design of drug administration for treatment of bacterial infections can benefit from experimental data on the effect of candidate drugs on infecting bacteria. In previous work, our group developed a model-based technique that renders usable optical density measurements of bacteria in suspension exposed to antibiotics over time, as these measurements count both dead and live bacterial cells and thus are practically useless. The fitted model can be used to design therapeutic treatment under clinically relevant pharmacokinetic conditions.

Objectives: Here we use in vitro simulation on a hollow-fiber infection model to validate predictions by the fitted model on the bactericidal efficacy of drugs that follow clinically relevant pharmacokinetics.

Methods: Optical-density measurements were taken over 20 hours in an automated instrument, based on which bacterial (Acinetobacter baumannii) cell counts over time were assessed under exposure to various time-invariant antibiotic concentrations. The model separated live from dead cell, thus estimating antibiotic kill rates at various concentrations. Subsequently, predictions were made on the bactericidal efficacy of the same antibiotic following a periodic profile of injection and exponential decline. These predictions were tested in an in vitro hollow-fiber infection model.

Results: Analysis suggested that suppression of the bacterial population could not be expected with clinically relevant ceftazidime concentrations alone. A combination of ceftazidime and amikacin (at a 2:1 ratio) would be necessary. These predictions were validated in the hollow-fiber infection model.

Conclusions: In vitro simulations substantiated the feasibility for clinicians to combine easily automated optical-density measurements with the proposed mathematical framework for the design of effective therapeutic treatments of challenging bacterial infections in a clinical setting. Additional validation experiments will be undertaken to consolidate the utility of the proposed method.