Development and validation of large animal models of Pseudomonas aeruginosa ventilator-associated pneumonia are needed for testing new drug candidates in a manner that mimics how they will be used clinically. We developed a new model in which rabbits were ventilated with low tidal volume and challenged with P. aeruginosa to recapitulate hallmark clinical features of acute respiratory distress syndrome (ARDS): acute lung injury and inflammation, progressive decrease in arterial oxygen partial pressure to fractional inspired oxygen PaO2:FiO2, leukopenia, neutropenia, thrombocytopenia, hyperlactatemia, severe hypotension, bacterial dissemination from lung to other organs, multiorgan dysfunction, and ultimately death. We evaluated the predictive power of this rabbit model for antibiotic efficacy testing by determining whether a humanized dosing regimen of meropenem, a potent antipseudomonal β-lactam antibiotic, when administered with or without intensive care unit (ICU)-supportive care (fluid challenge and norepinephrine), could halt or reverse natural disease progression. Our humanized meropenem dosing regimen produced a plasma concentration-time profile in the rabbit model similar to those reported in patients with ventilator-associated bacterial pneumonia. In this rabbit model, treatment with humanized meropenem and ICU-supportive care achieved the highest level of survival, halted the worsening of ARDS biomarkers, and reversed lethal hypotension, although treatment with humanized meropenem alone also conferred some protection compared to treatment with placebo (saline) alone or placebo plus ICU-supportive care. In conclusion, this rabbit model could help predict whether an antibiotic will be efficacious for the treatment of human ventilator-associated pneumonia.
Keywords: Pseudomonas aeruginosa; animal efficacy model validation; rabbit model.