Treatment of experimental pneumonia in rats caused by a PER-1 extended-spectrum beta-lactamase-producing strain of Pseudomonas aeruginosa

The antibacterial activity of imipenem, cefepime and piperacillin-tazobacta m alone or in combination with amikacin against a Pseudomonas aeruginosa st rain producing an extended-spectrum beta-lactamase (PER-1) were compared us ing an experimental model of pneumonia in non-leucopenic rats. Animals were infected intratracheally with 8.0 +/- 0.4 log(10) cfu of P. aeruginosa, an d therapy was initiated 3 h later, by which time animal lungs showed bilate ral pneumonia containing >7 log(10) P. aeruginosa cfu/g of tissue. Since ra ts eliminate antibiotics much more rapidly than humans, renal impairment wa s induced in all animals to simulate the pharmacokinetic parameters of huma ns. MICs determined using an inoculum of 4 log(10) cfu/mL were as follows: imipenem, 1 mg/L; cefepime, 8 mg/L; piperacillin-tazobactam, 32 mg/L; and a mikacin, 16 mg/L. A noticeable inoculum effect was observed with the four a ntimicrobial agents tested, which was greatest for cefepime and piperacilli n-tazobactam. In-vitro studies indicated that imipenem was the beta-lactam with the greatest bactericidal effect and that amikacin was synergic only i n combination with cefepime and imipenem. Cefepime and piperacillin-tazobac tam alone failed to decrease bacterial counts in the rats' lungs 60 h after therapy onset, whereas imipenem and, to a lesser extent, amikacin signific antly reduced the number of viable microorganisms. Combination of amikacin with any of the three beta-lactams tested was synergic, despite a high amik acin MIC for the infecting strain. These results paralleled our in-vitro da ta showing a marked inoculum effect for cefepime and piperacillin-tazobacta m. Based on the results of this study, the best treatment for infections ca used by this type of extended-spectrum beta-lactamase-possessing strain wou ld be imipenem plus amikacin.