Imipenem resistance of Pseudomonas in pneumonia: a systematic literature review

BACKGROUND: Pneumonia, and particularly nosocomial (NP) and ventilator-associated pneumonias (VAP), results in high morbidity and costs. NPs in particular are likely to be caused by Pseudomonas aeruginosa (PA), ~20% of which in observational studies are resistant to imipenem. We sought to identify the burden of PA imipenem resistance in pneumonia. METHODS: We conducted a systematic literature review of randomized controlled trials (RCT) of imipenem treatment for pneumonia published in English between 1993 and 2008. We extracted study, population and treatment characteristics, and proportions caused by PA. Endpoints of interest were: PA resistance to initial antimicrobial treatment, clinical success, microbiologic eradication and on-treatment emergence of resistance of PA. RESULTS: Of the 46 studies identified, 20 (N = 4,310) included patients with pneumonia (imipenem 1,667, PA 251; comparator 1,661, PA 270). Seven were double blind, and 7 included US data. Comparator arms included a β-lactam (17, [penicillin 6, carbapenem 4, cephalosporin 7, monobactam 1]), aminoglycoside 2, vancomycin 1, and a fluoroquinolone 5; 5 employed double coverage. Thirteen focused exclusively on pneumonia and 7 included pneumonia and other diagnoses. Initial resistance was present in 14.6% (range 4.2-24.0%) of PA isolates in imipenem and 2.5% (range 0.0-7.4%) in comparator groups. Pooled clinical success rates for PA were 45.2% (range 0.0-72.0%) for imipenem and 74.9% (range 0.0-100.0%) for comparator regimens. Microbiologic eradication was achieved in 47.6% (range 0.0%-100.0%) of isolates in the imipenem and 52.8% (range 0.0%-100.0%) in the comparator groups. Resistance emerged in 38.7% (range 5.6-77.8%) PA isolates in imipenem and 21.9% (range 4.8-56.5%) in comparator groups. CONCLUSIONS: In the 15 years of RCTs of imipenem for pneumonia, PA imipenem resistance rates are high, and PA clinical success and microbiologic eradication rates are directionally lower for imipenem than for comparators. Conversely, initial and treatment-emergent resistance is more likely with the imipenem than the comparator regimens.