Hospital-acquired pneumonia in critically ill patients: factors associated with episodes due to imipenem-resistant organisms.

BACKGROUND: Inadequate initial antimicrobial therapy represents one of the factors associated with mortality of patients suffering from hospital-acquired pneumonia. According to its wide antimicrobial spectrum, imipenem belongs to the usual antibiotics proposed by current guidelines for such a therapy. However, major changes in the antibiotic susceptibility patterns of bacteria in the intensive care unit (ICU) have occurred. Our goal was to determine the incidence of hospital-acquired pneumonia (HAP) due to imipenem-resistant organism(s) in our ICU and to identify factors associated with such a resistance.

PATIENTS AND METHODS: From January 1994 to December 2001, all consecutive patients admitted to our ICU for HAP or exhibiting HAP during their ICU stay were included in an observational cohort. Patients with a bacteriologically documented HAP were studied. For each causative pathogen, imipenem susceptibility was routinely determined. Patients with an HAP episode due to at least one imipenem-resistant causative organism were compared with patients who developed HAP in which all incriminated pathogens were imipenem susceptible.

RESULTS: 235 patients were included in our observational cohort. Among them, 168 had an HAP episode with a bacteriologically proven infection. In 42 patients (25%), at least one causative organism was resistant to imipenem. The 44 imipenem-resistant organisms were Staphylococcus aureus (n = 15), Pseudomonas aeruginosa (n = 14), Stenotrophomonas maltophilia (n = 13), and Acinetobacter baumannii (n = 2). Multivariate analysis identified four significant independent factors associated with resistance of causative organism(s) to imipenem: prior use of a fluoroquinolone (AOR = 3.9; 95% CI: 1.8 to 8.8; p < 0.0001), prior use of an aminoglycoside (AOR = 2.6; CI: 1.2 to 5.9; p = 0.02), use of invasive blood pressure monitoring (AOR = 2.7; CI: 1.0 to 7.0; p = 0.04) and bilateral chest X-ray involvement (AOR = 2.6; CI: 1.1 to 5.8; p = 0.02).

CONCLUSION: Factors associated with potential inadequacy of imipenem used as the single antibiotic for initial empiric treatment for HAP were identified. When they are present, imipenem should be either combined with antibiotics such as vancomycin and ciprofloxacin or replaced with another broad-spectrum antimicrobial regimen.