Empiric therapy for pneumonia in the surgical intensive care unit.

Empiri c therapy of ventilator-associated pneumonia (VAP) in surgical patients should be based on intensive care unit (ICU)-specific surveillance data, because microbial flora patterns vary widely between geographic regions as well as within hospitals. Surgical ICUs have higher VAP rates than other units. Data from the National Nosocomial Infection Surveillance (NNIS) System report Pseudomonas aeruginosa and Staphylococcus aureus to be the most frequent isolates (each 17.4%). Data from the NNIS documents high resistance patterns in ICUs compared with hospitals at large, as well as unit-specific patterns. VAP risk factors for surgical patients include thoracoabdominal surgery, altered level of consciousness, advanced age, diabetes mellitus, malnutrition, chronic obstructive pulmonary disease, and prior antibiotic administration. Promising prevention strategies include restricting ventilator circuit changes, in-line heat moisture exchange filters, semi-recumbant positioning, and continuous subglottic aspiration. Pharmacodynamics should be considered when choosing antibiotic regimens. Postantibiotic effect and time-dependent versus concentration-dependent killing should be studied in clinical trials. Current guidelines for choosing regimens have been well developed by the American Thoracic Society.