infection control and hospital epidemiology july 2007, vol. 28, no. 7 original article Microbiology of Ventilator-Associated Pneumonia Compared With That of Hospital-Acquired Pneumonia David J. Weber, MD, MPH; William A. Rutala, PhD, MPH; Emily E. Sickbert-Bennett, MS; Gregory P. Samsa, PhD; Vickie Brown, RN, MPH; Michael S. Niederman, MD objective. Nosocomial pneumonia is the leading cause of mortality attributed to nosocomial infection. Appropriate empirical therapy has been associated with improved survival, but data are limited regarding the etiologic agents of hospital-acquired pneumonia in non- ventilated patients (HAP). This evaluation assessed whether the currently recommended empirical therapy is appropriate for both ventilator- associated pneumonia (VAP) and HAP by evaluating the infecting flora. design. Prospectively collected hospitalwide surveillance data was obtained by infection control professionals using standard Centers for Disease Control and Prevention definitions. setting. A tertiary care academic hospital. patients. All patients admitted from 2000 through 2003. results. A total of 588 episodes of pneumonia were reported in 556 patients: 327 episodes of VAP in 309 patients, and 261 episodes of HAP in 247 patients. The infecting flora in ventilated patients included gram-positive cocci (32.0% [oxacillin-susceptible Staphylococcus aureus {OSSA}, 9.25%; oxacillin-resistant Staphylococcus aureus {ORSA}, 17.75%]), gram-negative bacilli (59.0% (Pseudomonas aeruginosa, 17.50%; Stenotrophomonas maltophilia, 6.75%; Acinetobacter species, 7.75%), and miscellaneous pathogens (9.0%). The infecting flora in nonventilated patients included gram-positive cocci (42.59% [OSSA, 13.33%; ORSA, 20.37%]), gram-negative bacilli (39.63% [P. aeruginosa, 9.26%; S. maltophilia, 1.11%; Acinetobacter species, 3.33%), and miscellaneous pathogens (17.78%). conclusions. Our data demonstrated that patients with HAP, compared with those with VAP, had a similar frequency of infection with ORSA but less commonly had infections due to P. aeruginosa, Acinetobacter species, and S. maltophilia. However, the overall frequency of infection with these pathogens was sufficiently high to warrant the use of empirical therapy likely to be active against them. Our data supports using the currently recommended empirical therapy for both HAP and VAP. Infect Control Hosp Epidemiol 2007; 28:825-831 From the Department of Hospital Epidemiology, University of North Carolina Health Care System (D.J.W., W.A.R., E.E.S.-B., V.B.), the Division of Infectious Diseases, University of North Carolina School of Medicine (D.J.W., W.A.R.), Chapel Hill, and the Department of Biostatistics and Bioinformatics, Duke University Medical School, Durham (G.P.S.), North Carolina; and Winthrop University Hospital and State University of New York at Stony Brook, New York, New York (M.S.N.). Received September 18, 2006; accepted November 16, 2006; electronically published May 17, 2007.  2007 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2007/2807-0009$15.00. DOI: 10.1086/518460 It has been estimated that each year nearly 2 million patients in the United States acquire infections in the hospital, re- sulting in about 90,000 deaths. 1 Nosocomial pneumonia has been reported to be the second most common healthcare- associated infection in US intensive care units (ICUs), 2 but in a prevalence study of European ICUs pneumonia was the most prevalent nosocomial infection. 3 Early administration of appropriate empirical therapy for nosocomial pneumonia has been demonstrated to significantly improve survival. 4-7 For this reason, the most recent guideline by the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA) on the treatment of nosocomial pneu- monia recommends empirical therapy. 8 The choice of anti- microbial agents is made on the basis of the most likely infecting flora and is modified according to time since ad- mission (ie, 0-4 days or 5 days or more), prior receipt of antibiotics, and the presence of certain risk factors (eg, res- idence in an extended care facility or receipt of dialysis within the past 90 days for a patient receiving long-term dialysis). The current guideline distinguishes ventilator-associated pneu- monia (VAP) from hospital-acquired pneumonia in nonven- tilated patients (HAP) and recommends similar therapy for both diagnoses. While the infecting flora associated with VAP has been well described, 9 only limited data are available re- garding the infecting flora associated with HAP. 10,11 To validate the use of the same empirical therapy for VAP and HAP, as modified by the presence of specified risk factors, we analyzed our prospectively obtained data regarding the infecting flora